Merge branch 'pandas-dev:main' into main

LirongMa · web-flow · commit 70a8357689a6 · 2025-09-28T15:04:26.000-04:00
diff --git a/doc/source/whatsnew/v0.21.0.rst b/doc/source/whatsnew/v0.21.0.rst
@@ -635,22 +635,17 @@ Previous behavior:
 
 New behavior:
 
-.. code-block:: ipython
+.. ipython:: python
 
-   In [1]: pi = pd.period_range('2017-01', periods=12, freq='M')
+   pi = pd.period_range('2017-01', periods=12, freq='M')
 
-   In [2]: s = pd.Series(np.arange(12), index=pi)
+   s = pd.Series(np.arange(12), index=pi)
 
-   In [3]: resampled = s.resample('2Q').mean()
+   resampled = s.resample('2Q').mean()
 
-   In [4]: resampled
-   Out[4]:
-   2017Q1    2.5
-   2017Q3    8.5
-   Freq: 2Q-DEC, dtype: float64
+   resampled
 
-   In [5]: resampled.index
-   Out[5]: PeriodIndex(['2017Q1', '2017Q3'], dtype='period[2Q-DEC]')
+   resampled.index
 
 Upsampling and calling ``.ohlc()`` previously returned a ``Series``, basically identical to calling ``.asfreq()``. OHLC upsampling now returns a DataFrame with columns ``open``, ``high``, ``low`` and ``close`` (:issue:`13083`). This is consistent with downsampling and ``DatetimeIndex`` behavior.
 
diff --git a/doc/source/whatsnew/v2.2.0.rst b/doc/source/whatsnew/v2.2.0.rst
@@ -664,7 +664,7 @@ Other Deprecations
 - Deprecated :meth:`DatetimeArray.__init__` and :meth:`TimedeltaArray.__init__`, use :func:`array` instead (:issue:`55623`)
 - Deprecated :meth:`Index.format`, use ``index.astype(str)`` or ``index.map(formatter)`` instead (:issue:`55413`)
 - Deprecated :meth:`Series.ravel`, the underlying array is already 1D, so ravel is not necessary (:issue:`52511`)
-- Deprecated :meth:`Series.resample` and :meth:`DataFrame.resample` with a :class:`PeriodIndex` (and the 'convention' keyword), convert to :class:`DatetimeIndex` (with ``.to_timestamp()``) before resampling instead (:issue:`53481`)
+- Deprecated :meth:`Series.resample` and :meth:`DataFrame.resample` with a :class:`PeriodIndex` (and the 'convention' keyword), convert to :class:`DatetimeIndex` (with ``.to_timestamp()``) before resampling instead (:issue:`53481`). Note: this deprecation was later undone in pandas 2.3.3 (:issue:`57033`)
 - Deprecated :meth:`Series.view`, use :meth:`Series.astype` instead to change the dtype (:issue:`20251`)
 - Deprecated :meth:`offsets.Tick.is_anchored`, use ``False`` instead (:issue:`55388`)
 - Deprecated ``core.internals`` members ``Block``, ``ExtensionBlock``, and ``DatetimeTZBlock``, use public APIs instead (:issue:`55139`)
diff --git a/doc/source/whatsnew/v2.3.3.rst b/doc/source/whatsnew/v2.3.3.rst
@@ -57,6 +57,14 @@ Bug fixes
 
 - The :meth:`DataFrame.iloc` now works correctly with ``copy_on_write`` option when assigning values after subsetting the columns of a homogeneous DataFrame (:issue:`60309`)
 
+Other changes
+~~~~~~~~~~~~~
+
+- The deprecation of using :meth:`Series.resample` and :meth:`DataFrame.resample`
+  with a :class:`PeriodIndex` (and the 'convention' keyword) has been undone.
+  Resampling with a :class:`PeriodIndex` is supported again, but a subset of
+  methods that return incorrect results will raise an error in pandas 3.0 (:issue:`57033`)
+
 
 .. ---------------------------------------------------------------------------
 .. _whatsnew_233.contributors:
diff --git a/pandas/_typing.py b/pandas/_typing.py
@@ -83,8 +83,7 @@
 
