Merge branch 'main' of github.com:mongodb-labs/mongo-arrow into python-infer-schema

Author: blink1073

.github/workflows/release-python.yml

@@ -32,15 +32,11 @@ jobs:
         - [ubuntu-20.04, manylinux_aarch64]
         - [macos-14, macosx_*]
         - [windows-2019, win_amd64]
-        python: ["cp38", "cp39", "cp310", "cp311", "cp312"]
+        python: ["cp39", "cp310", "cp311", "cp312"]
         exclude:
-          - buildplat: [macos-14, macosx_*]
-            python: "cp38"
           - buildplat: [macos-14, macosx_*]
             python: "cp39"
         include:
-          - buildplat: [macos-12, macosx_*]
-            python: "cp38"
           - buildplat: [macos-12, macosx_*]
             python: "cp39"
 
@@ -75,13 +71,6 @@ jobs:
       - name: Install cibuildwheel
         run: python -m pip install "cibuildwheel>=2.4,<3"
 
-      - name: Build MacOS Py38 Wheel
-        if: ${{ matrix.python == 'cp38' && matrix.buildplat[0] == 'macos-11' }}
-        env:
-          CIBW_BUILD: cp38-macosx_x86_64
-          MACOSX_DEPLOYMENT_TARGET: "10.14"
-        run: python -m cibuildwheel --output-dir wheelhouse
-
       - name: Build MacOS Py39 Wheels
         if: ${{ matrix.python == 'cp39' && matrix.buildplat[0] == 'macos-11' }}
         env:
@@ -114,7 +103,7 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           # Build sdist on lowest supported Python
-          python-version: '3.8'
+          python-version: '3.9'
 
       - name: Install tox
         run: |

.github/workflows/test-python.yml

@@ -33,7 +33,7 @@ jobs:
     strategy:
       matrix:
         os: ["ubuntu-latest", "macos-latest", "windows-latest"]
-        python-version: [3.8, 3.9, "3.10", "3.11", "3.12"]
+        python-version: ["3.9", "3.10", "3.11", "3.12"]
       fail-fast: false
     name: CPython ${{ matrix.python-version }}-${{ matrix.os }}
     steps:

bindings/python/CHANGELOG.md

@@ -3,6 +3,10 @@
 
 ---
 
+# Changes in Version 1.6.0
+
+- Drop support for Python 3.8.
+
 # Changes in Version 1.5.2
 
 - Fix support for PyMongo 4.9.

bindings/python/CONTRIBUTING.md

@@ -13,7 +13,7 @@ be of interest or that has already been addressed.
 
 ## Supported Interpreters
 
-PyMongoArrow supports CPython 3.7+ and PyPy3.8+. Language features not
+PyMongoArrow supports CPython 3.9+ and PyPy3.9+. Language features not
 supported by all interpreters can not be used.
 
 ## Style Guide

bindings/python/pymongoarrow/pandas_types.py

@@ -13,10 +13,10 @@
 # limitations under the License.
 
 # Pandas Extension Types
+from __future__ import annotations
 
 import numbers
 import re
-from typing import Type, Union
 
 import numpy as np
 import pandas as pd
@@ -38,7 +38,7 @@ class PandasBSONDtype(ExtensionDtype):
     def name(self) -> str:
         return f"bson_{self.__class__.__name__}"
 
-    def __from_arrow__(self, array: Union[pa.Array, pa.ChunkedArray]) -> ExtensionArray:
+    def __from_arrow__(self, array: pa.Array | pa.ChunkedArray) -> ExtensionArray:
         chunks = [array] if isinstance(array, pa.Array) else array.chunks
 
         arr_type = self.construct_array_type()
@@ -204,7 +204,7 @@ def name(self) -> str:
         return f"bson_{self.type.__name__}[{self.subtype}]"
 
     @classmethod
-    def construct_array_type(cls) -> Type["PandasBinaryArray"]:
+    def construct_array_type(cls) -> type[PandasBinaryArray]:
         return PandasBinaryArray
 
     @classmethod
@@ -242,7 +242,7 @@ class PandasObjectId(PandasBSONDtype):
     type = ObjectId
 
     @classmethod
-    def construct_array_type(cls) -> Type["PandasObjectIdArray"]:
+    def construct_array_type(cls) -> type[PandasObjectIdArray]:
         return PandasObjectIdArray
 
