DEPS: remove most of distutils usage pandas-dev#41199

Author: fangchenli

pandas/compat/_optional.py

@@ -1,11 +1,12 @@
 from __future__ import annotations
 
-import distutils.version
 import importlib
 import sys
 import types
 import warnings
 
+from packaging.version import Version
+
 # Update install.rst when updating versions!
 
 VERSIONS = {
@@ -128,7 +129,7 @@ def import_optional_dependency(
     minimum_version = min_version if min_version is not None else VERSIONS.get(parent)
     if minimum_version:
         version = get_version(module_to_get)
-        if distutils.version.LooseVersion(version) < minimum_version:
+        if Version(version) < Version(minimum_version):
             msg = (
                 f"Pandas requires version '{minimum_version}' or newer of '{parent}' "
                 f"(version '{version}' currently installed)."

pandas/compat/numpy/__init__.py

@@ -1,21 +1,21 @@
 """ support numpy compatibility across versions """
 
-from distutils.version import LooseVersion
 import re
 
 import numpy as np
+from packaging.version import Version
 
 # numpy versioning
 _np_version = np.__version__
-_nlv = LooseVersion(_np_version)
-np_version_under1p18 = _nlv < LooseVersion("1.18")
-np_version_under1p19 = _nlv < LooseVersion("1.19")
-np_version_under1p20 = _nlv < LooseVersion("1.20")
+_nlv = Version(_np_version)
+np_version_under1p18 = _nlv < Version("1.18")
+np_version_under1p19 = _nlv < Version("1.19")
+np_version_under1p20 = _nlv < Version("1.20")
 is_numpy_dev = ".dev" in str(_nlv)
 _min_numpy_ver = "1.17.3"
 
 
-if _nlv < _min_numpy_ver:
+if _nlv < Version(_min_numpy_ver):
     raise ImportError(
         f"this version of pandas is incompatible with numpy < {_min_numpy_ver}\n"
         f"your numpy version is {_np_version}.\n"

pandas/compat/numpy/function.py

@@ -15,7 +15,6 @@
 methods that are spread throughout the codebase. This module will make it
 easier to adjust to future upstream changes in the analogous numpy signatures.
 """
-from distutils.version import LooseVersion
 from typing import (
     Any,
     Dict,
@@ -27,6 +26,7 @@
     __version__,
     ndarray,
 )
+from packaging.version import Version
 
 from pandas._libs.lib import (
     is_bool,
@@ -128,7 +128,7 @@ def validate_argmax_with_skipna(skipna, args, kwargs):
 ARGSORT_DEFAULTS["kind"] = "quicksort"
 ARGSORT_DEFAULTS["order"] = None
 
-if LooseVersion(__version__) >= LooseVersion("1.17.0"):
+if Version(__version__) >= Version("1.17.0"):
     # GH-26361. NumPy added radix sort and changed default to None.
     ARGSORT_DEFAULTS["kind"] = None
 

pandas/core/arrays/_arrow_utils.py

@@ -1,12 +1,12 @@
-from distutils.version import LooseVersion
 import json
 
 import numpy as np
+from packaging.version import Version
 import pyarrow
 
 from pandas.core.arrays.interval import VALID_CLOSED
 
-_pyarrow_version_ge_015 = LooseVersion(pyarrow.__version__) >= LooseVersion("0.15")
+_pyarrow_version_ge_015 = Version(pyarrow.__version__) >= Version("0.15")
 
 
 def pyarrow_array_to_numpy_and_mask(arr, dtype):

pandas/core/arrays/string_arrow.py

@@ -1,6 +1,5 @@
 from __future__ import annotations
 
-from distutils.version import LooseVersion
 from typing import (
     TYPE_CHECKING,
     Any,
@@ -9,6 +8,7 @@
 )
 
 import numpy as np
+from packaging.version import Version
 
 from pandas._libs import (
     lib,
@@ -54,7 +54,7 @@
     # PyArrow backed StringArrays are available starting at 1.0.0, but this
     # file is imported from even if pyarrow is < 1.0.0, before pyarrow.compute
     # and its compute functions existed. GH38801
-    if LooseVersion(pa.__version__) >= "1.0.0":
+    if Version(pa.__version__) >= Version("1.0.0"):
         import pyarrow.compute as pc
 
