Merge branch 'main' of https://github.com/pacificdragon/pandas

Author: pacificdragon

.circleci/config.yml

@@ -26,20 +26,21 @@ jobs:
       image: ubuntu-2004:202101-01
     resource_class: arm.large
     environment:
-      TAG = << pipeline.git.tag >>
-      TRIGGER_SOURCE = << pipeline.trigger_source >>
+      ENV_FILE: ci/deps/circle-38-arm64.yaml
+      TRIGGER_SOURCE: << pipeline.trigger_source >>
     steps:
       - checkout
       - run:
           name: Check if build is necessary
           command: |
             # Check if tag is defined or TRIGGER_SOURCE is scheduled
-            if [[ -n ${TAG} ]]; then
-                export IS_PUSH="true"
+            if [[ -n "$CIRCLE_TAG" ]]; then
+                echo 'export IS_PUSH="true"' >> "$BASH_ENV"
             elif [[ $TRIGGER_SOURCE == "scheduled_pipeline" ]]; then
-                export IS_SCHEDULE_DISPATCH="true"
+                echo 'export IS_SCHEDULE_DISPATCH="true"' >> "$BASH_ENV"
             # Look for the build label/[wheel build] in commit
-            elif (git log --format=oneline -n 1 $CIRCLE_SHA1) | grep -q '[wheel build]'; then
+            # grep takes a regex, so need to escape brackets
+            elif (git log --format=oneline -n 1 $CIRCLE_SHA1) | grep -q '\[wheel build\]'; then
                 : # Do nothing
             elif ! (curl https://api.github.com/repos/pandas-dev/pandas/issues/$CIRCLE_PR_NUMBER | jq '.labels' | grep -q 'Build'); then
                 circleci-agent step halt
@@ -51,15 +52,24 @@ jobs:
             cibuildwheel --output-dir wheelhouse
           environment:
             CIBW_BUILD: << parameters.cibw-build >>
+
       - run:
-          name: Upload wheels
+          name: Install Anaconda Client & Upload Wheels
           command: |
-            if [[ -n ${TAG} ]]; then
-                export IS_PUSH="true"
-            fi
-            if [[ $TRIGGER_SOURCE == "scheduled_pipeline" ]]; then
-                export IS_SCHEDULE_DISPATCH="true"
-            fi
+            echo "Install Mambaforge"
+            MAMBA_URL="https://github.com/conda-forge/miniforge/releases/download/4.14.0-0/Mambaforge-4.14.0-0-Linux-aarch64.sh"
+            echo "Downloading $MAMBA_URL"
+            wget -q $MAMBA_URL -O minimamba.sh
+            chmod +x minimamba.sh
+
+            MAMBA_DIR="$HOME/miniconda3"
+            rm -rf $MAMBA_DIR
+            ./minimamba.sh -b -p $MAMBA_DIR
+
+            export PATH=$MAMBA_DIR/bin:$PATH
+
+            mamba install -y -c conda-forge anaconda-client
+
             source ci/upload_wheels.sh
             set_upload_vars
             upload_wheels

.pre-commit-config.yaml

@@ -15,15 +15,22 @@ default_stages: [
 ci:
     autofix_prs: false
 repos:
+-   repo: local
+    hooks:
+    # NOTE: we make `black` a local hook because if it's installed from
+    # PyPI (rather than from source) then it'll run twice as fast thanks to mypyc
+    -   id: black
+        name: black
+        description: "Black: The uncompromising Python code formatter"
+        entry: black
+        language: python
+        require_serial: true
+        types_or: [python, pyi]
+        additional_dependencies: [black==23.1.0]
 -   repo: https://github.com/charliermarsh/ruff-pre-commit
     rev: v0.0.244
     hooks:
     -   id: ruff
--   repo: https://github.com/MarcoGorelli/absolufy-imports
-    rev: v0.3.1
-    hooks:
-    -   id: absolufy-imports
-        files: ^pandas/
 -   repo: https://github.com/jendrikseipp/vulture
     rev: 'v2.7'
     hooks:
@@ -116,16 +123,6 @@ repos:
     - id: sphinx-lint
 -   repo: local
     hooks:
-    # NOTE: we make `black` a local hook because if it's installed from
-    # PyPI (rather than from source) then it'll run twice as fast thanks to mypyc
-    -   id: black
-        name: black
-        description: "Black: The uncompromising Python code formatter"
-        entry: black
-        language: python
-        require_serial: true
-        types_or: [python, pyi]
-        additional_dependencies: [black==22.10.0]
     -   id: pyright
         # note: assumes python env is setup and activated
         name: pyright

asv_bench/benchmarks/algorithms.py

@@ -15,7 +15,6 @@
 
 
 class Factorize:
-
     params = [
         [True, False],
         [True, False],
@@ -65,7 +64,6 @@ def time_factorize(self, unique, sort, dtype):
 
 
 class Duplicated:
-
     params = [
         [True, False],
         ["first", "last", False],
@@ -96,7 +94,6 @@ def time_duplicated(self, unique, keep, dtype):
 
 
 class DuplicatedMaskedArray:
-
     params = [
         [True, False],
         ["first", "last", False],

asv_bench/benchmarks/algos/isin.py

@@ -12,7 +12,6 @@
 
 
 class IsIn:
-
     params = [
         "int64",
         "uint64",
@@ -183,7 +182,6 @@ def time_isin(self, dtype, M, offset_factor):
 
 
 class IsInFloat64:
-
     params = [
         [np.float64, "Float64"],
         ["many_different_values", "few_different_values", "only_nans_values"],

asv_bench/benchmarks/arithmetic.py

@@ -173,7 +173,6 @@ def time_op_same_blocks(self, op, shape):
 
 
 class Ops:
-
     params = [[True, False], ["default", 1]]
     param_names = ["use_numexpr", "threads"]
 
