Merge remote-tracking branch 'upstream/main' into convert-dtype-pyarrow

Author: yuanx749

.github/workflows/wheels.yml

@@ -153,7 +153,7 @@ jobs:
         run: echo "sdist_name=$(cd ./dist && ls -d */)" >> "$GITHUB_ENV"
 
       - name: Build wheels
-        uses: pypa/[email protected].0
+        uses: pypa/[email protected].1
         with:
          package-dir: ./dist/${{ startsWith(matrix.buildplat[1], 'macosx') && env.sdist_name || needs.build_sdist.outputs.sdist_file }}
         env:

ci/code_checks.sh

@@ -72,13 +72,9 @@ if [[ -z "$CHECK" || "$CHECK" == "docstrings" ]]; then
         -i "pandas.Series.dt PR01" `# Accessors are implemented as classes, but we do not document the Parameters section` \
         -i "pandas.Period.freq GL08" \
         -i "pandas.Period.ordinal GL08" \
-        -i "pandas.Timedelta.max PR02" \
-        -i "pandas.Timedelta.min PR02" \
-        -i "pandas.Timedelta.resolution PR02" \
         -i "pandas.Timestamp.max PR02" \
         -i "pandas.Timestamp.min PR02" \
         -i "pandas.Timestamp.resolution PR02" \
-        -i "pandas.Timestamp.tzinfo GL08" \
         -i "pandas.core.groupby.DataFrameGroupBy.plot PR02" \
         -i "pandas.core.groupby.SeriesGroupBy.plot PR02" \
         -i "pandas.core.resample.Resampler.quantile PR01,PR07" \

doc/source/whatsnew/v3.0.0.rst

@@ -704,6 +704,7 @@ Indexing
 - Bug in :meth:`Index.get_indexer` and similar methods when ``NaN`` is located at or after position 128 (:issue:`58924`)
 - Bug in :meth:`MultiIndex.insert` when a new value inserted to a datetime-like level gets cast to ``NaT`` and fails indexing (:issue:`60388`)
 - Bug in printing :attr:`Index.names` and :attr:`MultiIndex.levels` would not escape single quotes (:issue:`60190`)
+- Bug in reindexing of :class:`DataFrame` with :class:`PeriodDtype` columns in case of consolidated block (:issue:`60980`, :issue:`60273`)
 
 Missing
 ^^^^^^^
@@ -839,6 +840,7 @@ Other
 - Bug in :meth:`Series.diff` allowing non-integer values for the ``periods`` argument. (:issue:`56607`)
 - Bug in :meth:`Series.dt` methods in :class:`ArrowDtype` that were returning incorrect values. (:issue:`57355`)
 - Bug in :meth:`Series.isin` raising ``TypeError`` when series is large (>10**6) and ``values`` contains NA (:issue:`60678`)
+- Bug in :meth:`Series.mode` where an exception was raised when taking the mode with nullable types with no null values in the series. (:issue:`58926`)
 - Bug in :meth:`Series.rank` that doesn't preserve missing values for nullable integers when ``na_option='keep'``. (:issue:`56976`)
 - Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` inconsistently replacing matching instances when ``regex=True`` and missing values are present. (:issue:`56599`)
 - Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` throwing ``ValueError`` when ``regex=True`` and all NA values. (:issue:`60688`)

pandas/_libs/hashtable_func_helper.pxi.in

@@ -430,7 +430,7 @@ def mode(ndarray[htfunc_t] values, bint dropna, const uint8_t[:] mask=None):
 
     if na_counter > 0:
         res_mask = np.zeros(j+1, dtype=np.bool_)
-        res_mask[j] = True
+        res_mask[j] = (na_counter == max_count)
     return modes[:j + 1], res_mask
 