         ARROW_CMP_FUNCS = {
@@ -222,7 +222,7 @@ def __init__(self, values):
     def _chk_pyarrow_available(cls) -> None:
         # TODO: maybe update import_optional_dependency to allow a minimum
         # version to be specified rather than use the global minimum
-        if pa is None or LooseVersion(pa.__version__) < "1.0.0":
+        if pa is None or Version(pa.__version__) < Version("1.0.0"):
             msg = "pyarrow>=1.0.0 is required for PyArrow backed StringArray."
             raise ImportError(msg)
 

pandas/core/computation/ops.py

@@ -5,7 +5,6 @@
 from __future__ import annotations
 
 from datetime import datetime
-from distutils.version import LooseVersion
 from functools import partial
 import operator
 from typing import (
@@ -14,6 +13,7 @@
 )
 
 import numpy as np
+from packaging.version import Version
 
 from pandas._libs.tslibs import Timestamp
 
@@ -623,7 +623,7 @@ def __init__(self, name: str):
 
         if name not in MATHOPS or (
             NUMEXPR_INSTALLED
-            and NUMEXPR_VERSION < LooseVersion("2.6.9")
+            and Version(NUMEXPR_VERSION) < Version("2.6.9")
             and name in ("floor", "ceil")
         ):
             raise ValueError(f'"{name}" is not a supported function')

pandas/core/util/numba_.py

@@ -1,5 +1,4 @@
 """Common utilities for Numba operations"""
-from distutils.version import LooseVersion
 import types
 from typing import (
     Callable,
@@ -9,6 +8,7 @@
 )
 
 import numpy as np
+from packaging.version import Version
 
 from pandas.compat._optional import import_optional_dependency
 from pandas.errors import NumbaUtilError
@@ -89,7 +89,7 @@ def jit_user_function(
     """
     numba = import_optional_dependency("numba")
 
-    if LooseVersion(numba.__version__) >= LooseVersion("0.49.0"):
+    if Version(numba.__version__) >= Version("0.49.0"):
         is_jitted = numba.extending.is_jitted(func)
     else:
         is_jitted = isinstance(func, numba.targets.registry.CPUDispatcher)

pandas/io/clipboard/__init__.py

@@ -51,9 +51,9 @@
     get_errno,
     sizeof,
 )
-import distutils.spawn
 import os
 import platform
+from shutil import which
 import subprocess
 import time
 import warnings
@@ -528,7 +528,7 @@ def determine_clipboard():
         return init_windows_clipboard()
 
     if platform.system() == "Linux":
-        if distutils.spawn.find_executable("wslconfig.exe"):
+        if which("wslconfig.exe"):
             return init_wsl_clipboard()
 
     # Setup for the MAC OS X platform:

pandas/io/excel/_base.py

@@ -2,7 +2,6 @@
 
 import abc
 import datetime
-from distutils.version import LooseVersion
 import inspect
 from io import BytesIO
 import os
@@ -15,6 +14,8 @@
 import warnings
 import zipfile
 
+from packaging.version import Version
+
 from pandas._config import config
 
 from pandas._libs.parsers import STR_NA_VALUES
@@ -1147,7 +1148,7 @@ def __init__(
         else:
             import xlrd
 
-            xlrd_version = LooseVersion(get_version(xlrd))
+            xlrd_version = Version(get_version(xlrd))
 
         ext = None
         if engine is None:

pandas/io/parquet.py

@@ -1,7 +1,6 @@
 """ parquet compat """
 from __future__ import annotations
 
-from distutils.version import LooseVersion
 import io
 import os
 from typing import (
@@ -10,6 +9,8 @@
 )
 from warnings import catch_warnings
 
+from packaging.version import Version
+
 from pandas._typing import (
     FilePathOrBuffer,
     StorageOptions,
@@ -210,7 +211,7 @@ def read(
 
         to_pandas_kwargs = {}
         if use_nullable_dtypes:
-            if LooseVersion(self.api.__version__) >= "0.16":
+            if Version(self.api.__version__) >= Version("0.16"):
                 import pandas as pd
 
                 mapping = {

pandas/io/sql.py

@@ -11,7 +11,6 @@
     datetime,
     time,
 )
-from distutils.version import LooseVersion
 from functools import partial
 import re
 from typing import (
@@ -24,6 +23,7 @@
 import warnings
 
 import numpy as np
+from packaging.version import Version
 
 import pandas._libs.lib as lib
 from pandas._typing import DtypeArg
@@ -83,7 +83,7 @@ def _gt14() -> bool:
     """
     import sqlalchemy
 