 
@@ -266,7 +266,7 @@ class PandasDecimal128(PandasBSONDtype):
     type = Decimal128
 
     @classmethod
-    def construct_array_type(cls) -> Type["PandasDecimal128Array"]:
+    def construct_array_type(cls) -> type[PandasDecimal128Array]:
         return PandasDecimal128Array
 
 
@@ -290,7 +290,7 @@ class PandasCode(PandasBSONDtype):
     type = Code
 
     @classmethod
-    def construct_array_type(cls) -> Type["PandasCodeArray"]:
+    def construct_array_type(cls) -> type[PandasCodeArray]:
         return PandasCodeArray
 
 

bindings/python/pyproject.toml

@@ -23,7 +23,6 @@ classifiers = [
     "Operating System :: POSIX",
     "Programming Language :: Python :: 3",
     "Programming Language :: Python :: 3 :: Only",
-    "Programming Language :: Python :: 3.8",
     "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
@@ -32,7 +31,7 @@ classifiers = [
     "Topic :: Database",
 ]
 readme = "README.md"
-requires-python = ">=3.8"
+requires-python = ">=3.9"
 dependencies = [
     # Must be kept in sync with "build_sytem.requires" above.
     "pyarrow >=17.0,<17.1",

bindings/python/test/test_arrow.py

@@ -114,7 +114,7 @@ def test_find_simple(self):
         self.assertEqual(find_cmd.command_name, "find")
         self.assertEqual(find_cmd.command["projection"], {"_id": True, "data": True})
 
-    def test_find_repeat_dtype(self):
+    def test_find_repeat_type(self):
         expected = Table.from_pydict(
             {"_id": [1, 2, 3, 4], "data": [10, 20, 30, None]},
             ArrowSchema([("_id", int32()), ("data", int32())]),

bindings/python/test/test_builders.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import calendar
-from datetime import date, datetime, timedelta
+from datetime import date, datetime, timedelta, timezone
 from unittest import TestCase
 
 from bson import Binary, Code, Decimal128, ObjectId
@@ -86,7 +86,7 @@ def test_simple(self):
         self.maxDiff = None
 
         builder = DatetimeBuilder(dtype=timestamp("ms"))
-        datetimes = [datetime.utcnow() + timedelta(days=k * 100) for k in range(5)]
+        datetimes = [datetime.now(timezone.utc) + timedelta(days=k * 100) for k in range(5)]
         builder.append(self._datetime_to_millis(datetimes[0]))
         builder.append_values([self._datetime_to_millis(k) for k in datetimes[1:]])
         builder.append_null()
@@ -97,7 +97,9 @@ def test_simple(self):
         self.assertEqual(len(arr), len(datetimes) + 1)
         for actual, expected in zip(arr, datetimes + [None]):
             if actual.is_valid:
-                self.assertEqual(actual.as_py(), self._millis_only(expected))
+                self.assertEqual(
+                    actual.as_py().timetuple(), self._millis_only(expected).timetuple()
+                )
             else:
                 self.assertIsNone(expected)
         self.assertEqual(arr.type, timestamp("ms"))

bindings/python/test/test_datetime.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import unittest
-from datetime import datetime, timedelta
+from datetime import datetime, timedelta, timezone
 from test import client_context
 
 import pytz
@@ -36,8 +36,8 @@ def setUp(self):
         self.coll.drop()
         self.coll.insert_many(
             [
-                {"_id": 1, "data": datetime.utcnow() + timedelta(milliseconds=10)},
-                {"_id": 2, "data": datetime.utcnow() + timedelta(milliseconds=25)},
+                {"_id": 1, "data": datetime.now(timezone.utc) + timedelta(milliseconds=10)},
+                {"_id": 2, "data": datetime.now(timezone.utc) + timedelta(milliseconds=25)},
             ]
         )
         self.expected_times = []

bindings/python/test/test_numpy.py

@@ -325,7 +325,7 @@ def table_from_dict(self, d, schema=None):
         out = {}
         for k, v in d.items():
             if any(isinstance(x, int) for x in v) and None in v:
-                out[k] = np.array(v, dtype=np.float_)
+                out[k] = np.array(v, dtype=np.float64)
             else:
                 out[k] = np.array(v, dtype=np.dtype(type(v[0])))  # Infer
         return out