Modernize IO using only the API

pandas-stubs/_libs/lib.pyi

@@ -1,5 +1,14 @@
 from __future__ import annotations
 
+from enum import Enum
+
 no_default = None
 
+from typing import Literal
+
+class _NoDefault(Enum):
+    no_default = ...
+
+NoDefault = Literal[_NoDefault.no_default]
+
 def infer_dtype(value: object, skipna: bool = ...) -> str: ...

pandas-stubs/_typing.pyi

@@ -16,6 +16,7 @@ from typing import (
     AnyStr,
     Callable,
     Hashable,
+    Iterator,
     Literal,
     Mapping,
     Optional,
@@ -67,6 +68,18 @@ class BaseBuffer(Protocol): ...
 class ReadBuffer(BaseBuffer, Protocol[AnyStr_cov]): ...
 class WriteBuffer(BaseBuffer, Protocol[AnyStr_cov]): ...
 
+class ReadPickleBuffer(ReadBuffer[bytes], Protocol):
+    def readline(self) -> AnyStr_cov: ...
+
+class WriteExcelBuffer(WriteBuffer[bytes], Protocol):
+    def truncate(self, size: int | None = ...) -> int: ...
+
+class ReadCsvBuffer(ReadBuffer[AnyStr_cov], Protocol[AnyStr_cov]):
+    def __iter__(self) -> Iterator[AnyStr_cov]: ...
+    def readline(self) -> AnyStr_cov: ...
+    @property
+    def closed(self) -> bool: ...
+
 FilePath = Union[str, PathLike[str]]
 
 Buffer = Union[IO[AnyStr], RawIOBase, BufferedIOBase, TextIOBase, TextIOWrapper, mmap]
@@ -208,4 +221,6 @@ GroupByObjectNonScalar = Union[
 ]
 GroupByObject = Union[Scalar, GroupByObjectNonScalar]
 
+CSVEngine = Literal["c", "python", "pyarrow", "python-fwf"]
+
 __all__ = ["npt", "type_t"]

pandas-stubs/core/frame.pyi

@@ -48,8 +48,10 @@ from pandas._typing import (
     Axes,
     Axis,
     AxisType,
+    CompressionOptions,
     Dtype,
     DtypeNp,
+    FilePath,
     FilePathOrBuffer,
     FilePathOrBytesBuffer,
     GroupByObjectNonScalar,
@@ -61,12 +63,15 @@ from pandas._typing import (
     Label,
     Level,
     MaskType,
+    ReadBuffer,
     Renamer,
     Scalar,
     ScalarT,
     SeriesAxisType,
+    StorageOptions,
     StrLike,
     T as TType,
+    WriteBuffer,
     np_ndarray_bool,
     np_ndarray_str,
     num,
@@ -240,15 +245,19 @@ class DataFrame(NDFrame, OpsMixin):
     ) -> np.recarray: ...
     def to_stata(
         self,
-        path: FilePathOrBuffer,
-        convert_dates: dict | None = ...,
+        path: FilePath | WriteBuffer[bytes],
+        convert_dates: dict[Hashable, str] | None = ...,
         write_index: _bool = ...,
-        byteorder: _str | Literal["<", ">", "little", "big"] | None = ...,
-        time_stamp=...,
+        byteorder: Literal["<", ">", "little", "big"] | None = ...,
+        time_stamp: _dt.datetime | None = ...,
         data_label: _str | None = ...,
-        variable_labels: dict | None = ...,
-        version: int = ...,
-        convert_strl: list[_str] | None = ...,
+        variable_labels: dict[Hashable, str] | None = ...,
+        version: int | None = ...,
+        convert_strl: list[HashableT] | None = ...,
+        compression: CompressionOptions = ...,
+        storage_options: StorageOptions = ...,
+        *,
+        value_labels: dict[Hashable, dict[float | int, str]] | None = ...,
     ) -> None: ...
     def to_feather(self, path: FilePathOrBuffer, **kwargs) -> None: ...
     @overload
@@ -332,6 +341,25 @@ class DataFrame(NDFrame, OpsMixin):
         render_links: _bool = ...,
         encoding: _str | None = ...,
     ) -> _str: ...
+    def to_xml(
+        self,
+        path_or_buffer: FilePath | WriteBuffer[bytes] | WriteBuffer[str] | None = ...,
+        index: bool = ...,
+        root_name: str | None = ...,
+        row_name: str | None = ...,
+        na_rep: str | None = ...,
+        attr_cols: list[str] | None = ...,
+        elem_cols: list[str] | None = ...,
+        namespaces: dict[str | None, str] | None = ...,
+        prefix: str | None = ...,
+        encoding: str = ...,
+        xml_declaration: bool | None = ...,
+        pretty_print: bool | None = ...,
+        parser: str | None = ...,
+        stylesheet: FilePath | ReadBuffer[str] | ReadBuffer[bytes] | None = ...,
+        compression: CompressionOptions = ...,
+        storage_options: StorageOptions = ...,
+    ) -> str | None: ...
     def info(
         self, verbose=..., buf=..., max_cols=..., memory_usage=..., null_counts=...
     ) -> None: ...
@@ -2116,18 +2144,6 @@ class DataFrame(NDFrame, OpsMixin):
         compression: _str | Literal["infer", "gzip", "bz2", "zip", "xz"] = ...,
         protocol: int = ...,
     ) -> None: ...
-    def to_sql(
-        self,
-        name: _str,
-        con,
-        schema: _str | None = ...,
-        if_exists: _str = ...,
-        index: _bool = ...,
-        index_label: _str | Sequence[_str] | None = ...,
-        chunksize: int | None = ...,
-        dtype: dict | Scalar | None = ...,
-        method: _str | Callable | None = ...,
-    ) -> None: ...
     @overload
     def to_string(
         self,

