Support for Python 3.12 (dask#10544)

.github/workflows/additional.yml

@@ -39,7 +39,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.9", "3.10", "3.11"]
+        python-version: ["3.9", "3.10", "3.11", "3.12"]
     steps:
       - name: Checkout source
         uses: actions/[email protected]

.github/workflows/tests.yml

@@ -24,13 +24,15 @@ jobs:
       fail-fast: false
       matrix:
         os: ["windows-latest", "ubuntu-latest", "macos-latest"]
-        environment: ["3.9", "3.10", "3.11"]
+        environment: ["3.9", "3.10", "3.11", "3.12"]
         extra: [null]
         exclude:
           - os: "macos-latest"
             environment: "3.9"
           - os: "macos-latest"
             environment: "3.10"
+          - os: "macos-latest"
+            environment: "3.11"
         include:
           # Minimum dependencies
           - os: "ubuntu-latest"
@@ -45,7 +47,7 @@ jobs:
             environment: "mindeps-optional"
           # Pyarrow strings turned on
           - os: "ubuntu-latest"
-            environment: "3.11"
+            environment: "3.12"
             extra: "pyarrow"
 
     env:

continuous_integration/environment-3.12.yaml

@@ -0,0 +1,79 @@
+# This job includes coverage
+name: test-environment
+channels:
+  - conda-forge
+  - nodefaults
+dependencies:
+  # required dependencies
+  - python=3.12
+  - packaging
+  - pyyaml
+  - click
+  - cloudpickle
+  - partd
+  - fsspec
+  - importlib_metadata
+  - toolz
+  # test dependencies
+  - pre-commit
+  - pytest
+  - pytest-cov
+  - pytest-rerunfailures
+  - pytest-timeout
+  - pytest-xdist
+  - moto
+  # Optional dependencies
+  - mimesis
+  - numpy
+  - pandas
+  - flask
+  - fastparquet>=0.8.0
+  - h5py
+  - pytables
+  - zarr
+  # `tiledb-py=0.17.5` lead to strange seg faults in CI, However 0.18 is needed for 3.11
+  # https://github.com/dask/dask/pull/9569
+  # - tiledb-py # crashes on Python 3.11
+  # - pyspark
+  # - tiledb>=2.5.0 # crashes on Python 3.11
+  - xarray
+  - sqlalchemy>=1.4.16
+  - pyarrow>=13
+  - coverage
+  - jsonschema
+  # # other -- IO
+  - boto3
+  - botocore
+  - bokeh
+  - httpretty
+  - aiohttp
+  - s3fs
+  # Need a new `crick` release with support for `numpy=1.24+`
+  # https://github.com/dask/crick/issues/25
+  # - crick
+  - cytoolz
+  - distributed
+  - ipython
+  - ipycytoscape
+  # until https://github.com/jupyter-widgets/ipywidgets/issues/3731 is fixed
+  - ipywidgets<8.0.5
+  - ipykernel<6.22.0
+  - lz4
+  # - numba no support for 3.12
+  - psutil
+  - requests
+  - scikit-image
+  - scikit-learn
+  - scipy
+  - python-snappy
+  - sparse
+  - cachey
+  - python-graphviz
+  # Not available on macOS for Python 3.12 yet
+  # - python-cityhash
+  - python-xxhash
+  - mmh3
+  - jinja2
+  - pip
+  - pip:
+    - git+https://github.com/dask/distributed

dask/array/tests/test_array_core.py

@@ -14,18 +14,21 @@
 import math
 import operator
 import os
+import sys
 import time
 import warnings
 from functools import reduce
 from io import StringIO
 from operator import add, sub
 from threading import Lock
 
+from packaging.version import Version
 from tlz import concat, merge
 from tlz.curried import identity
 
 import dask
 import dask.array as da
+from dask._compatibility import PY_VERSION
 from dask.array.chunk import getitem
 from dask.array.core import (
     Array,
@@ -3910,6 +3913,10 @@ def test_setitem_1d():
         dx[index] = 1
 
 
+@pytest.mark.xfail(
+    sys.platform == "win32" and PY_VERSION >= Version("3.12.0"),
+    reason="https://github.com/dask/dask/issues/10604",
+)
 def test_setitem_hardmask():
     x = np.ma.array([1, 2, 3, 4], dtype=int)
     x.harden_mask()

dask/dataframe/tests/test_multi.py

@@ -5,9 +5,11 @@
 import numpy as np
 import pandas as pd
 import pytest
+from packaging.version import Version
 from pandas.api.types import is_object_dtype
 
 import dask.dataframe as dd
+from dask._compatibility import PY_VERSION
 from dask.base import compute_as_if_collection
 from dask.dataframe._compat import (
     PANDAS_GE_140,
@@ -2135,7 +2137,7 @@ def test_concat5():
             False,
             True,
             marks=pytest.mark.xfail(
-                PANDAS_GE_220,
+                PANDAS_GE_220 or PY_VERSION >= Version("3.12.0"),
                 reason="fails on pandas dev: https://github.com/dask/dask/issues/10558",
                 raises=AssertionError,
                 strict=False,
@@ -2148,7 +2150,7 @@ def test_concat5():
             False,
             True,
             marks=pytest.mark.xfail(
-                PANDAS_GE_220,
+                PANDAS_GE_220 or PY_VERSION >= Version("3.12.0"),
                 reason="fails on pandas dev: https://github.com/dask/dask/issues/10558",
                 raises=AssertionError,
                 strict=False,

pyproject.toml

@@ -20,6 +20,7 @@ classifiers = [
     "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
     "Topic :: Scientific/Engineering",
     "Topic :: System :: Distributed Computing",
 ]