♻️ Move to Pandas v3 (#103)

Author: ddelange

.github/workflows/CI.yml

@@ -16,12 +16,12 @@ jobs:
       id-token: write  # codecov/codecov-action
     strategy:
       matrix:
-        python-version: ['3.9', '3.14']  # sync with make targets below
+        python-version: ['3.11', '3.14']  # sync with make targets below
 
     steps:
     - uses: actions/checkout@v6
       with:
-        fetch-depth: 0
+        fetch-depth: 0  # for setuptools_scm
 
     - name: Set up Python
       uses: actions/setup-python@v6

.github/workflows/codeql-analysis.yml

@@ -1,7 +1,7 @@
 ---
 repos:
 - repo: https://github.com/astral-sh/ruff-pre-commit
-  rev: v0.14.7
+  rev: v0.15.2
   hooks:
   - id: ruff-check
     args: [--fix, --exit-non-zero-on-fix]
@@ -66,7 +66,7 @@ repos:
   - id: debug-statements
 
 - repo: https://github.com/python-jsonschema/check-jsonschema
-  rev: 0.35.0
+  rev: 0.36.2
   hooks:
   - id: check-dependabot
   - id: check-github-workflows

.pre-commit-config.yaml

@@ -10,7 +10,7 @@ version: 2
 build:
   os: ubuntu-24.04
   tools:
-    python: '3.12'
+    python: '3.14'
 
 # Build documentation in the docs/ directory with Sphinx
 sphinx:

.readthedocs.yaml

@@ -83,7 +83,7 @@
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
 # ones.
 extensions = [
-    "recommonmark",
+    "myst_parser",
     "sphinx_rtd_theme",
     "sphinx.ext.autodoc",
     "sphinx.ext.napoleon",
@@ -92,8 +92,7 @@
 ]
 autodoc_typehints = "description"
 
-# recommonmark extension allows mixed filetypes
-source_suffix = [".rst", ".md"]
+suppress_warnings = ["myst.header"]
 
 # Add any paths that contain templates here, relative to this directory.
 # templates_path = ["_templates"]
@@ -169,7 +168,7 @@ def linkcode_resolve(  # noqa: C901
     for part in fullname.split("."):
         try:
             obj = getattr(obj, part)
-        except Exception:  # noqa: PERF203
+        except Exception:
             return None
 
     # strip decorators, which would resolve to the source of the decorator

docs/source/conf.py

@@ -8,16 +8,14 @@ description = "Sensible multi-core apply function for Pandas"
 readme = "README.md"
 urls = {Repository = "https://github.com/ddelange/mapply", Documentation = "https://mapply.readthedocs.io"}
 authors = [{name = "ddelange", email = "ddelange@delange.dev"}]
-requires-python = ">=3.9" # sync with classifiers below, and tool.ruff and tool.mypy
+requires-python = ">=3.11" # sync with classifiers below, and tool.ruff and tool.mypy
 license = "BSD-3-Clause"
 classifiers = [
   "Development Status :: 5 - Production/Stable",
   "Intended Audience :: Developers",
   "Operating System :: OS Independent",
   "Programming Language :: Python",
   "Programming Language :: Python :: 3",
-  "Programming Language :: Python :: 3.9",
-  "Programming Language :: Python :: 3.10",
   "Programming Language :: Python :: 3.11",
   "Programming Language :: Python :: 3.12",
   "Programming Language :: Python :: 3.13",
@@ -32,7 +30,7 @@ branch = true
 omit = ["site-packages"]
 
 [tool.mypy]
-python_version = "3.9"
+python_version = "3.11"
 ignore_missing_imports = true
 warn_no_return = false
 disallow_untyped_defs = false
@@ -45,8 +43,7 @@ check-return-types = false
 check-yield-types = false
 
