BUG: 2D ndarray of dtype 'object' is always copied upon construction

doc/source/whatsnew/v1.4.0.rst

@@ -178,6 +178,7 @@ Categorical
 Datetimelike
 ^^^^^^^^^^^^
 - Bug in :func:`to_datetime` returning pd.NaT for inputs that produce duplicated values, when ``cache=True`` (:issue:`42259`)
+- Bug in :class:`DataFrame` constructor unnecessarily copying non-datetimelike 2D object arrays (:issue:`39272`)
 -
 
 Timedelta

pandas/core/internals/construction.py

@@ -78,15 +78,11 @@
     ArrayManager,
     SingleArrayManager,
 )
-from pandas.core.internals.blocks import (
-    ensure_block_shape,
-    new_block,
-)
 from pandas.core.internals.managers import (
     BlockManager,
     SingleBlockManager,
+    create_block_manager_from_array,
     create_block_manager_from_arrays,
-    create_block_manager_from_blocks,
 )
 
 if TYPE_CHECKING:
@@ -342,37 +338,29 @@ def ndarray_to_mgr(
 
         return ArrayManager(arrays, [index, columns], verify_integrity=False)
 
-    values = values.T
+    array = values.T
 
     # if we don't have a dtype specified, then try to convert objects
     # on the entire block; this is to convert if we have datetimelike's
     # embedded in an object type
-    if dtype is None and is_object_dtype(values.dtype):
-
-        if values.ndim == 2 and values.shape[0] != 1:
+    if dtype is None and is_object_dtype(array.dtype):
+        if array.ndim == 2 and array.shape[0] != 1:
             # transpose and separate blocks
-
-            dtlike_vals = [maybe_infer_to_datetimelike(row) for row in values]
-            dvals_list = [ensure_block_shape(dval, 2) for dval in dtlike_vals]
-
-            # TODO: What about re-joining object columns?
-            block_values = [
-                new_block(dvals_list[n], placement=n, ndim=2)
-                for n in range(len(dvals_list))
+            maybe_datetime = [
+                maybe_infer_to_datetimelike(instance) for instance in array
             ]
-
+            # don't convert (and copy) the objects if no type inference occurs
+            if any(
+                not is_dtype_equal(instance.dtype, array.dtype)
+                for instance in maybe_datetime
+            ):
+                return create_block_manager_from_arrays(
+                    maybe_datetime, columns, [columns, index]
+                )
         else:
-            datelike_vals = maybe_infer_to_datetimelike(values)
-            nb = new_block(datelike_vals, placement=slice(len(columns)), ndim=2)
-            block_values = [nb]
-    else:
-        nb = new_block(values, placement=slice(len(columns)), ndim=2)
-        block_values = [nb]
-
-    if len(columns) == 0:
-        block_values = []
+            array = maybe_infer_to_datetimelike(array)
 
-    return create_block_manager_from_blocks(block_values, [columns, index])
+    return create_block_manager_from_array(array, [columns, index])
 
 
 def _check_values_indices_shape_match(

pandas/core/internals/managers.py

@@ -1791,6 +1791,26 @@ def create_block_manager_from_arrays(
     return mgr
 
 
+def create_block_manager_from_array(
+    array,
+    axes: list[Index],
+    consolidate: bool = True,
+) -> BlockManager:
+    assert isinstance(axes, list)
+    assert all(isinstance(x, Index) for x in axes)
+
+    array = _extract_array(array)
+
+    try:
+        block = new_block(values=array, placement=slice(0, len(axes[0])), ndim=2)
+        mgr = BlockManager([block], axes)
+    except ValueError as e:
+        raise construction_error(array.shape[0], array.shape[1:], axes, e)
+    if consolidate:
+        mgr._consolidate_inplace()
+    return mgr
+
+
 def construction_error(
     tot_items: int,
     block_shape: Shape,

pandas/tests/frame/test_constructors.py

@@ -253,6 +253,20 @@ def test_constructor_dtype_nocast_view_2d_array(self):
         should_be_view[0][0] = 97
         assert df.values[0, 0] == 97
 
+    @td.skip_array_manager_invalid_test
+    def test_1d_object_array_does_not_copy(self):
+        # https://github.com/pandas-dev/pandas/issues/39272
+        arr = np.array(["a", "b"], dtype="object")
+        df = DataFrame(arr)
+        assert np.shares_memory(df.values, arr)
+
+    @td.skip_array_manager_invalid_test
+    def test_2d_object_array_does_not_copy(self):
+        # https://github.com/pandas-dev/pandas/issues/39272
+        arr = np.array([["a", "b"], ["c", "d"]], dtype="object")
+        df = DataFrame(arr)
+        assert np.shares_memory(df.values, arr)
+
     def test_constructor_dtype_list_data(self):
         df = DataFrame([[1, "2"], [None, "a"]], dtype=object)
         assert df.loc[1, 0] is None