Merge branch 'pandas-dev:main' into main

Author: SFuller4

doc/_templates/pandas_footer.html

@@ -0,0 +1,3 @@
+<p class="copyright">
+    &copy {{copyright}} pandas via <a href="https://numfocus.org">NumFOCUS, Inc.</a> Hosted by <a href="https://www.ovhcloud.com">OVH Cloud</a>.
+</p>

doc/source/conf.py

@@ -163,11 +163,8 @@
 
 # General information about the project.
 project = "pandas"
-copyright = (
-    f"{datetime.now().year} "
-    'pandas via <a href="https://numfocus.org">NumFOCUS, Inc.</a> '
-    'Hosted by <a href="https://www.ovhcloud.com">OVH Cloud</a>'
-)
+# We have our custom "pandas_footer.html" template, using copyright for the current year
+copyright = f"{datetime.now().year}"
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
@@ -243,6 +240,7 @@
 
 html_theme_options = {
     "external_links": [],
+    "footer_items": ["pandas_footer", "sphinx-version"],
     "github_url": "https://github.com/pandas-dev/pandas",
     "twitter_url": "https://twitter.com/pandas_dev",
     "google_analytics_id": "UA-27880019-2",

doc/source/user_guide/indexing.rst

@@ -1723,12 +1723,13 @@ the given columns to a MultiIndex:
    frame
 
 Other options in ``set_index`` allow you not drop the index columns or to add
-the index without creating a copy of the underlying data:
+the index in-place (without creating a new object):
 
 .. ipython:: python
 
    data.set_index('c', drop=False)
-   data.set_index(['a', 'b'], copy=False)
+   data.set_index(['a', 'b'], inplace=True)
+   data
 
