Merge branch 'main' into fix-zip-strict-indexing

Author: kei-0021

ci/code_checks.sh

@@ -81,7 +81,6 @@ if [[ -z "$CHECK" || "$CHECK" == "docstrings" ]]; then
         -i "pandas.tseries.offsets.CustomBusinessMonthBegin PR02" \
         -i "pandas.tseries.offsets.CustomBusinessMonthEnd PR02" \
         -i "pandas.tseries.offsets.Easter.is_on_offset GL08" \
-        -i "pandas.tseries.offsets.FY5253.is_on_offset GL08" \
         -i "pandas.tseries.offsets.FY5253Quarter.is_on_offset GL08" \
         -i "pandas.tseries.offsets.FY5253Quarter.rule_code GL08" \
         -i "pandas.tseries.offsets.LastWeekOfMonth.is_on_offset GL08" \

ci/deps/actions-311.yaml

@@ -42,7 +42,7 @@ dependencies:
   - pyqt>=5.15.9
   - openpyxl>=3.1.5
   - psycopg2>=2.9.10
-  - pyarrow>=13.0.0
+  - pyarrow>=13.0.0,<=22.0.0 # https://github.com/apache/arrow/issues/49003
   - pyiceberg>=0.8.1
   - pydantic<2.12.0  # TMP pin to avoid pyiceberg/pydantic issues
   - pymysql>=1.1.1

ci/deps/actions-312.yaml

@@ -42,7 +42,7 @@ dependencies:
   - pyqt>=5.15.9
   - openpyxl>=3.1.5
   - psycopg2>=2.9.10
-  - pyarrow>=13.0.0
+  - pyarrow>=13.0.0,<=22.0.0 # https://github.com/apache/arrow/issues/49003
   - pyiceberg>=0.8.1
   - pydantic<2.12.0  # TMP pin to avoid pyiceberg/pydantic issues
   - pymysql>=1.1.1

ci/deps/actions-313-downstream_compat.yaml

@@ -41,7 +41,7 @@ dependencies:
   - qtpy>=2.4.2
   - openpyxl>=3.1.5
   - psycopg2>=2.9.10
-  - pyarrow>=13.0.0
+  - pyarrow>=13.0.0,<=22.0.0 # https://github.com/apache/arrow/issues/49003
   - pyiceberg>=0.8.1
   - pymysql>=1.1.1
   - pyqt>=5.15.9

ci/deps/actions-313.yaml

@@ -43,7 +43,7 @@ dependencies:
   - pyqt>=5.15.9
   - openpyxl>=3.1.5
   - psycopg2>=2.9.10
-  - pyarrow>=13.0.0
+  - pyarrow>=13.0.0,<=22.0.0 # https://github.com/apache/arrow/issues/49003
   - pymysql>=1.1.1
   - pyreadstat>=1.2.8
   - pytables>=3.10.1

pandas/_libs/tslibs/offsets.pyx

@@ -5328,6 +5328,38 @@ cdef class FY5253(FY5253Mixin):
     _attributes = tuple(["n", "normalize", "weekday", "startingMonth", "variation"])
 
     def is_on_offset(self, dt: datetime) -> bool:
+        """
+        Return boolean whether a timestamp intersects with this frequency.
+
+        This method checks if a given datetime falls on a fiscal year end date
+        as defined by the 52-53 week fiscal year calendar.
+
+        Parameters
+        ----------
+        dt : datetime
+            Timestamp to check.
+
+        Returns
+        -------
+        bool
+            True if the timestamp is on the fiscal year end, False otherwise.
+
+        See Also
+        --------
+        FY5253Quarter.is_on_offset : Check if timestamp is on fiscal quarter end.
+        DateOffset.is_on_offset : Check if timestamp intersects with frequency.
+
+        Examples
+        --------
+        >>> offset = pd.offsets.FY5253(weekday=4, startingMonth=12, variation="last")
+        >>> ts = pd.Timestamp(2022, 12, 30)
+        >>> offset.is_on_offset(ts)
+        True
+
+        >>> ts = pd.Timestamp(2022, 12, 31)
+        >>> offset.is_on_offset(ts)
+        False
+        """
         if self.normalize and not _is_normalized(dt):
             return False
         dt = datetime(dt.year, dt.month, dt.day)

pandas/core/dtypes/common.py

@@ -627,8 +627,8 @@ def is_string_dtype(arr_or_dtype) -> bool:
 
     See Also
     --------
-    api.types.is_string_dtype : Check whether the provided array or dtype
-                                is of the string dtype.
+    api.types.is_object_dtype : Check whether an array-like or dtype is of the
+        object dtype.
 
