CLN: astype_nansafe require dtype object (#38466)

Author: jbrockmendel

pandas/core/arrays/sparse/array.py

@@ -1061,7 +1061,7 @@ def astype(self, dtype=None, copy=True):
             else:
                 return self.copy()
         dtype = self.dtype.update_dtype(dtype)
-        subtype = dtype._subtype_with_str
+        subtype = pandas_dtype(dtype._subtype_with_str)
         # TODO copy=False is broken for astype_nansafe with int -> float, so cannot
         # passthrough copy keyword: https://github.com/pandas-dev/pandas/issues/34456
         sp_values = astype_nansafe(self.sp_values, subtype, copy=True)

pandas/core/dtypes/cast.py

@@ -70,7 +70,6 @@
     is_timedelta64_dtype,
     is_timedelta64_ns_dtype,
     is_unsigned_integer_dtype,
-    pandas_dtype,
 )
 from pandas.core.dtypes.dtypes import (
     DatetimeTZDtype,
@@ -962,7 +961,7 @@ def astype_nansafe(
     Parameters
     ----------
     arr : ndarray
-    dtype : np.dtype
+    dtype : np.dtype or ExtensionDtype
     copy : bool, default True
         If False, a view will be attempted but may fail, if
         e.g. the item sizes don't align.
@@ -975,11 +974,11 @@ def astype_nansafe(
         The dtype was a datetime64/timedelta64 dtype, but it had no unit.
     """
     # dispatch on extension dtype if needed
-    if is_extension_array_dtype(dtype):
+    if isinstance(dtype, ExtensionDtype):
         return dtype.construct_array_type()._from_sequence(arr, dtype=dtype, copy=copy)
 
-    if not isinstance(dtype, np.dtype):
-        dtype = pandas_dtype(dtype)
+    elif not isinstance(dtype, np.dtype):
+        raise ValueError("dtype must be np.dtype or ExtensionDtype")
 
     if issubclass(dtype.type, str):
         return lib.ensure_string_array(

pandas/io/parsers.py

@@ -1714,6 +1714,7 @@ def _convert_to_ndarrays(
                         except (AttributeError, TypeError):
                             # invalid input to is_bool_dtype
                             pass
+                    cast_type = pandas_dtype(cast_type)
                     cvals = self._cast_types(cvals, cast_type, c)
 
             result[c] = cvals

pandas/tests/dtypes/test_common.py

@@ -716,6 +716,8 @@ def test__is_dtype_type(input_param, result):
 def test_astype_nansafe(val, typ):
     arr = np.array([val])
 
+    typ = np.dtype(typ)
+
     msg = "Cannot convert NaT values to integer"
     with pytest.raises(ValueError, match=msg):
         astype_nansafe(arr, dtype=typ)
@@ -738,14 +740,16 @@ def test_astype_nansafe(val, typ):
 def test_astype_datetime64_bad_dtype_raises(from_type, to_type):
     arr = np.array([from_type("2018")])
 
+    to_type = np.dtype(to_type)
+
     with pytest.raises(TypeError, match="cannot astype"):
         astype_nansafe(arr, dtype=to_type)
 
 
 @pytest.mark.parametrize("from_type", [np.datetime64, np.timedelta64])
 def test_astype_object_preserves_datetime_na(from_type):
     arr = np.array([from_type("NaT")])
-    result = astype_nansafe(arr, dtype="object")
+    result = astype_nansafe(arr, dtype=np.dtype("object"))
 
     assert isna(result)[0]