Backport PR #41595: CI: Xfails on Python 3.10 (#42297)

Co-authored-by: Thomas Li <[email protected]>

.github/workflows/python-dev.yml

@@ -10,6 +10,12 @@ on:
     paths-ignore:
       - "doc/**"
 
+env:
+  PYTEST_WORKERS: "auto"
+  PANDAS_CI: 1
+  PATTERN: "not slow and not network and not clipboard"
+  COVERAGE: true
+
 jobs:
   build:
     runs-on: ubuntu-latest
@@ -32,7 +38,7 @@ jobs:
         pip install git+https://github.com/numpy/numpy.git
         pip install git+https://github.com/pytest-dev/pytest.git
         pip install git+https://github.com/nedbat/coveragepy.git
-        pip install cython python-dateutil pytz hypothesis pytest-xdist
+        pip install cython python-dateutil pytz hypothesis pytest-xdist pytest-cov
         pip list
 
     - name: Build Pandas
@@ -46,7 +52,8 @@ jobs:
 
     - name: Test with pytest
       run: |
-        coverage run -m pytest -m 'not slow and not network and not clipboard' pandas
+        ci/run_tests.sh
+      # GH 41935
       continue-on-error: true
 
     - name: Publish test results

pandas/tests/arrays/string_/test_string.py

@@ -2,9 +2,6 @@
 This module tests the functionality of StringArray and ArrowStringArray.
 Tests for the str accessors are in pandas/tests/strings/test_string_array.py
 """
-
-import re
-
 import numpy as np
 import pytest
 
@@ -314,7 +311,7 @@ def test_astype_int(dtype):
     tm.assert_numpy_array_equal(result, expected)
 
     arr = pd.array(["1", pd.NA, "3"], dtype=dtype)
-    msg = re.escape("int() argument must be a string, a bytes-like object or a number")
+    msg = r"int\(\) argument must be a string, a bytes-like object or a( real)? number"
     with pytest.raises(TypeError, match=msg):
         arr.astype("int64")
 

pandas/tests/extension/test_sparse.py

@@ -13,6 +13,7 @@
 be added to the array-specific tests in `pandas/tests/arrays/`.
 
 """
+
 import numpy as np
 import pytest
 

pandas/tests/indexes/test_base.py

@@ -12,6 +12,7 @@
 
 from pandas.compat import (
     IS64,
+    PY310,
     np_datetime64_compat,
 )
 from pandas.util._test_decorators import async_mark
@@ -992,7 +993,7 @@ def test_isin(self, values, index, expected):
         result = index.isin(values)
         tm.assert_numpy_array_equal(result, expected)
 
-    def test_isin_nan_common_object(self, nulls_fixture, nulls_fixture2):
+    def test_isin_nan_common_object(self, request, nulls_fixture, nulls_fixture2):
         # Test cartesian product of null fixtures and ensure that we don't
         # mangle the various types (save a corner case with PyPy)
 
@@ -1003,6 +1004,24 @@ def test_isin_nan_common_object(self, nulls_fixture, nulls_fixture2):
             and math.isnan(nulls_fixture)
             and math.isnan(nulls_fixture2)
         ):
+            if PY310:
+                if (
+                    # Failing cases are
+                    # np.nan, float('nan')
+                    # float('nan'), np.nan
+                    # float('nan'), float('nan')
+                    # Since only float('nan'), np.nan is float
+                    # Use not np.nan to identify float('nan')
+                    nulls_fixture is np.nan
+                    and nulls_fixture2 is not np.nan
+                    or nulls_fixture is not np.nan
+                ):
+                    request.applymarker(
+                        # This test is flaky :(
+                        pytest.mark.xfail(
+                            reason="Failing on Python 3.10 GH41940", strict=False
+                        )
+                    )
             tm.assert_numpy_array_equal(
                 Index(["a", nulls_fixture]).isin([nulls_fixture2]),
                 np.array([False, True]),

pandas/tests/io/json/test_ujson.py

@@ -16,6 +16,7 @@
 import pandas._libs.json as ujson
 from pandas.compat import (
     IS64,
+    PY310,
     is_platform_windows,
 )
 