     # numpy compatible types
     NumpyValueArrayLike: TypeAlias = ScalarLike_co | npt.ArrayLike
-    # Name "npt._ArrayLikeInt_co" is not defined  [name-defined]
-    NumpySorter: TypeAlias = npt._ArrayLikeInt_co | None  # type: ignore[name-defined]
+    NumpySorter: TypeAlias = npt._ArrayLikeInt_co | None
 
 
 P = ParamSpec("P")
diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
@@ -215,14 +215,15 @@ def _reconstruct_data(
         #  that values.dtype == dtype
         cls = dtype.construct_array_type()
 
-        # error: Incompatible types in assignment (expression has type
-        # "ExtensionArray", variable has type "ndarray[Any, Any]")
-        values = cls._from_sequence(values, dtype=dtype)  # type: ignore[assignment]
-
-    else:
-        values = values.astype(dtype, copy=False)
-
-    return values
+        # error: Incompatible return value type
+        # (got "ExtensionArray",
+        # expected "ndarray[tuple[Any, ...], dtype[Any]]")
+        return cls._from_sequence(values, dtype=dtype)  # type: ignore[return-value]
+
+    # error: Incompatible return value type
+    # (got "ndarray[tuple[Any, ...], dtype[Any]]",
+    # expected "ExtensionArray")
+    return values.astype(dtype, copy=False)  # type: ignore[return-value]
 
 
 def _ensure_arraylike(values, func_name: str) -> ArrayLike:
diff --git a/pandas/core/array_algos/quantile.py b/pandas/core/array_algos/quantile.py
@@ -102,7 +102,7 @@ def quantile_with_mask(
             interpolation=interpolation,
         )
 
-        result = np.asarray(result)  # type: ignore[assignment]
+        result = np.asarray(result)
         result = result.T
 
     return result
diff --git a/pandas/core/arrays/_mixins.py b/pandas/core/arrays/_mixins.py
@@ -151,7 +151,9 @@ def view(self, dtype: Dtype | None = None) -> ArrayLike:
 
             td64_values = arr.view(dtype)
             return TimedeltaArray._simple_new(td64_values, dtype=dtype)
-        return arr.view(dtype=dtype)
+        # error: Argument "dtype" to "view" of "ndarray" has incompatible type
+        # "ExtensionDtype | dtype[Any]"; expected "dtype[Any] | _HasDType[dtype[Any]]"
+        return arr.view(dtype=dtype)  # type: ignore[arg-type]
 
     def take(
         self,
diff --git a/pandas/core/arrays/arrow/_arrow_utils.py b/pandas/core/arrays/arrow/_arrow_utils.py
@@ -44,7 +44,7 @@ def pyarrow_array_to_numpy_and_mask(
         mask = pyarrow.BooleanArray.from_buffers(
             pyarrow.bool_(), len(arr), [None, bitmask], offset=arr.offset
         )
-        mask = np.asarray(mask)  # type: ignore[assignment]
+        mask = np.asarray(mask)
     else:
         mask = np.ones(len(arr), dtype=bool)
     return data, mask
diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
@@ -657,7 +657,7 @@ def _box_pa_array(
             ):
                 arr_value = np.asarray(value, dtype=object)
                 # similar to isna(value) but exclude NaN, NaT, nat-like, nan-like
-                mask = is_pdna_or_none(arr_value)  # type: ignore[assignment]
+                mask = is_pdna_or_none(arr_value)
 
             try:
                 pa_array = pa.array(value, type=pa_type, mask=mask)
@@ -2738,7 +2738,7 @@ def _str_get_dummies(self, sep: str = "|", dtype: NpDtype | None = None):
             dummies_dtype = np.bool_
         dummies = np.zeros(n_rows * n_cols, dtype=dummies_dtype)
         dummies[indices] = True
-        dummies = dummies.reshape((n_rows, n_cols))  # type: ignore[assignment]
+        dummies = dummies.reshape((n_rows, n_cols))
         result = self._from_pyarrow_array(pa.array(list(dummies)))
         return result, uniques_sorted.to_pylist()
 