-    return LooseVersion(sqlalchemy.__version__) >= LooseVersion("1.4.0")
+    return Version(sqlalchemy.__version__) >= Version("1.4.0")
 
 
 def _convert_params(sql, params):

pandas/plotting/_matplotlib/compat.py

@@ -1,7 +1,8 @@
 # being a bit too dynamic
-from distutils.version import LooseVersion
 import operator
 
+from packaging.version import Version
+
 
 def _mpl_version(version, op):
     def inner():
@@ -10,7 +11,7 @@ def inner():
         except ImportError:
             return False
         return (
-            op(LooseVersion(mpl.__version__), LooseVersion(version))
+            op(Version(mpl.__version__), Version(version))
             and str(mpl.__version__)[0] != "0"
         )
 

pandas/tests/computation/test_compat.py

@@ -1,5 +1,4 @@
-from distutils.version import LooseVersion
-
+from packaging.version import Version
 import pytest
 
 from pandas.compat._optional import VERSIONS
@@ -18,7 +17,7 @@ def test_compat():
         import numexpr as ne
 
         ver = ne.__version__
-        if LooseVersion(ver) < LooseVersion(VERSIONS["numexpr"]):
+        if Version(ver) < Version(VERSIONS["numexpr"]):
             assert not NUMEXPR_INSTALLED
         else:
             assert NUMEXPR_INSTALLED

pandas/tests/computation/test_eval.py

@@ -1,4 +1,3 @@
-from distutils.version import LooseVersion
 from functools import reduce
 from itertools import product
 import operator
@@ -10,6 +9,7 @@
 import warnings
 
 import numpy as np
+from packaging.version import Version
 import pytest
 
 from pandas.errors import PerformanceWarning
@@ -78,14 +78,14 @@ def parser(request):
 
 @pytest.fixture
 def ne_lt_2_6_9():
-    if NUMEXPR_INSTALLED and NUMEXPR_VERSION >= LooseVersion("2.6.9"):
+    if NUMEXPR_INSTALLED and NUMEXPR_VERSION >= Version("2.6.9"):
         pytest.skip("numexpr is >= 2.6.9")
     return "numexpr"
 
 
 def _get_unary_fns_for_ne():
     if NUMEXPR_INSTALLED:
-        if NUMEXPR_VERSION >= LooseVersion("2.6.9"):
+        if NUMEXPR_VERSION >= Version("2.6.9"):
             return list(_unary_math_ops)
         else:
             return [x for x in _unary_math_ops if x not in ["floor", "ceil"]]

pandas/tests/io/excel/__init__.py

@@ -1,5 +1,4 @@
-from distutils.version import LooseVersion
-
+from packaging.version import Version
 import pytest
 
 from pandas.compat._optional import (
@@ -32,4 +31,4 @@
 else:
     import xlrd
 
-    xlrd_version = LooseVersion(get_version(xlrd))
+    xlrd_version = Version(get_version(xlrd))

pandas/tests/io/generate_legacy_storage_files.py

@@ -33,13 +33,13 @@
 """
 
 from datetime import timedelta
-from distutils.version import LooseVersion
 import os
 import pickle
 import platform as pl
 import sys
 
 import numpy as np
+from packaging.version import Version
 
 import pandas
 from pandas import (
@@ -88,7 +88,7 @@
     from pandas.core.sparse.api import SparseArray
 
 
-_loose_version = LooseVersion(pandas.__version__)
+_loose_version = Version(pandas.__version__)
 
 
 def _create_sp_series():
@@ -155,7 +155,7 @@ def create_data():
 
     index["range"] = RangeIndex(10)
 
-    if _loose_version >= LooseVersion("0.21"):
+    if _loose_version >= Version("0.21"):
         from pandas import interval_range
 
         index["interval"] = interval_range(0, periods=10)

pandas/tests/io/pytables/test_select.py

@@ -1,7 +1,7 @@
-from distutils.version import LooseVersion
 from warnings import catch_warnings
 
 import numpy as np
+from packaging.version import Version
 import pytest
 
 from pandas._libs.tslibs import Timestamp
@@ -862,7 +862,7 @@ def test_select_as_multiple(setup_path):
 
 
 @pytest.mark.skipif(
-    LooseVersion(tables.__version__) < LooseVersion("3.1.0"),
+    Version(tables.__version__) < Version("3.1.0"),
     reason=("tables version does not support fix for nan selection bug: GH 4858"),
 )
 def test_nan_selection_bug_4858(setup_path):