REF: unify casting logic in Categorical.__init__

pandas/core/arrays/categorical.py

@@ -48,8 +48,6 @@
 from pandas.core.dtypes.cast import (
     coerce_indexer_dtype,
     maybe_cast_to_extension_array,
-    maybe_infer_to_datetimelike,
-    sanitize_to_nanoseconds,
 )
 from pandas.core.dtypes.common import (
     ensure_int64,
@@ -396,24 +394,27 @@ def __init__(
             if dtype.categories is None:
                 dtype = CategoricalDtype(values.categories, dtype.ordered)
         elif not isinstance(values, (ABCIndex, ABCSeries, ExtensionArray)):
-            # sanitize_array coerces np.nan to a string under certain versions
-            # of numpy
-            if not isinstance(values, (np.ndarray, list)):
-                # convert e.g. range, tuple to allow for stronger typing
-                #  of maybe_infer_to_datetimelike
-                values = list(values)
-            values = maybe_infer_to_datetimelike(values)
-            if isinstance(values, np.ndarray):
-                values = sanitize_to_nanoseconds(values)
-            elif not isinstance(values, ExtensionArray):
-                values = com.convert_to_list_like(values)
-
+            values = com.convert_to_list_like(values)
+            if isinstance(values, list) and len(values) == 0:
                 # By convention, empty lists result in object dtype:
-                sanitize_dtype = np.dtype("O") if len(values) == 0 else None
-                null_mask = isna(values)
+                values = np.array([], dtype=object)
+            elif isinstance(values, np.ndarray):
+                if values.ndim > 1:
+                    # preempt sanitize_array from raising ValueError
+                    raise NotImplementedError(
+                        "> 1 ndim Categorical are not supported at this time"
+                    )
+                values = sanitize_array(values, None)
+            else:
+                # i.e. must be a list
+                arr = sanitize_array(values, None)
+                null_mask = isna(arr)
                 if null_mask.any():
-                    values = [values[idx] for idx in np.where(~null_mask)[0]]
-                values = sanitize_array(values, None, dtype=sanitize_dtype)
+                    # We remove null values here, then below will re-insert
+                    #  them, grep "full_codes"
+                    arr = [values[idx] for idx in np.where(~null_mask)[0]]
+                    arr = sanitize_array(arr, None)
+                values = arr
 
         if dtype.categories is None:
             try:

pandas/tests/arrays/categorical/test_constructors.py

@@ -42,6 +42,12 @@ def test_categorical_scalar_deprecated(self):
         with tm.assert_produces_warning(FutureWarning):
             Categorical("A", categories=["A", "B"])
 
+    def test_categorical_1d_only(self):
+        # ndim > 1
+        msg = "> 1 ndim Categorical are not supported at this time"
+        with pytest.raises(NotImplementedError, match=msg):
+            Categorical(np.array([list("abcd")]))
+
     def test_validate_ordered(self):
         # see gh-14058
         exp_msg = "'ordered' must either be 'True' or 'False'"

pandas/tests/frame/test_constructors.py

@@ -2062,13 +2062,6 @@ def test_construct_from_listlikes_mismatched_lengths(self):
         with pytest.raises(ValueError, match=msg):
             DataFrame([Categorical(list("abc")), Categorical(list("abdefg"))])
 
-    def test_categorical_1d_only(self):
-        # TODO: belongs in Categorical tests
-        # ndim > 1
-        msg = "> 1 ndim Categorical are not supported at this time"
-        with pytest.raises(NotImplementedError, match=msg):
-            Categorical(np.array([list("abcd")]))
-
     def test_constructor_categorical_series(self):
 
         items = [1, 2, 3, 1]