pandas-stubs/core/generic.pyi

@@ -21,12 +21,12 @@ from pandas._typing import (
     ArrayLike,
     Axis,
     Dtype,
+    DtypeArg,
     FilePathOrBuffer,
     FrameOrSeries,
     FrameOrSeriesUnion,
     IgnoreRaise,
     Level,
-    Scalar,
     SeriesAxisType,
     T,
 )
@@ -152,13 +152,13 @@ class NDFrame(PandasObject, indexing.IndexingMixin):
     def to_sql(
         self,
         name: _str,
-        con,
+        con: str | Any,
         schema: _str | None = ...,
         if_exists: _str = ...,
         index: _bool = ...,
-        index_label: _str | Sequence[_str] | None = ...,
+        index_label: Hashable | Sequence[Hashable] | None = ...,
         chunksize: int | None = ...,
-        dtype: dict | Scalar | None = ...,
+        dtype: DtypeArg | None = ...,
         method: _str | Callable | None = ...,
     ) -> None: ...
     def to_pickle(

pandas-stubs/io/clipboards.pyi

@@ -1,6 +1,10 @@
 from __future__ import annotations
 
+from typing import Any
+
 from pandas.core.frame import DataFrame
 
-def read_clipboard(sep: str = ..., **kwargs) -> DataFrame: ...
-def to_clipboard(obj, excel: bool = ..., sep=..., **kwargs) -> None: ...
+def read_clipboard(sep: str = ..., **kwargs: Any) -> DataFrame: ...
+def to_clipboard(
+    obj, excel: bool = ..., sep: str | None = ..., **kwargs: Any
+) -> None: ...

pandas-stubs/io/gbq.pyi

@@ -14,10 +14,10 @@ def read_gbq(
     dialect: str | None = ...,
     location: str | None = ...,
     configuration: dict[str, Any] | None = ...,
-    credentials=...,
+    # Google type, not available
+    credentials: Any = ...,
     use_bqstorage_api: bool | None = ...,
-    private_key=...,
-    verbose=...,
+    max_results: int | None = ...,
     progress_bar_type: str | None = ...,
 ) -> DataFrame: ...
 def to_gbq(
@@ -31,7 +31,6 @@ def to_gbq(
     table_schema: list[dict[str, str]] | None = ...,
     location: str | None = ...,
     progress_bar: bool = ...,
-    credentials=...,
-    verbose=...,
-    private_key=...,
+    # Google type, not available
+    credentials: Any = ...,
 ) -> None: ...