BUG: DataFrame.__setitem__ silently dropping columns of a 2D value (GH#46544)

doc/source/whatsnew/v3.1.0.rst

@@ -311,6 +311,7 @@ Indexing
 - Bug in :meth:`Index.get_indexer` where ``method="pad"``, ``"backfill"``, or ``"nearest"`` returned incorrect results when the target contained ``NaT`` or ``NaN`` instead of ``-1`` (:issue:`32572`)
 - Bugs in setitem-with-expansion when adding new rows failing to keep the original dtype in some cases (:issue:`32346`, :issue:`15231`, :issue:`47503`, :issue:`6485`, :issue:`25383`, :issue:`52235`, :issue:`17026`, :issue:`56010`)
 - Bug in :meth:`DataFrame.__getitem__` raising ``InvalidIndexError`` when indexing with a tuple containing a ``slice`` on a :class:`DataFrame` with :class:`MultiIndex` columns (e.g., ``df[:, "t1"]``) (:issue:`26511`)
+- Bug in :meth:`DataFrame.__setitem__` silently keeping only the first column when assigning a 2D array to an existing column label, instead of raising as the new-column and :meth:`DataFrame.loc` cases already did (:issue:`46544`)
 - Bug in :meth:`DataFrame.at` raising ``TypeError`` when accessing a :class:`MultiIndex` with a partial date string on a :class:`DatetimeIndex` level (:issue:`43395`)
 - Bug in :meth:`DataFrame.duplicated` returning an empty :class:`Series` without the DataFrame's index when the DataFrame had no columns (:issue:`61191`)
 - Bug in :meth:`DataFrame.iloc` setitem raising ``AttributeError`` when assigning a :class:`Series` or :class:`Index` with a nullable EA dtype (e.g. ``Int64``, ``Float64``, ``boolean``) into a column with a NumPy dtype (:issue:`47776`)

pandas/core/internals/managers.py

@@ -1315,6 +1315,13 @@ def iset(
 
             # Check if we can use _iset_single fastpath
             loc = cast("int", loc)
+            if not value_is_extension_type and len(value) > 1:
+                # GH#46544 setting a single column with a 2D value; matches the
+                #  check in self.insert. value has been transposed above, so
+                #  len(value) is the number of columns in the original value.
+                raise ValueError(
+                    f"Expected a 1D array, got an array with shape {value.T.shape}"
+                )
             blkno = self.blknos[loc]
             blk = self.blocks[blkno]
             if len(blk._mgr_locs) == 1:  # TODO: fastest way to check this?

pandas/tests/frame/indexing/test_setitem.py

@@ -29,6 +29,7 @@
     PeriodIndex,
     Series,
     Timestamp,
+    concat,
     cut,
     date_range,
     notna,
@@ -740,6 +741,38 @@ def test_setitem_npmatrix_2d(self):
 
         tm.assert_frame_equal(df, expected)
 
+    @pytest.mark.parametrize("cols", [3, 1])
+    def test_setitem_2d_existing_column_raises(self, cols):
+        # GH#46544 setting an existing column with a multi-column 2D array used
+        #  to silently keep only the first column; it should raise like the
+        #  new-column and .loc cases do. cols=1 exercises the single-column
+        #  Block fastpath, cols=3 the consolidated-block path.
+        df = DataFrame(np.zeros((4, cols)))
+        msg = r"Expected a 1D array, got an array with shape \(4, 2\)"
+        with pytest.raises(ValueError, match=msg):
+            df[0] = np.arange(8).reshape(4, 2)
+
+    def test_setitem_2d_single_column_ok(self):
+        # GH#46544 an (N, 1) array is still a single column and is allowed
+        df = DataFrame(np.zeros((4, 2)))
+        df[0] = np.arange(4).reshape(4, 1)
+        expected = DataFrame({0: np.arange(4), 1: np.zeros(4)})
+        tm.assert_frame_equal(df, expected)
+
+    def test_setitem_2d_duplicate_columns_ok(self):
+        # GH#46544 when the key maps to multiple columns, a matching 2D value
+        #  fills them and must keep working
+        df = DataFrame(np.zeros((4, 3)), columns=[0, 0, 1])
+        df[0] = np.arange(8).reshape(4, 2)
+        expected = concat(
+            [
+                DataFrame(np.arange(8).reshape(4, 2), columns=[0, 0]),
+                DataFrame(np.zeros((4, 1)), columns=[1]),
+            ],
+            axis=1,
+        )
+        tm.assert_frame_equal(df, expected)
+
     @pytest.mark.parametrize("vals", [{}, {"d": "a"}])
     def test_setitem_aligning_dict_with_index(self, vals):
         # GH#47216