BUG: ignoring sort in DTA.factorize (pandas-dev#53992)

pandas/core/arrays/datetimelike.py

@@ -2211,7 +2211,15 @@ def factorize(
                 codes = codes[::-1]
                 uniques = uniques[::-1]
             return codes, uniques
-        # FIXME: shouldn't get here; we are ignoring sort
+
+        if sort:
+            # algorithms.factorize only passes sort=True here when freq is
+            #  not None, so this should not be reached.
+            raise NotImplementedError(
+                f"The 'sort' keyword in {type(self).__name__}.factorize is "
+                "ignored unless arr.freq is not None. To factorize with sort, "
+                "call pd.factorize(obj, sort=True) instead."
+            )
         return super().factorize(use_na_sentinel=use_na_sentinel)
 
     @classmethod

pandas/tests/arrays/test_datetimes.py

@@ -745,3 +745,16 @@ def test_iter_zoneinfo_fold(self, tz):
         right2 = dta.astype(object)[2]
         assert str(left) == str(right2)
         assert left.utcoffset() == right2.utcoffset()
+
+
+def test_factorize_sort_without_freq():
+    dta = DatetimeArray._from_sequence([0, 2, 1])
+
+    msg = r"call pd.factorize\(obj, sort=True\) instead"
+    with pytest.raises(NotImplementedError, match=msg):
+        dta.factorize(sort=True)
+
+    # Do TimedeltaArray while we're here
+    tda = dta - dta[0]
+    with pytest.raises(NotImplementedError, match=msg):
+        tda.factorize(sort=True)