REGR: DataFrame.sort_index not producing stable sort (#57169)

DataFrame.sort_index not producing stable sort

Author: lukemanley

doc/source/whatsnew/v2.2.1.rst

@@ -18,6 +18,7 @@ Fixed regressions
 - Fixed regression in :func:`merge_ordered` raising ``TypeError`` for ``fill_method="ffill"`` and ``how="left"`` (:issue:`57010`)
 - Fixed regression in :func:`wide_to_long` raising an ``AttributeError`` for string columns (:issue:`57066`)
 - Fixed regression in :meth:`DataFrame.loc` raising ``IndexError`` for non-unique, masked dtype indexes where result has more than 10,000 rows (:issue:`57027`)
+- Fixed regression in :meth:`DataFrame.sort_index` not producing a stable sort for a index with duplicates (:issue:`57151`)
 - Fixed regression in :meth:`DataFrame.to_dict` with ``orient='list'`` and datetime or timedelta types returning integers (:issue:`54824`)
 - Fixed regression in :meth:`DataFrameGroupBy.idxmin`, :meth:`DataFrameGroupBy.idxmax`, :meth:`SeriesGroupBy.idxmin`, :meth:`SeriesGroupBy.idxmax` ignoring the ``skipna`` argument (:issue:`57040`)
 - Fixed regression in :meth:`DataFrameGroupBy.idxmin`, :meth:`DataFrameGroupBy.idxmax`, :meth:`SeriesGroupBy.idxmin`, :meth:`SeriesGroupBy.idxmax` where values containing the minimum or maximum value for the dtype could produce incorrect results (:issue:`57040`)

pandas/core/indexes/base.py

@@ -5952,17 +5952,14 @@ def sort_values(
         (Index([1000, 100, 10, 1], dtype='int64'), array([3, 1, 0, 2]))
         """
         if key is None and (
-            self.is_monotonic_increasing or self.is_monotonic_decreasing
+            (ascending and self.is_monotonic_increasing)
+            or (not ascending and self.is_monotonic_decreasing)
         ):
-            reverse = ascending != self.is_monotonic_increasing
-            sorted_index = self[::-1] if reverse else self.copy()
             if return_indexer:
                 indexer = np.arange(len(self), dtype=np.intp)
-                if reverse:
-                    indexer = indexer[::-1]
-                return sorted_index, indexer
+                return self.copy(), indexer
             else:
-                return sorted_index
+                return self.copy()
 
         # GH 35584. Sort missing values according to na_position kwarg
         # ignore na_position for MultiIndex

pandas/tests/frame/methods/test_sort_index.py

@@ -1001,3 +1001,27 @@ def test_axis_columns_ignore_index():
     result = df.sort_index(axis="columns", ignore_index=True)
     expected = DataFrame([[2, 1]])
     tm.assert_frame_equal(result, expected)
+
+
+def test_sort_index_stable_sort():
+    # GH 57151
+    df = DataFrame(
+        data=[
+            (Timestamp("2024-01-30 13:00:00"), 13.0),
+            (Timestamp("2024-01-30 13:00:00"), 13.1),
+            (Timestamp("2024-01-30 12:00:00"), 12.0),
+            (Timestamp("2024-01-30 12:00:00"), 12.1),
+        ],
+        columns=["dt", "value"],
+    ).set_index(["dt"])
+    result = df.sort_index(level="dt", kind="stable")
+    expected = DataFrame(
+        data=[
+            (Timestamp("2024-01-30 12:00:00"), 12.0),
+            (Timestamp("2024-01-30 12:00:00"), 12.1),
+            (Timestamp("2024-01-30 13:00:00"), 13.0),
+            (Timestamp("2024-01-30 13:00:00"), 13.1),
+        ],
+        columns=["dt", "value"],
+    ).set_index(["dt"])
+    tm.assert_frame_equal(result, expected)