 

pandas/_libs/tslibs/timedeltas.pyx

@@ -998,8 +998,9 @@ class MinMaxReso:
     and Timedelta class.  On an instance, these depend on the object's _reso.
     On the class, we default to the values we would get with nanosecond _reso.
     """
-    def __init__(self, name):
+    def __init__(self, name, docstring):
         self._name = name
+        self.__doc__ = docstring
 
     def __get__(self, obj, type=None):
         if self._name == "min":
@@ -1012,9 +1013,13 @@ class MinMaxReso:
 
         if obj is None:
             # i.e. this is on the class, default to nanos
-            return Timedelta(val)
+            result = Timedelta(val)
         else:
-            return Timedelta._from_value_and_reso(val, obj._creso)
+            result = Timedelta._from_value_and_reso(val, obj._creso)
+
+        result.__doc__ = self.__doc__
+
+        return result
 
     def __set__(self, obj, value):
         raise AttributeError(f"{self._name} is not settable.")
@@ -1033,9 +1038,75 @@ cdef class _Timedelta(timedelta):
 
     # higher than np.ndarray and np.matrix
     __array_priority__ = 100
-    min = MinMaxReso("min")
-    max = MinMaxReso("max")
-    resolution = MinMaxReso("resolution")
+
+    _docstring_min = """
+    Returns the minimum bound possible for Timedelta.
+
+    This property provides access to the smallest possible value that
+    can be represented by a Timedelta object.
+
+    Returns
+    -------
+    Timedelta
+
+    See Also
+    --------
+    Timedelta.max: Returns the maximum bound possible for Timedelta.
+    Timedelta.resolution: Returns the smallest possible difference between
+        non-equal Timedelta objects.
+
+    Examples
+    --------
+    >>> pd.Timedelta.min
+    -106752 days +00:12:43.145224193
+    """
+
+    _docstring_max = """
+    Returns the maximum bound possible for Timedelta.
+
+    This property provides access to the largest possible value that
+    can be represented by a Timedelta object.
+
+    Returns
+    -------
+    Timedelta
+
+    See Also
+    --------
+    Timedelta.min: Returns the minimum bound possible for Timedelta.
+    Timedelta.resolution: Returns the smallest possible difference between
+        non-equal Timedelta objects.
+
+    Examples
+    --------
+    >>> pd.Timedelta.max
+    106751 days 23:47:16.854775807
+    """
+
+    _docstring_reso = """
+    Returns the smallest possible difference between non-equal Timedelta objects.
+
+    The resolution value is determined by the underlying representation of time
+    units and is equivalent to Timedelta(nanoseconds=1).
+
+    Returns
+    -------
+    Timedelta
+
+    See Also
+    --------
+    Timedelta.max: Returns the maximum bound possible for Timedelta.
+    Timedelta.min: Returns the minimum bound possible for Timedelta.
+
+    Examples
+    --------
+    >>> pd.Timedelta.resolution
+    0 days 00:00:00.000000001
+    """
+
+    min = MinMaxReso("min", _docstring_min)
+    max = MinMaxReso("max", _docstring_max)
+    resolution = MinMaxReso("resolution", _docstring_reso)
 
     @property
     def value(self):

pandas/_libs/tslibs/timestamps.pyx

@@ -2208,6 +2208,34 @@ class Timestamp(_Timestamp):
         """
         return super().tzname()
 
+    @property
+    def tzinfo(self):
+        """
+        Returns the timezone info of the Timestamp.
+
+        This property returns a `datetime.tzinfo` object if the Timestamp
+        is timezone-aware. If the Timestamp has no timezone, it returns `None`.
+        If the Timestamp is in UTC or a fixed-offset timezone,
+        it returns `datetime.timezone`. If the Timestamp uses an
+        IANA timezone (e.g., "America/New_York"), it returns `zoneinfo.ZoneInfo`.
+
+        See Also
+        --------
+        Timestamp.tz : Alias for `tzinfo`, may return a `zoneinfo.ZoneInfo` object.
+        Timestamp.tz_convert : Convert timezone-aware Timestamp to another time zone.
+        Timestamp.tz_localize : Localize the Timestamp to a specific timezone.
+
+        Examples
+        --------
+        >>> ts = pd.Timestamp("2023-01-01 12:00:00", tz="UTC")
+        >>> ts.tzinfo
+        datetime.timezone.utc
+
+        >>> ts_naive = pd.Timestamp("2023-01-01 12:00:00")
+        >>> ts_naive.tzinfo
+        """
+        return super().tzinfo
+
     def utcoffset(self):
         """
         Return utc offset.

pandas/core/algorithms.py

@@ -987,7 +987,7 @@ def duplicated(
 
 def mode(
     values: ArrayLike, dropna: bool = True, mask: npt.NDArray[np.bool_] | None = None
-) -> ArrayLike:
+) -> tuple[np.ndarray, npt.NDArray[np.bool_]] | ExtensionArray:
     """
     Returns the mode(s) of an array.
 
