Test for is_dtype_equal matching dtype.__eq__

Author: jbrockmendel

pandas/core/arrays/categorical.py

@@ -36,6 +36,7 @@
     is_scalar,
     is_timedelta64_dtype,
     needs_i8_conversion,
+    pandas_dtype,
 )
 from pandas.core.dtypes.dtypes import CategoricalDtype
 from pandas.core.dtypes.generic import ABCIndex, ABCSeries
@@ -403,6 +404,7 @@ def astype(self, dtype: Dtype, copy: bool = True) -> ArrayLike:
             If copy is set to False and dtype is categorical, the original
             object is returned.
         """
+        dtype = pandas_dtype(dtype)
         if self.dtype is dtype:
             result = self.copy() if copy else self
 

pandas/core/dtypes/common.py

@@ -639,6 +639,19 @@ def is_dtype_equal(source, target) -> bool:
     >>> is_dtype_equal(DatetimeTZDtype(tz="UTC"), "datetime64")
     False
     """
+    if isinstance(target, str):
+        if not isinstance(source, str):
+            # GH#38516 ensure we get the same behavior from
+            #  is_dtype_equal(CDT, "category") and CDT == "category"
+            try:
+                src = get_dtype(source)
+                if isinstance(src, ExtensionDtype):
+                    return src == target
+            except (TypeError, AttributeError):
+                return False
+    elif isinstance(source, str):
+        return is_dtype_equal(target, source)
+
     try:
         source = get_dtype(source)
         target = get_dtype(target)

pandas/core/dtypes/dtypes.py

@@ -355,7 +355,7 @@ def __eq__(self, other: Any) -> bool:
             return False
         elif self.categories is None or other.categories is None:
             # For non-fully-initialized dtypes, these are only equal to
-            #  - the string "categorical" (handled above)
+            #  - the string "category" (handled above)
             #  - other CategoricalDtype with categories=None
             return self.categories is other.categories
         elif self.ordered or other.ordered:

pandas/core/indexes/category.py

@@ -201,8 +201,13 @@ def __new__(
 
         if not isinstance(data, Categorical):
             data = Categorical(data, dtype=dtype)
-        elif isinstance(dtype, CategoricalDtype) and dtype != data.dtype:
+        elif (
+            isinstance(dtype, CategoricalDtype)
+            and dtype != data.dtype
+            and dtype.categories is not None
+        ):
             # we want to silently ignore dtype='category'
+            # TODO: what if dtype.ordered is not None but dtype.categories is?
             data = data._set_dtype(dtype)
 
         data = data.copy() if copy else data

pandas/tests/arrays/categorical/test_dtypes.py

@@ -127,7 +127,7 @@ def test_astype(self, ordered):
         expected = np.array(cat)
         tm.assert_numpy_array_equal(result, expected)
 
-        msg = r"Cannot cast object dtype to <class 'float'>"
+        msg = r"Cannot cast object dtype to float64"
         with pytest.raises(ValueError, match=msg):
             cat.astype(float)
 

pandas/tests/dtypes/test_dtypes.py

@@ -90,9 +90,20 @@ def test_hash_vs_equality(self, dtype):
         assert hash(dtype) == hash(dtype2)
 
     def test_equality(self, dtype):
+        assert dtype == "category"
         assert is_dtype_equal(dtype, "category")
+        assert "category" == dtype
+        assert is_dtype_equal("category", dtype)
+
+        assert dtype == CategoricalDtype()
         assert is_dtype_equal(dtype, CategoricalDtype())
+        assert CategoricalDtype() == dtype
+        assert is_dtype_equal(CategoricalDtype(), dtype)
+
+        assert dtype != "foo"
         assert not is_dtype_equal(dtype, "foo")
+        assert "foo" != dtype
+        assert not is_dtype_equal("foo", dtype)
 
     def test_construction_from_string(self, dtype):
         result = CategoricalDtype.construct_from_string("category")