COMPAT: properly handle objects of datetimelike in astype_nansafe (#19232)

reclose #19116

Author: jreback

pandas/core/dtypes/cast.py

@@ -649,10 +649,12 @@ def astype_nansafe(arr, dtype, copy=True):
     if issubclass(dtype.type, text_type):
         # in Py3 that's str, in Py2 that's unicode
         return lib.astype_unicode(arr.ravel()).reshape(arr.shape)
+
     elif issubclass(dtype.type, string_types):
         return lib.astype_str(arr.ravel()).reshape(arr.shape)
+
     elif is_datetime64_dtype(arr):
-        if dtype == object:
+        if is_object_dtype(dtype):
             return tslib.ints_to_pydatetime(arr.view(np.int64))
         elif dtype == np.int64:
             return arr.view(dtype)
@@ -666,10 +668,10 @@ def astype_nansafe(arr, dtype, copy=True):
                                                  to_dtype=dtype))
 
     elif is_timedelta64_dtype(arr):
-        if dtype == np.int64:
-            return arr.view(dtype)
-        elif dtype == object:
+        if is_object_dtype(dtype):
             return tslib.ints_to_pytimedelta(arr.view(np.int64))
+        elif dtype == np.int64:
+            return arr.view(dtype)
 
         # in py3, timedelta64[ns] are int64
         if ((PY3 and dtype not in [_INT64_DTYPE, _TD_DTYPE]) or
@@ -696,9 +698,21 @@ def astype_nansafe(arr, dtype, copy=True):
             raise ValueError('Cannot convert non-finite values (NA or inf) to '
                              'integer')
 
-    elif is_object_dtype(arr.dtype) and np.issubdtype(dtype.type, np.integer):
+    elif is_object_dtype(arr):
+
         # work around NumPy brokenness, #1987
-        return lib.astype_intsafe(arr.ravel(), dtype).reshape(arr.shape)
+        if np.issubdtype(dtype.type, np.integer):
+            return lib.astype_intsafe(arr.ravel(), dtype).reshape(arr.shape)
+
+        # if we have a datetime/timedelta array of objects
+        # then coerce to a proper dtype and recall astype_nansafe
+
+        elif is_datetime64_dtype(dtype):
+            from pandas import to_datetime
+            return astype_nansafe(to_datetime(arr).values, dtype, copy=copy)
+        elif is_timedelta64_dtype(dtype):
+            from pandas import to_timedelta
+            return astype_nansafe(to_timedelta(arr).values, dtype, copy=copy)
 
     if dtype.name in ("datetime64", "timedelta64"):
         msg = ("Passing in '{dtype}' dtype with no frequency is "
@@ -709,7 +723,6 @@ def astype_nansafe(arr, dtype, copy=True):
         dtype = np.dtype(dtype.name + "[ns]")
 
     if copy:
-
         return arr.astype(dtype, copy=True)
     return arr.view(dtype)
 

pandas/tests/frame/test_dtypes.py

@@ -640,6 +640,22 @@ def test_astype_categoricaldtype_class_raises(self, cls):
         with tm.assert_raises_regex(TypeError, xpr):
             df['A'].astype(cls)
 
+    @pytest.mark.parametrize("dtype", ["M8", "m8"])
+    @pytest.mark.parametrize("unit", ['ns', 'us', 'ms', 's', 'h', 'm', 'D'])
+    def test_astype_from_datetimelike_to_objectt(self, dtype, unit):
+        # tests astype to object dtype
+        # gh-19223 / gh-12425
+        dtype = "{}[{}]".format(dtype, unit)
+        arr = np.array([[1, 2, 3]], dtype=dtype)
+        df = DataFrame(arr)
+        result = df.astype(object)
+        assert (result.dtypes == object).all()
+
+        if dtype.startswith('M8'):
+            assert result.iloc[0, 0] == pd.to_datetime(1, unit=unit)
+        else:
+            assert result.iloc[0, 0] == pd.to_timedelta(1, unit=unit)
+
     @pytest.mark.parametrize("arr_dtype", [np.int64, np.float64])
     @pytest.mark.parametrize("dtype", ["M8", "m8"])
     @pytest.mark.parametrize("unit", ['ns', 'us', 'ms', 's', 'h', 'm', 'D'])