@@ -1000,7 +1000,7 @@ def mode(
 
     Returns
     -------
-    np.ndarray or ExtensionArray
+    Union[Tuple[np.ndarray, npt.NDArray[np.bool_]], ExtensionArray]
     """
     values = _ensure_arraylike(values, func_name="mode")
     original = values
@@ -1014,8 +1014,10 @@ def mode(
     values = _ensure_data(values)
 
     npresult, res_mask = htable.mode(values, dropna=dropna, mask=mask)
-    if res_mask is not None:
-        return npresult, res_mask  # type: ignore[return-value]
+    if res_mask is None:
+        res_mask = np.zeros(npresult.shape, dtype=np.bool_)
+    else:
+        return npresult, res_mask
 
     try:
         npresult = safe_sort(npresult)
@@ -1026,7 +1028,7 @@ def mode(
         )
 
     result = _reconstruct_data(npresult, original.dtype, original)
-    return result
+    return result, res_mask
 
 
 def rank(

pandas/core/arrays/base.py

@@ -2511,8 +2511,9 @@ def _mode(self, dropna: bool = True) -> Self:
             Sorted, if possible.
         """
         # error: Incompatible return value type (got "Union[ExtensionArray,
-        # ndarray[Any, Any]]", expected "Self")
-        return mode(self, dropna=dropna)  # type: ignore[return-value]
+        # Tuple[np.ndarray, npt.NDArray[np.bool_]]", expected "Self")
+        result, _ = mode(self, dropna=dropna)
+        return result  # type: ignore[return-value]
 
     def __array_ufunc__(self, ufunc: np.ufunc, method: str, *inputs, **kwargs):
         if any(

pandas/core/arrays/categorical.py

@@ -2477,7 +2477,7 @@ def _mode(self, dropna: bool = True) -> Categorical:
         if dropna:
             mask = self.isna()
 
-        res_codes = algorithms.mode(codes, mask=mask)
+        res_codes, _ = algorithms.mode(codes, mask=mask)
         res_codes = cast(np.ndarray, res_codes)
         assert res_codes.dtype == codes.dtype
         res = self._from_backing_data(res_codes)

pandas/core/arrays/datetimelike.py

@@ -1635,7 +1635,7 @@ def _mode(self, dropna: bool = True):
         if dropna:
             mask = self.isna()
 
-        i8modes = algorithms.mode(self.view("i8"), mask=mask)
+        i8modes, _ = algorithms.mode(self.view("i8"), mask=mask)
         npmodes = i8modes.view(self._ndarray.dtype)
         npmodes = cast(np.ndarray, npmodes)
         return self._from_backing_data(npmodes)

pandas/core/arrays/masked.py

@@ -1099,12 +1099,8 @@ def value_counts(self, dropna: bool = True) -> Series:
         return Series(arr, index=index, name="count", copy=False)
 
     def _mode(self, dropna: bool = True) -> Self:
-        if dropna:
-            result = mode(self._data, dropna=dropna, mask=self._mask)
-            res_mask = np.zeros(result.shape, dtype=np.bool_)
-        else:
-            result, res_mask = mode(self._data, dropna=dropna, mask=self._mask)
-        result = type(self)(result, res_mask)  # type: ignore[arg-type]
+        result, res_mask = mode(self._data, dropna=dropna, mask=self._mask)
+        result = type(self)(result, res_mask)
         return result[result.argsort()]
 
     @doc(ExtensionArray.equals)

pandas/core/internals/blocks.py

@@ -155,12 +155,9 @@ class Block(PandasObject, libinternals.Block):
     def _validate_ndim(self) -> bool:
         """
         We validate dimension for blocks that can hold 2D values, which for now
-        means numpy dtypes or DatetimeTZDtype.
+        means numpy dtypes or EA dtypes like DatetimeTZDtype and PeriodDtype.
         """
-        dtype = self.dtype
-        return not isinstance(dtype, ExtensionDtype) or isinstance(
-            dtype, DatetimeTZDtype
-        )
+        return not is_1d_only_ea_dtype(self.dtype)
 