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
@@ -1869,7 +1869,7 @@ def value_counts(self, dropna: bool = True) -> Series:
             count = np.bincount(obs, minlength=ncat or 0)
         else:
             count = np.bincount(np.where(mask, code, ncat))
-            ix = np.append(ix, -1)  # type: ignore[assignment]
+            ix = np.append(ix, -1)
 
         ix = coerce_indexer_dtype(ix, self.dtype.categories)
         ix_categorical = self._from_backing_data(ix)
diff --git a/pandas/core/arrays/datetimes.py b/pandas/core/arrays/datetimes.py
@@ -804,7 +804,11 @@ def _add_offset(self, offset: BaseOffset) -> Self:
         try:
             res_values = offset._apply_array(values._ndarray)
             if res_values.dtype.kind == "i":
-                res_values = res_values.view(values.dtype)
+                # error: Argument 1 to "view" of "ndarray" has
+                # incompatible type
+                # "dtype[datetime64[date | int | None]] | DatetimeTZDtype";
+                # expected "dtype[Any] | _HasDType[dtype[Any]]"  [arg-type]
+                res_values = res_values.view(values.dtype)  # type: ignore[arg-type]
         except NotImplementedError:
             if get_option("performance_warnings"):
                 warnings.warn(
diff --git a/pandas/core/arrays/string_.py b/pandas/core/arrays/string_.py
@@ -764,7 +764,7 @@ def _cast_pointwise_result(self, values) -> ArrayLike:
         result = super()._cast_pointwise_result(values)
         if isinstance(result.dtype, StringDtype):
             # Ensure we retain our same na_value/storage
-            result = result.astype(self.dtype)  # type: ignore[call-overload]
+            result = result.astype(self.dtype)
         return result
 
     @classmethod
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
@@ -8811,7 +8811,7 @@ def resample(
         rule,
         closed: Literal["right", "left"] | None = None,
         label: Literal["right", "left"] | None = None,
-        convention: Literal["start", "end", "s", "e"] | lib.NoDefault = lib.no_default,
+        convention: Literal["start", "end", "s", "e"] = "start",
         on: Level | None = None,
         level: Level | None = None,
         origin: str | TimestampConvertibleTypes = "start_day",
@@ -8841,9 +8841,6 @@ def resample(
         convention : {{'start', 'end', 's', 'e'}}, default 'start'
             For `PeriodIndex` only, controls whether to use the start or
             end of `rule`.
-
-            .. deprecated:: 2.2.0
-                Convert PeriodIndex to DatetimeIndex before resampling instead.
         on : str, optional
             For a DataFrame, column to use instead of index for resampling.
             Column must be datetime-like.
@@ -8999,6 +8996,55 @@ def resample(
         2000-01-01 00:06:00    26
         Freq: 3min, dtype: int64
 
+        For a Series with a PeriodIndex, the keyword `convention` can be
+        used to control whether to use the start or end of `rule`.
+
+        Resample a year by quarter using 'start' `convention`. Values are
+        assigned to the first quarter of the period.
+
+        >>> s = pd.Series(
+        ...     [1, 2], index=pd.period_range("2012-01-01", freq="Y", periods=2)
+        ... )
+        >>> s
+        2012    1
+        2013    2
+        Freq: Y-DEC, dtype: int64
+        >>> s.resample("Q", convention="start").asfreq()
+        2012Q1    1.0
+        2012Q2    NaN
+        2012Q3    NaN
+        2012Q4    NaN
+        2013Q1    2.0
+        2013Q2    NaN
+        2013Q3    NaN
+        2013Q4    NaN
+        Freq: Q-DEC, dtype: float64
+
+        Resample quarters by month using 'end' `convention`. Values are
+        assigned to the last month of the period.
+
+        >>> q = pd.Series(
+        ...     [1, 2, 3, 4], index=pd.period_range("2018-01-01", freq="Q", periods=4)
+        ... )
+        >>> q
+        2018Q1    1
+        2018Q2    2
+        2018Q3    3
+        2018Q4    4
+        Freq: Q-DEC, dtype: int64
+        >>> q.resample("M", convention="end").asfreq()
+        2018-03    1.0
+        2018-04    NaN
+        2018-05    NaN
+        2018-06    2.0
+        2018-07    NaN
+        2018-08    NaN
+        2018-09    3.0
+        2018-10    NaN
+        2018-11    NaN
+        2018-12    4.0
+        Freq: M, dtype: float64
+
         For DataFrame objects, the keyword `on` can be used to specify the
         column instead of the index for resampling.
 