     Examples
     --------

pandas/tests/extension/test_arrow.py

@@ -36,8 +36,6 @@
 from pandas._libs.tslibs import timezones
 from pandas.compat import (
     PY312,
-    is_ci_environment,
-    is_platform_windows,
     pa_version_under14p0,
     pa_version_under19p0,
     pa_version_under20p0,
@@ -72,18 +70,6 @@
 from pandas.core.arrays.arrow.extension_types import ArrowPeriodType
 
 
-def _require_timezone_database(request):
-    if is_platform_windows() and is_ci_environment():
-        mark = pytest.mark.xfail(
-            raises=pa.ArrowInvalid,
-            reason=(
-                "TODO: Set ARROW_TIMEZONE_DATABASE environment variable "
-                "on CI to path to the tzdata for pyarrow."
-            ),
-        )
-        request.applymarker(mark)
-
-
 @pytest.fixture(params=tm.ALL_PYARROW_DTYPES, ids=str)
 def dtype(request):
     return ArrowDtype(pyarrow_dtype=request.param)
@@ -363,10 +349,6 @@ def test_from_sequence_pa_array_notimplemented(self, request):
             ArrowExtensionArray._from_sequence_of_strings(["12-1"], dtype=dtype)
 
     def test_from_sequence_of_strings_pa_array(self, data, request):
-        pa_dtype = data.dtype.pyarrow_dtype
-        if pa.types.is_timestamp(pa_dtype) and pa_dtype.tz is not None:
-            _require_timezone_database(request)
-
         pa_array = data._pa_array.cast(pa.string())
         result = type(data)._from_sequence_of_strings(pa_array, dtype=data.dtype)
         tm.assert_extension_array_equal(result, data)
@@ -2608,17 +2590,13 @@ def test_dt_isocalendar():
 )
 def test_dt_day_month_name(method, exp, request):
     # GH 52388
-    _require_timezone_database(request)
-
     ser = pd.Series([datetime(2023, 1, 1), None], dtype=ArrowDtype(pa.timestamp("ms")))
     result = getattr(ser.dt, method)()
     expected = pd.Series([exp, None], dtype=ArrowDtype(pa.string()))
     tm.assert_series_equal(result, expected)
 
 
 def test_dt_strftime(request):
-    _require_timezone_database(request)
-
     ser = pd.Series(
         [datetime(year=2023, month=1, day=2, hour=3), None],
         dtype=ArrowDtype(pa.timestamp("ns")),
@@ -2705,8 +2683,6 @@ def test_dt_tz_localize_unsupported_tz_options():
 
 
 def test_dt_tz_localize_none(request):
-    _require_timezone_database(request)
-
     ser = pd.Series(
         [datetime(year=2023, month=1, day=2, hour=3), None],
         dtype=ArrowDtype(pa.timestamp("ns", tz="US/Pacific")),
@@ -2721,8 +2697,6 @@ def test_dt_tz_localize_none(request):
 
 @pytest.mark.parametrize("unit", ["us", "ns"])
 def test_dt_tz_localize(unit, request):
-    _require_timezone_database(request)
-
     ser = pd.Series(
         [datetime(year=2023, month=1, day=2, hour=3), None],
         dtype=ArrowDtype(pa.timestamp(unit)),
@@ -2744,8 +2718,6 @@ def test_dt_tz_localize(unit, request):
     ],
 )
 def test_dt_tz_localize_nonexistent(nonexistent, exp_date, request):
-    _require_timezone_database(request)
-
     ser = pd.Series(
         [datetime(year=2023, month=3, day=12, hour=2, minute=30), None],
         dtype=ArrowDtype(pa.timestamp("ns")),

pandas/tests/interchange/test_impl.py

@@ -7,10 +7,6 @@
 import pytest
 
 from pandas._libs.tslibs import iNaT
-from pandas.compat import (
-    is_ci_environment,
-    is_platform_windows,
-)
 
 import pandas as pd
 import pandas._testing as tm
@@ -380,21 +376,11 @@ def test_datetimetzdtype(tz, unit):
         tm.assert_frame_equal(df, from_dataframe(df.__dataframe__()))
 
 
-def test_interchange_from_non_pandas_tz_aware(request):
+def test_interchange_from_non_pandas_tz_aware():
     # GH 54239, 54287
     pa = pytest.importorskip("pyarrow", "11.0.0")
     import pyarrow.compute as pc
 
-    if is_platform_windows() and is_ci_environment():
-        mark = pytest.mark.xfail(
-            raises=pa.ArrowInvalid,
-            reason=(
-                "TODO: Set ARROW_TIMEZONE_DATABASE environment variable "
-                "on CI to path to the tzdata for pyarrow."
-            ),
-        )
-        request.applymarker(mark)
-
     arr = pa.array([datetime(2020, 1, 1), None, datetime(2020, 1, 2)])
     arr = pc.assume_timezone(arr, "Asia/Kathmandu")
     table = pa.table({"arr": arr})

pandas/tests/resample/test_datetime_index.py

@@ -8,7 +8,6 @@
 from pandas._libs import lib
 from pandas._libs.tslibs import Day
 from pandas._typing import DatetimeNaTType
-from pandas.compat import is_platform_windows
 from pandas.errors import Pandas4Warning
 import pandas.util._test_decorators as td
 
@@ -2138,16 +2137,7 @@ def test_resample_b_55282(unit):
 @td.skip_if_no("pyarrow")
 @pytest.mark.parametrize(
     "tz",
-    [
-        None,
-        pytest.param(
-            "UTC",
-            marks=pytest.mark.xfail(
-                condition=is_platform_windows(),
-                reason="TODO: Set ARROW_TIMEZONE_DATABASE env var in CI",
-            ),
-        ),
-    ],
+    [None, "UTC"],
 )
 def test_arrow_timestamp_resample(tz):
     # GH 56371