     @final
     @cache_readonly

pandas/core/series.py

@@ -2071,7 +2071,7 @@ def mode(self, dropna: bool = True) -> Series:
         # TODO: Add option for bins like value_counts()
         values = self._values
         if isinstance(values, np.ndarray):
-            res_values = algorithms.mode(values, dropna=dropna)
+            res_values, _ = algorithms.mode(values, dropna=dropna)
         else:
             res_values = values._mode(dropna=dropna)
 

pandas/tests/indexing/test_indexing.py

@@ -763,6 +763,27 @@ def test_partial_boolean_frame_indexing(self):
         )
         tm.assert_frame_equal(result, expected)
 
+    def test_period_column_slicing(self):
+        # GH#60273 The transpose operation creates a single 5x1 block of PeriodDtype
+        # Make sure it is reindexed correctly
+        df = DataFrame(
+            pd.period_range("2021-01-01", periods=5, freq="D"),
+            columns=["A"],
+        ).T
+        result = df[[0, 1, 2]]
+        expected = DataFrame(
+            [
+                [
+                    pd.Period("2021-01-01", freq="D"),
+                    pd.Period("2021-01-02", freq="D"),
+                    pd.Period("2021-01-03", freq="D"),
+                ]
+            ],
+            index=["A"],
+            columns=[0, 1, 2],
+        )
+        tm.assert_frame_equal(result, expected)
+
     def test_no_reference_cycle(self):
         df = DataFrame({"a": [0, 1], "b": [2, 3]})
         for name in ("loc", "iloc", "at", "iat"):

pandas/tests/internals/test_internals.py

@@ -1320,6 +1320,16 @@ def test_period_can_hold_element(self, element):
         with pytest.raises(TypeError, match="Invalid value"):
             self.check_series_setitem(elem, pi, False)
 
+    def test_period_reindex_axis(self):
+        # GH#60273 Test reindexing of block with PeriodDtype
+        pi = period_range("2020", periods=5, freq="Y")
+        blk = new_block(pi._data.reshape(5, 1), BlockPlacement(slice(5)), ndim=2)
+        mgr = BlockManager(blocks=(blk,), axes=[Index(np.arange(5)), Index(["a"])])
+        reindexed = mgr.reindex_axis(Index([0, 2, 4]), axis=0)
+        result = DataFrame._from_mgr(reindexed, axes=reindexed.axes)
+        expected = DataFrame([[pi[0], pi[2], pi[4]]], columns=[0, 2, 4], index=["a"])
+        tm.assert_frame_equal(result, expected)
+
     def check_can_hold_element(self, obj, elem, inplace: bool):
         blk = obj._mgr.blocks[0]
         if inplace:

pandas/tests/series/test_reductions.py

@@ -51,6 +51,29 @@ def test_mode_nullable_dtype(any_numeric_ea_dtype):
     tm.assert_series_equal(result, expected)
 
 
+def test_mode_nullable_dtype_edge_case(any_numeric_ea_dtype):
+    # GH##58926
+    ser = Series([1, 2, 3, 1], dtype=any_numeric_ea_dtype)
+    result = ser.mode(dropna=False)
+    expected = Series([1], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+    ser2 = Series([1, 1, 2, 3, pd.NA], dtype=any_numeric_ea_dtype)
+    result = ser2.mode(dropna=False)
+    expected = Series([1], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+    ser3 = Series([1, pd.NA, pd.NA], dtype=any_numeric_ea_dtype)
+    result = ser3.mode(dropna=False)
+    expected = Series([pd.NA], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+    ser4 = Series([1, 1, pd.NA, pd.NA], dtype=any_numeric_ea_dtype)
+    result = ser4.mode(dropna=False)
+    expected = Series([1, pd.NA], dtype=any_numeric_ea_dtype)
+    tm.assert_series_equal(result, expected)
+
+
 def test_mode_infer_string():
     # GH#56183
     pytest.importorskip("pyarrow")