@@ -257,7 +256,6 @@ def time_frame_series_dot(self):
 
 
 class Timeseries:
-
     params = [None, "US/Eastern"]
     param_names = ["tz"]
 
@@ -316,7 +314,6 @@ def time_categorical_op(self, op):
 
 
 class IndexArithmetic:
-
     params = ["float", "int"]
     param_names = ["dtype"]
 
@@ -387,7 +384,6 @@ def time_add_timedeltas(self, df):
 
 
 class AddOverflowScalar:
-
     params = [1, -1, 0]
     param_names = ["scalar"]
 
@@ -455,7 +451,6 @@ def time_add_overflow_both_arg_nan(self):
 
 
 class OffsetArrayArithmetic:
-
     params = offsets
     param_names = ["offset"]
 

asv_bench/benchmarks/array.py

@@ -72,7 +72,6 @@ def time_from_list(self):
 
 
 class ArrowStringArray:
-
     params = [False, True]
     param_names = ["multiple_chunks"]
 
@@ -107,7 +106,6 @@ def time_tolist(self, multiple_chunks):
 
 
 class ArrowExtensionArray:
-
     params = [
         [
             "boolean[pyarrow]",

asv_bench/benchmarks/attrs_caching.py

@@ -22,7 +22,6 @@ def time_set_index(self):
 
 
 class SeriesArrayAttribute:
-
     params = [["numeric", "object", "category", "datetime64", "datetime64tz"]]
     param_names = ["dtype"]
 

asv_bench/benchmarks/categoricals.py

@@ -143,7 +143,6 @@ def time_concat_non_overlapping_index(self):
 
 
 class ValueCounts:
-
     params = [True, False]
     param_names = ["dropna"]
 
@@ -254,7 +253,6 @@ def time_categorical_contains(self):
 
 
 class CategoricalSlicing:
-
     params = ["monotonic_incr", "monotonic_decr", "non_monotonic"]
     param_names = ["index"]
 

asv_bench/benchmarks/ctors.py

@@ -49,7 +49,6 @@ def list_of_lists_with_none(arr):
 
 
 class SeriesConstructors:
-
     param_names = ["data_fmt", "with_index", "dtype"]
     params = [
         [
@@ -124,7 +123,6 @@ def time_multiindex_from_iterables(self):
 
 class DatetimeIndexConstructor:
     def setup(self):
-
         N = 20_000
         dti = date_range("1900-01-01", periods=N)
 

asv_bench/benchmarks/dtypes.py

@@ -49,7 +49,6 @@ def time_pandas_dtype_invalid(self, dtype):
 
 
 class SelectDtypes:
-
     try:
         params = [
             tm.ALL_INT_NUMPY_DTYPES

asv_bench/benchmarks/eval.py

@@ -9,7 +9,6 @@
 
 
 class Eval:
-
     params = [["numexpr", "python"], [1, "all"]]
     param_names = ["engine", "threads"]
 

asv_bench/benchmarks/frame_ctor.py

@@ -74,7 +74,6 @@ def time_mi_series(self):
 
 
 class FromDictwithTimestamp:
-
     params = [Nano(1), Hour(1)]
     param_names = ["offset"]
 
@@ -89,7 +88,6 @@ def time_dict_with_timestamp_offsets(self, offset):
 
 
 class FromRecords:
-
     params = [None, 1000]
     param_names = ["nrows"]
 
@@ -116,7 +114,6 @@ def time_frame_from_ndarray(self):
 
 
 class FromLists:
-
     goal_time = 0.2
 
     def setup(self):
@@ -129,7 +126,6 @@ def time_frame_from_lists(self):
 
 
 class FromRange:
-
     goal_time = 0.2
 
     def setup(self):
@@ -162,7 +158,6 @@ def time_frame_from_scalar_ea_float64_na(self):
 
 
 class FromArrays:
-
     goal_time = 0.2
 
     def setup(self):

asv_bench/benchmarks/frame_methods.py

@@ -371,7 +371,6 @@ def time_isnull_obj(self):
 
 
 class Fillna:
-
     params = (
         [True, False],
         ["pad", "bfill"],
@@ -412,7 +411,6 @@ def time_frame_fillna(self, inplace, method, dtype):
 
 
 class Dropna:
-
     params = (["all", "any"], [0, 1])
     param_names = ["how", "axis"]
 
@@ -432,7 +430,6 @@ def time_dropna_axis_mixed_dtypes(self, how, axis):
 
 
 class Count:
-
     params = [0, 1]
     param_names = ["axis"]
 
@@ -531,7 +528,6 @@ def time_frame_object_unequal(self):
 
 
 class Interpolate:
-
     params = [None, "infer"]
     param_names = ["downcast"]
 
@@ -616,7 +612,6 @@ def time_frame_duplicated_subset(self):
 
 
 class XS:
-
     params = [0, 1]
     param_names = ["axis"]
 
@@ -629,7 +624,6 @@ def time_frame_xs(self, axis):
 
 
 class SortValues:
-
     params = [True, False]
     param_names = ["ascending"]
 
@@ -657,7 +651,6 @@ def time_frame_sort_values_by_columns(self):
 
 
 class Quantile:
-
     params = [0, 1]
     param_names = ["axis"]
 
@@ -697,7 +690,6 @@ def time_info(self):
 
 
 class NSort:
-
     params = ["first", "last", "all"]
     param_names = ["keep"]