BUG: assign consensus name to index union in array case GH13475 (#35338)

Author: iamlemec

doc/source/whatsnew/v1.2.0.rst

@@ -162,6 +162,7 @@ Reshaping
 ^^^^^^^^^
 
 - Bug in :meth:`DataFrame.pivot_table` with ``aggfunc='count'`` or ``aggfunc='sum'`` returning ``NaN`` for missing categories when pivoted on a ``Categorical``. Now returning ``0`` (:issue:`31422`)
+- Bug in :func:`union_indexes` where input index names are not preserved in some cases. Affects :func:`concat` and :class:`DataFrame` constructor (:issue:`13475`)
 -
 
 Sparse

pandas/core/indexes/api.py

@@ -218,9 +218,8 @@ def conv(i):
             return result
     elif kind == "array":
         index = indexes[0]
-        for other in indexes[1:]:
-            if not index.equals(other):
-                return _unique_indices(indexes)
+        if not all(index.equals(other) for other in indexes[1:]):
+            index = _unique_indices(indexes)
 
         name = get_consensus_names(indexes)[0]
         if name != index.name:

pandas/tests/frame/test_constructors.py

@@ -1618,6 +1618,42 @@ def test_constructor_Series_differently_indexed(self):
         tm.assert_index_equal(df2.index, other_index)
         tm.assert_frame_equal(df2, exp2)
 
+    @pytest.mark.parametrize(
+        "name_in1,name_in2,name_in3,name_out",
+        [
+            ("idx", "idx", "idx", "idx"),
+            ("idx", "idx", None, "idx"),
+            ("idx", None, None, "idx"),
+            ("idx1", "idx2", None, None),
+            ("idx1", "idx1", "idx2", None),
+            ("idx1", "idx2", "idx3", None),
+            (None, None, None, None),
+        ],
+    )
+    def test_constructor_index_names(self, name_in1, name_in2, name_in3, name_out):
+        # GH13475
+        indices = [
+            pd.Index(["a", "b", "c"], name=name_in1),
+            pd.Index(["b", "c", "d"], name=name_in2),
+            pd.Index(["c", "d", "e"], name=name_in3),
+        ]
+        series = {
+            c: pd.Series([0, 1, 2], index=i) for i, c in zip(indices, ["x", "y", "z"])
+        }
+        result = pd.DataFrame(series)
+
+        exp_ind = pd.Index(["a", "b", "c", "d", "e"], name=name_out)
+        expected = pd.DataFrame(
+            {
+                "x": [0, 1, 2, np.nan, np.nan],
+                "y": [np.nan, 0, 1, 2, np.nan],
+                "z": [np.nan, np.nan, 0, 1, 2],
+            },
+            index=exp_ind,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
     def test_constructor_manager_resize(self, float_frame):
         index = list(float_frame.index[:5])
         columns = list(float_frame.columns[:3])

pandas/tests/reshape/test_concat.py

@@ -1279,6 +1279,43 @@ def test_concat_ignore_index(self, sort):
 
         tm.assert_frame_equal(v1, expected)
 
+    @pytest.mark.parametrize(
+        "name_in1,name_in2,name_in3,name_out",
+        [
+            ("idx", "idx", "idx", "idx"),
+            ("idx", "idx", None, "idx"),
+            ("idx", None, None, "idx"),
+            ("idx1", "idx2", None, None),
+            ("idx1", "idx1", "idx2", None),
+            ("idx1", "idx2", "idx3", None),
+            (None, None, None, None),
+        ],
+    )
+    def test_concat_same_index_names(self, name_in1, name_in2, name_in3, name_out):
+        # GH13475
+        indices = [
+            pd.Index(["a", "b", "c"], name=name_in1),
+            pd.Index(["b", "c", "d"], name=name_in2),
+            pd.Index(["c", "d", "e"], name=name_in3),
+        ]
+        frames = [
+            pd.DataFrame({c: [0, 1, 2]}, index=i)
+            for i, c in zip(indices, ["x", "y", "z"])
+        ]
+        result = pd.concat(frames, axis=1)
+
+        exp_ind = pd.Index(["a", "b", "c", "d", "e"], name=name_out)
+        expected = pd.DataFrame(
+            {
+                "x": [0, 1, 2, np.nan, np.nan],
+                "y": [np.nan, 0, 1, 2, np.nan],
+                "z": [np.nan, np.nan, 0, 1, 2],
+            },
+            index=exp_ind,
+        )
+
+        tm.assert_frame_equal(result, expected)
+
     def test_concat_multiindex_with_keys(self):
         index = MultiIndex(
             levels=[["foo", "bar", "baz", "qux"], ["one", "two", "three"]],