@@ -9135,19 +9181,6 @@ def resample(
         """
         from pandas.core.resample import get_resampler
 
-        if convention is not lib.no_default:
-            # TODO: Enforce in 3.0 (#55968)
-            warnings.warn(
-                f"The 'convention' keyword in {type(self).__name__}.resample is "
-                "deprecated and will be removed in a future version. "
-                "Explicitly cast PeriodIndex to DatetimeIndex before resampling "
-                "instead.",
-                FutureWarning,  # pdlint: ignore[warning_class]
-                stacklevel=find_stack_level(),
-            )
-        else:
-            convention = "start"
-
         return get_resampler(
             cast("Series | DataFrame", self),
             freq=rule,
diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
@@ -1886,7 +1886,7 @@ def _apply_filter(self, indices, dropna):
             mask.fill(False)
             mask[indices.astype(int)] = True
             # mask fails to broadcast when passed to where; broadcast manually.
-            mask = np.tile(mask, list(self._selected_obj.shape[1:]) + [1]).T  # type: ignore[assignment]
+            mask = np.tile(mask, list(self._selected_obj.shape[1:]) + [1]).T
             filtered = self._selected_obj.where(mask)  # Fill with NaNs.
         return filtered
 
diff --git a/pandas/core/indexers/objects.py b/pandas/core/indexers/objects.py
@@ -131,8 +131,8 @@ def get_window_bounds(
         if closed in ["left", "neither"]:
             end -= 1
 
-        end = np.clip(end, 0, num_values)  # type: ignore[assignment]
-        start = np.clip(start, 0, num_values)  # type: ignore[assignment]
+        end = np.clip(end, 0, num_values)
+        start = np.clip(start, 0, num_values)
 
         return start, end
 
@@ -402,7 +402,7 @@ def get_window_bounds(
         start = np.arange(0, num_values, step, dtype="int64")
         end = start + self.window_size
         if self.window_size:
-            end = np.clip(end, 0, num_values)  # type: ignore[assignment]
+            end = np.clip(end, 0, num_values)
 
         return start, end
 
@@ -488,7 +488,7 @@ def get_window_bounds(
             )
             window_indices_start += len(indices)
             # Extend as we'll be slicing window like [start, end)
-            window_indices = np.append(window_indices, [window_indices[-1] + 1]).astype(  # type: ignore[assignment]
+            window_indices = np.append(window_indices, [window_indices[-1] + 1]).astype(
                 np.int64, copy=False
             )
             start_arrays.append(window_indices.take(ensure_platform_int(start)))
diff --git a/pandas/core/methods/describe.py b/pandas/core/methods/describe.py
@@ -353,14 +353,11 @@ def _refine_percentiles(
     if percentiles is None:
         return np.array([0.25, 0.5, 0.75])
 
-    # explicit conversion of `percentiles` to list
-    percentiles = list(percentiles)
+    percentiles = np.asarray(percentiles)
 
     # get them all to be in [0, 1]
     validate_percentile(percentiles)
 
-    percentiles = np.asarray(percentiles)
-
     # sort and check for duplicates
     unique_pcts = np.unique(percentiles)
     assert percentiles is not None
diff --git a/pandas/core/nanops.py b/pandas/core/nanops.py
@@ -651,9 +651,7 @@ def _mask_datetimelike_result(
         # we need to apply the mask
         result = result.astype("i8").view(orig_values.dtype)
         axis_mask = mask.any(axis=axis)
-        # error: Unsupported target for indexed assignment ("Union[ndarray[Any, Any],
-        # datetime64, timedelta64]")
-        result[axis_mask] = iNaT  # type: ignore[index]
+        result[axis_mask] = iNaT
     else:
         if mask.any():
             return np.int64(iNaT).view(orig_values.dtype)
diff --git a/pandas/core/resample.py b/pandas/core/resample.py
@@ -1932,13 +1932,6 @@ class PeriodIndexResampler(DatetimeIndexResampler):
 
