REF: Categorical.fillna match patterns in other methods

doc/source/whatsnew/v1.2.0.rst

@@ -238,7 +238,7 @@ Bug fixes
 
 Categorical
 ^^^^^^^^^^^
-
+- :meth:`Categorical.fillna` will always return a copy, will validate a passed fill value regardless of whether there are any NAs to fill, and will disallow a ``NaT`` as a fill value for numeric categories (:issue:`36530`)
 -
 -
 

pandas/core/arrays/categorical.py

@@ -37,7 +37,6 @@
 )
 from pandas.core.dtypes.dtypes import CategoricalDtype
 from pandas.core.dtypes.generic import ABCIndexClass, ABCSeries
-from pandas.core.dtypes.inference import is_hashable
 from pandas.core.dtypes.missing import is_valid_nat_for_dtype, isna, notna
 
 from pandas.core import ops
@@ -1213,7 +1212,7 @@ def _validate_fill_value(self, fill_value):
         ValueError
         """
 
-        if isna(fill_value):
+        if is_valid_nat_for_dtype(fill_value, self.categories.dtype):
             fill_value = -1
         elif fill_value in self.categories:
             fill_value = self._unbox_scalar(fill_value)
@@ -1636,6 +1635,7 @@ def fillna(self, value=None, method=None, limit=None):
         value, method = validate_fillna_kwargs(
             value, method, validate_scalar_dict_value=False
         )
+        value = extract_array(value, extract_numpy=True)
 
         if value is None:
             value = np.nan
@@ -1644,10 +1644,8 @@ def fillna(self, value=None, method=None, limit=None):
                 "specifying a limit for fillna has not been implemented yet"
             )
 
-        codes = self._codes
-
-        # pad / bfill
         if method is not None:
+            # pad / bfill
 
             # TODO: dispatch when self.categories is EA-dtype
             values = np.asarray(self).reshape(-1, len(self))
@@ -1657,40 +1655,25 @@ def fillna(self, value=None, method=None, limit=None):
             codes = _get_codes_for_values(values, self.categories)
 
         else:
+            # We copy even if there is nothing to fill
+            codes = self._ndarray.copy()
+            mask = self.isna()
 
-            # If value is a dict or a Series (a dict value has already
-            # been converted to a Series)
-            if isinstance(value, (np.ndarray, Categorical, ABCSeries)):
+            if isinstance(value, (np.ndarray, Categorical)):
                 # We get ndarray or Categorical if called via Series.fillna,
                 #  where it will unwrap another aligned Series before getting here
 
-                mask = ~algorithms.isin(value, self.categories)
-                if not isna(value[mask]).all():
+                not_categories = ~algorithms.isin(value, self.categories)
+                if not isna(value[not_categories]).all():
+                    # All entries in `value` must either be a category or NA
                     raise ValueError("fill value must be in categories")
 
                 values_codes = _get_codes_for_values(value, self.categories)
-                indexer = np.where(codes == -1)
-                codes = codes.copy()
-                codes[indexer] = values_codes[indexer]
-
-            # If value is not a dict or Series it should be a scalar
-            elif is_hashable(value):
-                if not isna(value) and value not in self.categories:
-                    raise ValueError("fill value must be in categories")
-
-                mask = codes == -1
-                if mask.any():
-                    codes = codes.copy()
-                    if isna(value):
-                        codes[mask] = -1
-                    else:
-                        codes[mask] = self._unbox_scalar(value)
+                codes[mask] = values_codes[mask]
 
             else:
-                raise TypeError(
-                    f"'value' parameter must be a scalar, dict "
-                    f"or Series, but you passed a {type(value).__name__}"
-                )
+                new_code = self._validate_fill_value(value)
+                codes[mask] = new_code
 
         return self._from_backing_data(codes)
 

pandas/tests/frame/test_missing.py

@@ -362,7 +362,8 @@ def test_na_actions_categorical(self):
         res = df.fillna(value={"cats": 3, "vals": "b"})
         tm.assert_frame_equal(res, df_exp_fill)
 
-        with pytest.raises(ValueError, match=("fill value must be in categories")):
+        msg = "'fill_value=4' is not present in this Categorical's categories"
+        with pytest.raises(ValueError, match=msg):
             df.fillna(value={"cats": 4, "vals": "c"})
 
         res = df.fillna(method="pad")

pandas/tests/indexes/categorical/test_fillna.py

@@ -14,6 +14,32 @@ def test_fillna_categorical(self):
         tm.assert_index_equal(idx.fillna(1.0), exp)
 
         # fill by value not in categories raises ValueError
-        msg = "fill value must be in categories"
+        msg = "'fill_value=2.0' is not present in this Categorical's categories"
         with pytest.raises(ValueError, match=msg):
             idx.fillna(2.0)
+
+    def test_fillna_copies_with_no_nas(self):
+        # Nothing to fill, should still get a copy
+        ci = CategoricalIndex([0, 1, 1])
+        cat = ci._data
+        result = ci.fillna(0)
+        assert result._values._ndarray is not cat._ndarray
+        assert result._values._ndarray.base is None
+
+        # Same check directly on the Categorical object
+        result = cat.fillna(0)
+        assert result._ndarray is not cat._ndarray
+        assert result._ndarray.base is None
+
+    def test_fillna_validates_with_no_nas(self):
+        # We validate the fill value even if fillna is a no-op
+        ci = CategoricalIndex([2, 3, 3])
+        cat = ci._data
+
+        msg = "'fill_value=False' is not present in this Categorical's categories"
+        with pytest.raises(ValueError, match=msg):
+            ci.fillna(False)
+
+        # Same check directly on the Categorical
+        with pytest.raises(ValueError, match=msg):
+            cat.fillna(False)

pandas/tests/series/methods/test_fillna.py

@@ -125,7 +125,8 @@ def test_fillna_categorical_raises(self):
         data = ["a", np.nan, "b", np.nan, np.nan]
         ser = Series(Categorical(data, categories=["a", "b"]))
 
-        with pytest.raises(ValueError, match="fill value must be in categories"):
+        msg = "'fill_value=d' is not present in this Categorical's categories"
+        with pytest.raises(ValueError, match=msg):
             ser.fillna("d")
 
         with pytest.raises(ValueError, match="fill value must be in categories"):