@@ -248,7 +249,21 @@ def test_double_precision(self):
             assert rounded_input == json.loads(output)
             assert rounded_input == ujson.decode(output)
 
-    @pytest.mark.parametrize("invalid_val", [20, -1, "9", None])
+    @pytest.mark.parametrize(
+        "invalid_val",
+        [
+            20,
+            -1,
+            pytest.param(
+                "9",
+                marks=pytest.mark.xfail(PY310, reason="Failing on Python 3.10 GH41940"),
+            ),
+            pytest.param(
+                None,
+                marks=pytest.mark.xfail(PY310, reason="Failing on Python 3.10 GH41940"),
+            ),
+        ],
+    )
     def test_invalid_double_precision(self, invalid_val):
         double_input = 30.12345678901234567890
         expected_exception = ValueError if isinstance(invalid_val, int) else TypeError

pandas/tests/reshape/test_get_dummies.py

@@ -3,6 +3,8 @@
 import numpy as np
 import pytest
 
+from pandas.compat import PY310
+
 from pandas.core.dtypes.common import is_integer_dtype
 
 import pandas as pd
@@ -428,6 +430,8 @@ def test_dataframe_dummies_unicode(self, get_dummies_kwargs, expected):
         result = get_dummies(**get_dummies_kwargs)
         tm.assert_frame_equal(result, expected)
 
+    # This is flaky on Python 3.10
+    @pytest.mark.xfail(PY310, reason="Failing on Python 3.10 GH41940", strict=False)
     def test_get_dummies_basic_drop_first(self, sparse):
         # GH12402 Add a new parameter `drop_first` to avoid collinearity
         # Basic case
@@ -467,6 +471,7 @@ def test_get_dummies_basic_drop_first_one_level(self, sparse):
         result = get_dummies(s_series_index, drop_first=True, sparse=sparse)
         tm.assert_frame_equal(result, expected)
 
+    @pytest.mark.xfail(PY310, reason="Failing on Python 3.10 GH41940", strict=False)
     def test_get_dummies_basic_drop_first_NA(self, sparse):
         # Test NA handling together with drop_first
         s_NA = ["a", "b", np.nan]

pandas/tests/test_algos.py

@@ -9,7 +9,10 @@
     algos as libalgos,
     hashtable as ht,
 )
-from pandas.compat import np_array_datetime64_compat
+from pandas.compat import (
+    PY310,
+    np_array_datetime64_compat,
+)
 import pandas.util._test_decorators as td
 
 from pandas.core.dtypes.common import (
@@ -783,6 +786,8 @@ def test_different_nans(self):
         expected = np.array([np.nan])
         tm.assert_numpy_array_equal(result, expected)
 
+    # Flaky on Python 3.10 -> Don't make strict
+    @pytest.mark.xfail(PY310, reason="Failing on Python 3.10 GH41940", strict=False)
     def test_first_nan_kept(self):
         # GH 22295
         # create different nans from bit-patterns:
@@ -988,6 +993,8 @@ def __hash__(self):
         # different objects -> False
         tm.assert_numpy_array_equal(algos.isin([a], [b]), np.array([False]))
 
+    # Flaky on Python 3.10 -> Don't make strict
+    @pytest.mark.xfail(PY310, reason="Failing on Python 3.10 GH41940", strict=False)
     def test_different_nans(self):
         # GH 22160
         # all nans are handled as equivalent
@@ -1030,6 +1037,8 @@ def test_empty(self, empty):
         result = algos.isin(vals, empty)
         tm.assert_numpy_array_equal(expected, result)
 
+    # Flaky on Python 3.10 -> Don't make strict
+    @pytest.mark.xfail(PY310, reason="Failing on Python 3.10 GH41940", strict=False)
     def test_different_nan_objects(self):
         # GH 22119
         comps = np.array(["nan", np.nan * 1j, float("nan")], dtype=object)