 [tool.pytest.ini_options]
-addopts = "-s --strict-markers -vv --cache-clear --doctest-modules --cov=mapply --cov-report=term --cov-report=html --cov-branch --no-cov-on-fail --ignore=docs"
-filterwarnings = ["ignore::DeprecationWarning"]
+addopts = "-s --strict-markers -vv --cache-clear --doctest-modules --cov=mapply --cov-report=term --cov-report=html --cov-report=xml --cov-branch --no-cov-on-fail --ignore=docs"
 
 [tool.ruff]
 output-format = "concise"

pyproject.toml

@@ -1,5 +1,5 @@
 detect-secrets~=1.4
 mypy~=1.6
-pre-commit~=3.5
-pytest-cov~=4.1
-pytest~=7.4
+pre-commit~=4.5
+pytest-cov~=7.0
+pytest~=9.0

requirements/ci.txt

@@ -1,3 +1,3 @@
-recommonmark~=0.7
-sphinx-rtd-theme~=1.3
-sphinx~=7.0
+myst-parser~=5.0
+sphinx-rtd-theme~=3.0
+sphinx~=8.0

requirements/docs.txt

@@ -2,4 +2,4 @@ pathos>=0.3.1  # https://github.com/uqfoundation/pathos/pull/252
 multiprocess
 psutil
 tqdm>=4.27  # from tqdm.auto import tqdm
-pandas~=2.1
+pandas~=3.0

requirements/prod.txt

@@ -32,8 +32,9 @@
 # SPDX-License-Identifier: BSD-3-Clause
 # ruff: noqa: ERA001
 import logging
+from collections.abc import Callable
 from types import MethodType
-from typing import Any, Callable
+from typing import Any
 
 from pandas.core.groupby.ops import _is_indexed_like
 
@@ -51,18 +52,18 @@ def run_groupwise_apply(
     args: tuple[Any, ...] = (),
     **kwargs: Any,
 ):
-    """Patch GroupBy.grouper.apply, applying func to each group in parallel."""
+    """Patch GroupBy._grouper.apply_groupwise, applying func to each group in parallel."""
 
-    def apply(self, f, data, axis=0):
-        # patching https://github.com/pandas-dev/pandas/blob/v2.1.4/pandas/core/groupby/ops.py#L890
+    def apply(self, f, data):
+        # patching https://github.com/pandas-dev/pandas/blob/v3.0.1/pandas/core/groupby/ops.py#L1014
         # with a multiprocessing_imap
         mutated = False
-        splitter = self._get_splitter(data, axis=axis)
-        group_keys = self.group_keys_seq
+        splitter = self._get_splitter(data)
+        group_keys = self.result_index
         result_values = []
 
         # This calls DataSplitter.__iter__
-        zipped = zip(group_keys, splitter)
+        zipped = zip(group_keys, splitter, strict=True)
 
         # rewrite the original for-loop into an imap
         def _run_apply(args):
@@ -91,12 +92,13 @@ def _run_apply(args):
         # original for-loop leftover
         for res, group_axes in zipped:
             # no changes made below this line
-            if not mutated and not _is_indexed_like(res, group_axes, axis):
+            if not mutated and not _is_indexed_like(res, group_axes):
                 mutated = True
             result_values.append(res)
         # getattr pattern for __name__ is needed for functools.partial objects
         if len(group_keys) == 0 and getattr(f, "__name__", None) in [
             "skew",
+            "kurt",
             "sum",
             "prod",
         ]:
@@ -109,9 +111,10 @@ def _run_apply(args):
 
     # overwrite apply method and restore after execution
     attr = "apply_groupwise"
-    original_apply = getattr(df_or_series.grouper, attr)
-    setattr(df_or_series.grouper, attr, MethodType(apply, df_or_series.grouper))
+    grouper = df_or_series._grouper  # noqa: SLF001
+    original_apply = getattr(grouper, attr)
+    setattr(grouper, attr, MethodType(apply, grouper))
     try:
         return df_or_series.apply(func, *args, **kwargs)
     finally:
-        setattr(df_or_series.grouper, attr, original_apply)
+        setattr(grouper, attr, original_apply)