     @property
     def _resampler_for_grouping(self):
-        # TODO: Enforce in 3.0 (#55968)
-        warnings.warn(
-            "Resampling a groupby with a PeriodIndex is deprecated. "
-            "Cast to DatetimeIndex before resampling instead.",
-            FutureWarning,  # pdlint: ignore[warning_class]
-            stacklevel=find_stack_level(),
-        )
         return PeriodIndexResamplerGroupby
 
     def _get_binner_for_time(self):
@@ -2265,15 +2258,6 @@ def _get_resampler(self, obj: NDFrame) -> Resampler:
                 gpr_index=ax,
             )
         elif isinstance(ax, PeriodIndex):
-            if isinstance(ax, PeriodIndex):
-                # TODO: Enforce in 3.0 (#53481)
-                # GH#53481
-                warnings.warn(
-                    "Resampling with a PeriodIndex is deprecated. "
-                    "Cast index to DatetimeIndex before resampling instead.",
-                    FutureWarning,  # pdlint: ignore[warning_class]
-                    stacklevel=find_stack_level(),
-                )
             return PeriodIndexResampler(
                 obj,
                 timegrouper=self,
diff --git a/pandas/core/reshape/encoding.py b/pandas/core/reshape/encoding.py
@@ -359,7 +359,7 @@ def get_empty_frame(data) -> DataFrame:
 
         if drop_first:
             # remove first GH12042
-            dummy_mat = dummy_mat[:, 1:]  # type: ignore[assignment]
+            dummy_mat = dummy_mat[:, 1:]
             dummy_cols = dummy_cols[1:]
         return DataFrame(dummy_mat, index=index, columns=dummy_cols, dtype=_dtype)
 
diff --git a/pandas/core/reshape/merge.py b/pandas/core/reshape/merge.py
@@ -1746,7 +1746,9 @@ def _maybe_coerce_merge_keys(self) -> None:
 
                     mask = ~np.isnan(lk)
                     match = lk == casted
-                    if not match[mask].all():
+                    # error: Item "ExtensionArray" of
+                    # "ExtensionArray | Any" has no attribute "all"
+                    if not match[mask].all():  # type: ignore[union-attr]
                         warnings.warn(
                             "You are merging on int and float "
                             "columns where the float values "
@@ -1766,7 +1768,9 @@ def _maybe_coerce_merge_keys(self) -> None:
 
                     mask = ~np.isnan(rk)
                     match = rk == casted
-                    if not match[mask].all():
+                    # error: Item "ExtensionArray" of
+                    # "ExtensionArray | Any" has no attribute "all"
+                    if not match[mask].all():  # type: ignore[union-attr]
                         warnings.warn(
                             "You are merging on int and float "
                             "columns where the float values "
diff --git a/pandas/core/util/hashing.py b/pandas/core/util/hashing.py
@@ -324,8 +324,8 @@ def _hash_ndarray(
             )
 
             codes, categories = factorize(vals, sort=False)
-            dtype = CategoricalDtype(categories=Index(categories), ordered=False)
-            cat = Categorical._simple_new(codes, dtype)
+            tdtype = CategoricalDtype(categories=Index(categories), ordered=False)
+            cat = Categorical._simple_new(codes, tdtype)
             return cat._hash_pandas_object(
                 encoding=encoding, hash_key=hash_key, categorize=False
             )
diff --git a/pandas/io/pytables.py b/pandas/io/pytables.py
diff --git a/pandas/plotting/_matplotlib/timeseries.py b/pandas/plotting/_matplotlib/timeseries.py
diff --git a/pandas/tests/resample/test_base.py b/pandas/tests/resample/test_base.py
diff --git a/pandas/tests/resample/test_datetime_index.py b/pandas/tests/resample/test_datetime_index.py
diff --git a/pandas/tests/resample/test_period_index.py b/pandas/tests/resample/test_period_index.py

Original file line number	Diff line number	Diff line change
`@@ -102,7 +102,7 @@ def quantile_with_mask(`
`102`	`102`	`interpolation=interpolation,`
`103`	`103`	`)`
`104`	`104`
`105`		`- result = np.asarray(result) # type: ignore[assignment]`
	`105`	`+ result = np.asarray(result)`
`106`	`106`	`result = result.T`
`107`	`107`
`108`	`108`	`return result`
Original file line number	Diff line number	Diff line change
`@@ -44,7 +44,7 @@ def pyarrow_array_to_numpy_and_mask(`
`44`	`44`	`mask = pyarrow.BooleanArray.from_buffers(`
`45`	`45`	`pyarrow.bool_(), len(arr), [None, bitmask], offset=arr.offset`
`46`	`46`	`)`
`47`		`- mask = np.asarray(mask) # type: ignore[assignment]`
	`47`	`+ mask = np.asarray(mask)`
`48`	`48`	`else:`
`49`	`49`	`mask = np.ones(len(arr), dtype=bool)`
`50`	`50`	`return data, mask`