 Reset the index
 ~~~~~~~~~~~~~~~

doc/source/whatsnew/v1.5.0.rst

@@ -330,7 +330,6 @@ Other enhancements
 - :meth:`DataFrame.quantile` gained a ``method`` argument that can accept ``table`` to evaluate multi-column quantiles (:issue:`43881`)
 - :class:`Interval` now supports checking whether one interval is contained by another interval (:issue:`46613`)
 - Added ``copy`` keyword to :meth:`Series.set_axis` and :meth:`DataFrame.set_axis` to allow user to set axis on a new object without necessarily copying the underlying data (:issue:`47932`)
-- :meth:`DataFrame.set_index` now supports a ``copy`` keyword. If ``False``, the underlying data is not copied when a new :class:`DataFrame` is returned (:issue:`48043`)
 - The method :meth:`.ExtensionArray.factorize` accepts ``use_na_sentinel=False`` for determining how null values are to be treated (:issue:`46601`)
 - The ``Dockerfile`` now installs a dedicated ``pandas-dev`` virtual environment for pandas development instead of using the ``base`` environment (:issue:`48427`)
 
@@ -934,7 +933,6 @@ Other Deprecations
 - Deprecated the ``inplace`` keyword in :meth:`DataFrame.set_axis` and :meth:`Series.set_axis`, use ``obj = obj.set_axis(..., copy=False)`` instead (:issue:`48130`)
 - Deprecated producing a single element when iterating over a :class:`DataFrameGroupBy` or a :class:`SeriesGroupBy` that has been grouped by a list of length 1; A tuple of length one will be returned instead (:issue:`42795`)
 - Fixed up warning message of deprecation of :meth:`MultiIndex.lesort_depth` as public method, as the message previously referred to :meth:`MultiIndex.is_lexsorted` instead (:issue:`38701`)
-- Deprecated the ``inplace`` keyword in :meth:`DataFrame.set_index`, use ``df = df.set_index(..., copy=False)`` instead (:issue:`48115`)
 - Deprecated the ``sort_columns`` argument in :meth:`DataFrame.plot` and :meth:`Series.plot` (:issue:`47563`).
 - Deprecated positional arguments for all but the first argument of :meth:`DataFrame.to_stata` and :func:`read_stata`, use keyword arguments instead (:issue:`48128`).
 - Deprecated the ``mangle_dupe_cols`` argument in :func:`read_csv`, :func:`read_fwf`, :func:`read_table` and :func:`read_excel`. The argument was never implemented, and a new argument where the renaming pattern can be specified will be added instead (:issue:`47718`)

pandas/core/frame.py

@@ -5869,9 +5869,8 @@ def set_index(
         *,
         drop: bool = ...,
         append: bool = ...,
-        inplace: Literal[False] | lib.NoDefault = ...,
+        inplace: Literal[False] = ...,
         verify_integrity: bool = ...,
-        copy: bool | lib.NoDefault = ...,
     ) -> DataFrame:
         ...
 
@@ -5884,7 +5883,6 @@ def set_index(
         append: bool = ...,
         inplace: Literal[True],
         verify_integrity: bool = ...,
-        copy: bool | lib.NoDefault = ...,
     ) -> None:
         ...
 
@@ -5894,9 +5892,8 @@ def set_index(
         keys,
         drop: bool = True,
         append: bool = False,
-        inplace: bool | lib.NoDefault = lib.no_default,
+        inplace: bool = False,
         verify_integrity: bool = False,
-        copy: bool | lib.NoDefault = lib.no_default,
     ) -> DataFrame | None:
         """
         Set the DataFrame index using existing columns.
@@ -5919,18 +5916,10 @@ def set_index(
             Whether to append columns to existing index.
         inplace : bool, default False
             Whether to modify the DataFrame rather than creating a new one.
-
-            .. deprecated:: 1.5.0
-
         verify_integrity : bool, default False
             Check the new index for duplicates. Otherwise defer the check until
             necessary. Setting to False will improve the performance of this
             method.
-        copy : bool, default True
-            Whether to make a copy of the underlying data when returning a new
-            DataFrame.
-
-            .. versionadded:: 1.5.0
 
         Returns
         -------
@@ -5995,25 +5984,7 @@ def set_index(
         3 9       7  2013    84
         4 16     10  2014    31
         """
-        if inplace is not lib.no_default:
-            inplace = validate_bool_kwarg(inplace, "inplace")
-            warnings.warn(
-                "The 'inplace' keyword in DataFrame.set_index is deprecated "
-                "and will be removed in a future version. Use "
-                "`df = df.set_index(..., copy=False)` instead.",
-                FutureWarning,
-                stacklevel=find_stack_level(inspect.currentframe()),
-            )
-        else:
-            inplace = False
-
-        if inplace:
-            if copy is not lib.no_default:
-                raise ValueError("Cannot specify copy when inplace=True")
-            copy = False
-        elif copy is lib.no_default:
-            copy = True
-
+        inplace = validate_bool_kwarg(inplace, "inplace")
         self._check_inplace_and_allows_duplicate_labels(inplace)
         if not isinstance(keys, list):
             keys = [keys]
@@ -6049,7 +6020,7 @@ def set_index(
         if inplace:
             frame = self
         else:
-            frame = self.copy(deep=copy)
+            frame = self.copy()
 
         arrays = []
         names: list[Hashable] = []

pandas/core/internals/blocks.py

@@ -881,8 +881,13 @@ def take_nd(
         )
 
         # Called from three places in managers, all of which satisfy
-        #  this assertion
+        #  these assertions
+        if isinstance(self, ExtensionBlock):
+            # NB: in this case, the 'axis' kwarg will be ignored in the
+            #  algos.take_nd call above.
+            assert not (self.ndim == 1 and new_mgr_locs is None)
         assert not (axis == 0 and new_mgr_locs is None)
+
         if new_mgr_locs is None:
             new_mgr_locs = self._mgr_locs
 
@@ -1753,33 +1758,6 @@ def is_view(self) -> bool:
     def is_numeric(self):
         return self.values.dtype._is_numeric
 
-    def take_nd(
-        self,
-        indexer: npt.NDArray[np.intp],
-        axis: int = 0,
-        new_mgr_locs: BlockPlacement | None = None,
-        fill_value=lib.no_default,
-    ) -> Block:
-        """
-        Take values according to indexer and return them as a block.
-        """
-        if fill_value is lib.no_default:
-            fill_value = None
-
-        # TODO(EA2D): special case not needed with 2D EAs
-        # axis doesn't matter; we are really a single-dim object
-        # but are passed the axis depending on the calling routing
-        # if its REALLY axis 0, then this will be a reindex and not a take
-        new_values = self.values.take(indexer, fill_value=fill_value, allow_fill=True)
-
-        # Called from three places in managers, all of which satisfy
-        #  this assertion
-        assert not (self.ndim == 1 and new_mgr_locs is None)
-        if new_mgr_locs is None:
-            new_mgr_locs = self._mgr_locs
-
-        return self.make_block_same_class(new_values, new_mgr_locs)
-
     def _slice(
         self, slicer: slice | npt.NDArray[np.bool_] | npt.NDArray[np.intp]
     ) -> ExtensionArray:

pandas/core/internals/managers.py

@@ -726,6 +726,9 @@ def reindex_indexer(
             result.axes[axis] = new_axis
             return result
 
+        # Should be intp, but in some cases we get int64 on 32bit builds
+        assert isinstance(indexer, np.ndarray)
+
         # some axes don't allow reindexing with dups
         if not allow_dups:
             self.axes[axis]._validate_can_reindex(indexer)

pandas/core/reshape/merge.py

@@ -783,9 +783,9 @@ def get_result(self, copy: bool = True) -> DataFrame:
         if self.indicator:
             result = self._indicator_post_merge(result)
 
-        result = self._maybe_add_join_keys(result, left_indexer, right_indexer)
+        self._maybe_add_join_keys(result, left_indexer, right_indexer)
 
-        result = self._maybe_restore_index_levels(result)
+        self._maybe_restore_index_levels(result)
 
         self._maybe_drop_cross_column(result, self._cross)
 
@@ -852,7 +852,7 @@ def _indicator_post_merge(self, result: DataFrame) -> DataFrame:
         result = result.drop(labels=["_left_indicator", "_right_indicator"], axis=1)
         return result
 
-    def _maybe_restore_index_levels(self, result: DataFrame) -> DataFrame:
+    def _maybe_restore_index_levels(self, result: DataFrame) -> None:
         """
         Restore index levels specified as `on` parameters
 
@@ -870,7 +870,7 @@ def _maybe_restore_index_levels(self, result: DataFrame) -> DataFrame:
 
         Returns
         -------
-        DataFrame
+        None
         """
         names_to_restore = []
         for name, left_key, right_key in zip(
@@ -894,15 +894,14 @@ def _maybe_restore_index_levels(self, result: DataFrame) -> DataFrame:
                 names_to_restore.append(name)
 
         if names_to_restore:
-            result = result.set_index(names_to_restore, copy=False)
-        return result
+            result.set_index(names_to_restore, inplace=True)
 
     def _maybe_add_join_keys(
         self,
         result: DataFrame,
         left_indexer: np.ndarray | None,
         right_indexer: np.ndarray | None,
-    ) -> DataFrame:
+    ) -> None:
 
         left_has_missing = None
