Backport PR pandas-dev#45668: BUG: unnecessary FutureWarning in sort_values

Author: jbrockmendel

doc/source/whatsnew/v1.4.1.rst

@@ -28,6 +28,7 @@ Fixed regressions
 Bug fixes
 ~~~~~~~~~
 - Fixed segfault in :meth:``DataFrame.to_json`` when dumping tz-aware datetimes in Python 3.10 (:issue:`42130`)
+- Stopped emitting unnecessary ``FutureWarning`` in :meth:`DataFrame.sort_values` with sparse columns (:issue:`45618`)
 - Fixed window aggregations in :meth:`DataFrame.rolling` and :meth:`Series.rolling` to skip over unused elements (:issue:`45647`)
 -
 

pandas/core/sorting.py

@@ -10,6 +10,7 @@
     Iterable,
     Sequence,
 )
+import warnings
 
 import numpy as np
 
@@ -320,7 +321,12 @@ def lexsort_indexer(
     keys = [ensure_key_mapped(k, key) for k in keys]
 
     for k, order in zip(keys, orders):
-        cat = Categorical(k, ordered=True)
+        with warnings.catch_warnings():
+            # TODO(2.0): unnecessary once deprecation is enforced
+            # GH#45618 don't issue warning user can't do anything about
+            warnings.filterwarnings("ignore", ".*SparseArray.*", category=FutureWarning)
+
+            cat = Categorical(k, ordered=True)
 
         if na_position not in ["last", "first"]:
             raise ValueError(f"invalid na_position: {na_position}")

pandas/tests/extension/test_sparse.py

@@ -278,14 +278,6 @@ def test_fillna_frame(self, data_missing):
 
 
 class TestMethods(BaseSparseTests, base.BaseMethodsTests):
-    @pytest.mark.parametrize("ascending", [True, False])
-    def test_sort_values_frame(self, data_for_sorting, ascending):
-        msg = "will store that array directly"
-        with tm.assert_produces_warning(
-            FutureWarning, match=msg, check_stacklevel=False
-        ):
-            super().test_sort_values_frame(data_for_sorting, ascending)
-
     def test_combine_le(self, data_repeated):
         # We return a Series[SparseArray].__le__ returns a
         # Series[Sparse[bool]]

pandas/tests/frame/methods/test_sort_values.py

@@ -15,6 +15,16 @@
 
 
 class TestDataFrameSortValues:
+    def test_sort_values_sparse_no_warning(self):
+        # GH#45618
+        # TODO(2.0): test will be unnecessary
+        ser = pd.Series(Categorical(["a", "b", "a"], categories=["a", "b", "c"]))
+        df = pd.get_dummies(ser, sparse=True)
+
+        with tm.assert_produces_warning(None):
+            # No warnings about constructing Index from SparseArray
+            df.sort_values(by=df.columns.tolist())
+
     def test_sort_values(self):
         frame = DataFrame(
             [[1, 1, 2], [3, 1, 0], [4, 5, 6]], index=[1, 2, 3], columns=list("ABC")