REF: avoid ravel/reshape in astype_nansafe, ndarray_to_mgr

pandas/core/construction.py

@@ -786,7 +786,16 @@ def _try_cast(
 
     elif dtype.kind == "U":
         # TODO: test cases with arr.dtype.kind in ["m", "M"]
-        return lib.ensure_string_array(arr, convert_na_value=False, copy=copy)
+        if is_ndarray:
+            arr = cast(np.ndarray, arr)
+            shape = arr.shape
+            if arr.ndim > 1:
+                arr = arr.ravel()
+        else:
+            shape = (len(arr),)
+        return lib.ensure_string_array(arr, convert_na_value=False, copy=copy).reshape(
+            shape
+        )
 
     elif dtype.kind in ["m", "M"]:
         return maybe_cast_to_datetime(arr, dtype)

pandas/core/dtypes/astype.py

@@ -83,12 +83,6 @@ def astype_nansafe(
     ValueError
         The dtype was a datetime64/timedelta64 dtype, but it had no unit.
     """
-    if arr.ndim > 1:
-        flat = arr.ravel()
-        result = astype_nansafe(flat, dtype, copy=copy, skipna=skipna)
-        # error: Item "ExtensionArray" of "Union[ExtensionArray, ndarray]" has no
-        # attribute "reshape"
-        return result.reshape(arr.shape)  # type: ignore[union-attr]
 
     # We get here with 0-dim from sparse
     arr = np.atleast_1d(arr)
@@ -109,7 +103,12 @@ def astype_nansafe(
         return arr.astype(dtype, copy=copy)
 
     if issubclass(dtype.type, str):
-        return lib.ensure_string_array(arr, skipna=skipna, convert_na_value=False)
+        shape = arr.shape
+        if arr.ndim > 1:
+            arr = arr.ravel()
+        return lib.ensure_string_array(
+            arr, skipna=skipna, convert_na_value=False
+        ).reshape(shape)
 
     elif is_datetime64_dtype(arr.dtype):
         if dtype == np.int64:
@@ -146,7 +145,7 @@ def astype_nansafe(
             from pandas import to_datetime
 
             return astype_nansafe(
-                to_datetime(arr).values,
+                to_datetime(arr.ravel()).values.reshape(arr.shape),
                 dtype,
                 copy=copy,
             )

pandas/core/internals/construction.py

@@ -329,18 +329,18 @@ def ndarray_to_mgr(
         values = _prep_ndarray(values, copy=copy_on_sanitize)
 
     if dtype is not None and not is_dtype_equal(values.dtype, dtype):
-        shape = values.shape
-        flat = values.ravel()
-
         # GH#40110 see similar check inside sanitize_array
         rcf = not (is_integer_dtype(dtype) and values.dtype.kind == "f")
 
         values = sanitize_array(
-            flat, None, dtype=dtype, copy=copy_on_sanitize, raise_cast_failure=rcf
+            values,
+            None,
+            dtype=dtype,
+            copy=copy_on_sanitize,
+            raise_cast_failure=rcf,
+            allow_2d=True,
         )
 
-        values = values.reshape(shape)
-
     # _prep_ndarray ensures that values.ndim == 2 at this point
     index, columns = _get_axes(
         values.shape[0], values.shape[1], index=index, columns=columns

pandas/tests/frame/test_constructors.py

@@ -70,6 +70,14 @@
 
 
 class TestDataFrameConstructors:
+    def test_constructor_from_ndarray_with_str_dtype(self):
+        # If we don't ravel/reshape around ensure_str_array, we end up
+        #  with an array of strings each of which is e.g. "[0 1 2]"
+        arr = np.arange(12).reshape(4, 3)
+        df = DataFrame(arr, dtype=str)
+        expected = DataFrame(arr.astype(str))
+        tm.assert_frame_equal(df, expected)
+
     def test_constructor_from_2d_datetimearray(self, using_array_manager):
         dti = date_range("2016-01-01", periods=6, tz="US/Pacific")
         dta = dti._data.reshape(3, 2)