Backport PR #29393: BUG: GH25495 incorrect dtype when using .loc to set Categorical value for column in 1-row DataFrame (#31393)

Co-authored-by: Kee Chong Tan <[email protected]>

Author: 2 people

doc/source/whatsnew/v1.0.0.rst

@@ -1091,6 +1091,7 @@ Indexing
 - Bug when indexing with ``.loc`` where the index was a :class:`CategoricalIndex` with non-string categories didn't work (:issue:`17569`, :issue:`30225`)
 - :meth:`Index.get_indexer_non_unique` could fail with ``TypeError`` in some cases, such as when searching for ints in a string index (:issue:`28257`)
 - Bug in :meth:`Float64Index.get_loc` incorrectly raising ``TypeError`` instead of ``KeyError`` (:issue:`29189`)
+- Bug in :meth:`DataFrame.loc` with incorrect dtype when setting Categorical value in 1-row DataFrame (:issue:`25495`)
 - :meth:`MultiIndex.get_loc` can't find missing values when input includes missing values (:issue:`19132`)
 - Bug in :meth:`Series.__setitem__` incorrectly assigning values with boolean indexer when the length of new data matches the number of ``True`` values and new data is not a ``Series`` or an ``np.array`` (:issue:`30567`)
 - Bug in indexing with a :class:`PeriodIndex` incorrectly accepting integers representing years, use e.g. ``ser.loc["2007"]`` instead of ``ser.loc[2007]`` (:issue:`30763`)

pandas/core/internals/blocks.py

@@ -872,7 +872,11 @@ def setitem(self, indexer, value):
 
         # length checking
         check_setitem_lengths(indexer, value, values)
-
+        exact_match = (
+            len(arr_value.shape)
+            and arr_value.shape[0] == values.shape[0]
+            and arr_value.size == values.size
+        )
         if is_empty_indexer(indexer, arr_value):
             # GH#8669 empty indexers
             pass
@@ -882,14 +886,21 @@ def setitem(self, indexer, value):
             #  be e.g. a list; see GH#6043
             values[indexer] = value
 
-        # if we are an exact match (ex-broadcasting),
-        # then use the resultant dtype
         elif (
-            len(arr_value.shape)
-            and arr_value.shape[0] == values.shape[0]
-            and arr_value.size == values.size
+            exact_match
+            and is_categorical_dtype(arr_value.dtype)
+            and not is_categorical_dtype(values)
         ):
+            # GH25495 - If the current dtype is not categorical,
+            # we need to create a new categorical block
             values[indexer] = value
+            return self.make_block(Categorical(self.values, dtype=arr_value.dtype))
+
+        # if we are an exact match (ex-broadcasting),
+        # then use the resultant dtype
+        elif exact_match:
+            values[indexer] = value
+
             try:
                 values = values.astype(arr_value.dtype)
             except ValueError:

pandas/tests/frame/indexing/test_categorical.py

@@ -354,6 +354,16 @@ def test_functions_no_warnings(self):
                 df.value, range(0, 105, 10), right=False, labels=labels
             )
 
+    def test_setitem_single_row_categorical(self):
+        # GH 25495
+        df = DataFrame({"Alpha": ["a"], "Numeric": [0]})
+        categories = pd.Categorical(df["Alpha"], categories=["a", "b", "c"])
+        df.loc[:, "Alpha"] = categories
+
+        result = df["Alpha"]
+        expected = Series(categories, index=df.index, name="Alpha")
+        tm.assert_series_equal(result, expected)
+
     def test_loc_indexing_preserves_index_category_dtype(self):
         # GH 15166
         df = DataFrame(