         right_has_missing = None
@@ -993,12 +992,11 @@ def _maybe_add_join_keys(
                             for level_name in result.index.names
                         ]
 
-                        result = result.set_index(idx_list, copy=False)
+                        result.set_index(idx_list, inplace=True)
                     else:
                         result.index = Index(key_col, name=name)
                 else:
                     result.insert(i, name or f"key_{i}", key_col)
-        return result
 
     def _get_join_indexers(self) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]]:
         """return the join indexers"""
@@ -1771,8 +1769,7 @@ def get_result(self, copy: bool = True) -> DataFrame:
         result = self._reindex_and_concat(
             join_index, left_join_indexer, right_join_indexer, copy=copy
         )
-
-        result = self._maybe_add_join_keys(result, left_indexer, right_indexer)
+        self._maybe_add_join_keys(result, left_indexer, right_indexer)
 
         return result
 

pandas/io/parsers/arrow_parser_wrapper.py

@@ -117,7 +117,7 @@ def _finalize_output(self, frame: DataFrame) -> DataFrame:
                     # String case
                     if item not in frame.columns:
                         raise ValueError(f"Index {item} invalid")
-            frame = frame.set_index(self.index_col, drop=True, copy=False)
+            frame.set_index(self.index_col, drop=True, inplace=True)
             # Clear names if headerless and no name given
             if self.header is None and not multi_index_named:
                 frame.index.names = [None] * len(frame.index.names)

pandas/io/pytables.py

@@ -4673,7 +4673,7 @@ def read(
                     columns.insert(0, n)
         s = super().read(where=where, columns=columns, start=start, stop=stop)
         if is_multi_index:
-            s = s.set_index(self.levels, copy=False)
+            s.set_index(self.levels, inplace=True)
 
         s = s.iloc[:, 0]
 

pandas/io/sql.py

@@ -152,7 +152,7 @@ def _wrap_result(
     frame = _parse_date_columns(frame, parse_dates)
 
     if index_col is not None:
-        frame = frame.set_index(index_col, copy=False)
+        frame.set_index(index_col, inplace=True)
 
     return frame
 
@@ -980,7 +980,7 @@ def _query_iterator(
                 self._harmonize_columns(parse_dates=parse_dates)
 
                 if self.index is not None:
-                    self.frame = self.frame.set_index(self.index, copy=False)
+                    self.frame.set_index(self.index, inplace=True)
 
                 yield self.frame
 
@@ -1021,7 +1021,7 @@ def read(
             self._harmonize_columns(parse_dates=parse_dates)
 
             if self.index is not None:
-                self.frame = self.frame.set_index(self.index, copy=False)
+                self.frame.set_index(self.index, inplace=True)
 
             return self.frame
 

pandas/tests/extension/base/dim2.py

@@ -18,6 +18,9 @@
 
 
 class Dim2CompatTests(BaseExtensionTests):
+    # Note: these are ONLY for ExtensionArray subclasses that support 2D arrays.
+    #  i.e. not for pyarrow-backed EAs.
+
     def test_transpose(self, data):
         arr2d = data.repeat(2).reshape(-1, 2)
         shape = arr2d.shape

pandas/tests/extension/test_arrow.py

@@ -319,20 +319,6 @@ def test_from_sequence_of_strings_pa_array(self, data, request):
         tm.assert_extension_array_equal(result, data)
 
 
-@pytest.mark.xfail(
-    raises=NotImplementedError, reason="pyarrow.ChunkedArray backing is 1D."
-)
-class TestDim2Compat(base.Dim2CompatTests):
-    pass
-
-
-@pytest.mark.xfail(
-    raises=NotImplementedError, reason="pyarrow.ChunkedArray backing is 1D."
-)
-class TestNDArrayBacked2D(base.NDArrayBacked2DTests):
-    pass
-
-
 class TestGetitemTests(base.BaseGetitemTests):
     @pytest.mark.xfail(
         reason=(

pandas/tests/frame/methods/test_combine_first.py

@@ -394,12 +394,12 @@ def test_combine_first_string_dtype_only_na(self, nullable_string_dtype):
             PerformanceWarning,
             pa_version_under7p0 and nullable_string_dtype == "string[pyarrow]",
         ):
-            df = df.set_index(["a", "b"], copy=False)
+            df.set_index(["a", "b"], inplace=True)
         with tm.maybe_produces_warning(
             PerformanceWarning,
             pa_version_under7p0 and nullable_string_dtype == "string[pyarrow]",
         ):
-            df2 = df2.set_index(["a", "b"], copy=False)
+            df2.set_index(["a", "b"], inplace=True)
         result = df.combine_first(df2)
         with tm.maybe_produces_warning(
             PerformanceWarning,