Backport PR pandas-dev#57250: REGR/DOC: pd.concat should special case…

… DatetimeIndex to sort even when sort=False
meeseeksmachine · Feb 9, 2024 · 385f6a7 · 385f6a7
1 parent 11a6136
commit 385f6a7
Show file tree

Hide file tree

Showing 4 changed files with 12 additions and 8 deletions.
diff --git a/doc/source/whatsnew/v2.2.1.rst b/doc/source/whatsnew/v2.2.1.rst
@@ -15,6 +15,7 @@ Fixed regressions
 ~~~~~~~~~~~~~~~~~
 - Fixed memory leak in :func:`read_csv` (:issue:`57039`)
 - Fixed performance regression in :meth:`Series.combine_first` (:issue:`55845`)
+- Fixed regression in :func:`concat` changing long-standing behavior that always sorted the non-concatenation axis when the axis was a :class:`DatetimeIndex` (:issue:`57006`)
 - 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`)

diff --git a/pandas/core/indexes/api.py b/pandas/core/indexes/api.py
@@ -295,6 +295,7 @@ def _find_common_index_dtype(inds):
             raise TypeError("Cannot join tz-naive with tz-aware DatetimeIndex")
 
         if len(dtis) == len(indexes):
+            sort = True
             result = indexes[0]
 
         elif len(dtis) > 1:

diff --git a/pandas/core/reshape/concat.py b/pandas/core/reshape/concat.py
@@ -205,8 +205,10 @@ def concat(
         Check whether the new concatenated axis contains duplicates. This can
         be very expensive relative to the actual data concatenation.
     sort : bool, default False
-        Sort non-concatenation axis if it is not already aligned.
-
+        Sort non-concatenation axis if it is not already aligned. One exception to
+        this is when the non-concatentation axis is a DatetimeIndex and join='outer'
+        and the axis is not already aligned. In that case, the non-concatenation
+        axis is always sorted lexicographically.
     copy : bool, default True
         If False, do not copy data unnecessarily.
 

diff --git a/pandas/tests/reshape/concat/test_datetimes.py b/pandas/tests/reshape/concat/test_datetimes.py
@@ -73,23 +73,23 @@ def test_concat_datetime_timezone(self):
 
         exp_idx = DatetimeIndex(
             [
-                "2010-12-31 23:00:00+00:00",
-                "2011-01-01 00:00:00+00:00",
-                "2011-01-01 01:00:00+00:00",
                 "2010-12-31 15:00:00+00:00",
                 "2010-12-31 16:00:00+00:00",
                 "2010-12-31 17:00:00+00:00",
+                "2010-12-31 23:00:00+00:00",
+                "2011-01-01 00:00:00+00:00",
+                "2011-01-01 01:00:00+00:00",
             ]
         ).as_unit("ns")
 
         expected = DataFrame(
             [
-                [1, np.nan],
-                [2, np.nan],
-                [3, np.nan],
                 [np.nan, 1],
                 [np.nan, 2],
                 [np.nan, 3],
+                [1, np.nan],
+                [2, np.nan],
+                [3, np.nan],
             ],
             index=exp_idx,
             columns=["a", "b"],