Backport PR pandas-dev#35590: BUG: validate index/data length match in DataFrame construction (pandas-dev#35597)

Co-authored-by: jbrockmendel <[email protected]>

Author: meeseeksmachine

doc/source/whatsnew/v1.1.1.rst

@@ -50,6 +50,10 @@ Categorical
 
 -
 
+**DataFrame**
+- Bug in :class:`DataFrame` constructor failing to raise ``ValueError`` in some cases when data and index have mismatched lengths (:issue:`33437`)
+-
+
 .. ---------------------------------------------------------------------------
 
 .. _whatsnew_111.contributors:

pandas/core/internals/blocks.py

@@ -105,7 +105,6 @@ class Block(PandasObject):
     is_extension = False
     _can_hold_na = False
     _can_consolidate = True
-    _verify_integrity = True
     _validate_ndim = True
 
     @classmethod
@@ -1525,7 +1524,6 @@ class ExtensionBlock(Block):
     """
 
     _can_consolidate = False
-    _verify_integrity = False
     _validate_ndim = False
     is_extension = True
 
@@ -2613,7 +2611,6 @@ def _replace_coerce(
 class CategoricalBlock(ExtensionBlock):
     __slots__ = ()
     is_categorical = True
-    _verify_integrity = True
     _can_hold_na = True
 
     should_store = Block.should_store

pandas/core/internals/managers.py

@@ -312,7 +312,7 @@ def _verify_integrity(self) -> None:
         mgr_shape = self.shape
         tot_items = sum(len(x.mgr_locs) for x in self.blocks)
         for block in self.blocks:
-            if block._verify_integrity and block.shape[1:] != mgr_shape[1:]:
+            if block.shape[1:] != mgr_shape[1:]:
                 raise construction_error(tot_items, block.shape[1:], self.axes)
         if len(self.items) != tot_items:
             raise AssertionError(

pandas/tests/frame/test_constructors.py

@@ -2619,6 +2619,12 @@ class DatetimeSubclass(datetime):
         data = pd.DataFrame({"datetime": [DatetimeSubclass(2020, 1, 1, 1, 1)]})
         assert data.datetime.dtype == "datetime64[ns]"
 
+    def test_with_mismatched_index_length_raises(self):
+        # GH#33437
+        dti = pd.date_range("2016-01-01", periods=3, tz="US/Pacific")
+        with pytest.raises(ValueError, match="Shape of passed values"):
+            DataFrame(dti, index=range(4))
+
 
 class TestDataFrameConstructorWithDatetimeTZ:
     def test_from_dict(self):