DEPR: DataFrame([categorical, ...]) special casing

doc/source/whatsnew/v1.3.0.rst

@@ -648,6 +648,7 @@ Deprecations
 - Deprecated setting :attr:`Categorical._codes`, create a new :class:`Categorical` with the desired codes instead (:issue:`40606`)
 - Deprecated behavior of :meth:`DatetimeIndex.union` with mixed timezones; in a future version both will be cast to UTC instead of object dtype (:issue:`39328`)
 - Deprecated using ``usecols`` with out of bounds indices for ``read_csv`` with ``engine="c"`` (:issue:`25623`)
+- Deprecated special treatment of lists with first element a Categorical in the :class:`DataFrame` constructor; pass as ``pd.DataFrame({col: categorical, ...})`` instead (:issue:`38845`)
 - Deprecated passing arguments as positional (except for ``"method"``) in :meth:`DataFrame.interpolate` and :meth:`Series.interpolate` (:issue:`41485`)
 
 .. ---------------------------------------------------------------------------

pandas/core/internals/construction.py

@@ -11,6 +11,7 @@
     Hashable,
     Sequence,
 )
+import warnings
 
 import numpy as np
 import numpy.ma as ma
@@ -772,6 +773,16 @@ def to_arrays(
         return [], ensure_index([])
 
     elif isinstance(data[0], Categorical):
+        # GH#38845 deprecate special case
+        warnings.warn(
+            "The behavior of DataFrame([categorical, ...]) is deprecated and "
+            "in a future version will be changed to match the behavior of "
+            "DataFrame([any_listlike, ...]). "
+            "To retain the old behavior, pass as a dictionary "
+            "DataFrame({col: categorical, ..})",
+            FutureWarning,
+            stacklevel=4,
+        )
         if columns is None:
             columns = ibase.default_index(len(data))
         return data, columns

pandas/io/pytables.py

@@ -4571,7 +4571,7 @@ def read(
                 df = DataFrame(values, columns=cols_, index=index_)
             else:
                 # Categorical
-                df = DataFrame([values], columns=cols_, index=index_)
+                df = DataFrame._from_arrays([values], columns=cols_, index=index_)
             assert (df.dtypes == values.dtype).all(), (df.dtypes, values.dtype)
             frames.append(df)
 

pandas/tests/frame/test_constructors.py

@@ -2089,12 +2089,16 @@ def test_constructor_categorical(self):
 
     def test_construct_from_1item_list_of_categorical(self):
         # ndim != 1
-        df = DataFrame([Categorical(list("abc"))])
+        msg = "will be changed to match the behavior"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            df = DataFrame([Categorical(list("abc"))])
         expected = DataFrame({0: Series(list("abc"), dtype="category")})
         tm.assert_frame_equal(df, expected)
 
     def test_construct_from_list_of_categoricals(self):
-        df = DataFrame([Categorical(list("abc")), Categorical(list("abd"))])
+        msg = "will be changed to match the behavior"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            df = DataFrame([Categorical(list("abc")), Categorical(list("abd"))])
         expected = DataFrame(
             {
                 0: Series(list("abc"), dtype="category"),
@@ -2106,7 +2110,9 @@ def test_construct_from_list_of_categoricals(self):
 
     def test_from_nested_listlike_mixed_types(self):
         # mixed
-        df = DataFrame([Categorical(list("abc")), list("def")])
+        msg = "will be changed to match the behavior"
+        with tm.assert_produces_warning(FutureWarning, match=msg):
+            df = DataFrame([Categorical(list("abc")), list("def")])
         expected = DataFrame(
             {0: Series(list("abc"), dtype="category"), 1: list("def")}, columns=[0, 1]
         )
@@ -2120,8 +2126,10 @@ def test_construct_from_listlikes_mismatched_lengths(self):
                 "Passed arrays should have the same length as the rows Index",
             ]
         )
+        msg2 = "will be changed to match the behavior"
         with pytest.raises(ValueError, match=msg):
-            DataFrame([Categorical(list("abc")), Categorical(list("abdefg"))])
+            with tm.assert_produces_warning(FutureWarning, match=msg2):
+                DataFrame([Categorical(list("abc")), Categorical(list("abdefg"))])
 
     def test_constructor_categorical_series(self):