BUG: DataFrame._item_cache not cleared on on .copy()

doc/source/whatsnew/v1.1.0.rst

@@ -364,6 +364,7 @@ Indexing
 - Bug in :class:`Index` constructor where an unhelpful error message was raised for ``numpy`` scalars (:issue:`33017`)
 - Bug in :meth:`DataFrame.lookup` incorrectly raising an ``AttributeError`` when ``frame.index`` or ``frame.columns`` is not unique; this will now raise a ``ValueError`` with a helpful error message (:issue:`33041`)
 - Bug in :meth:`DataFrame.iloc.__setitem__` creating a new array instead of overwriting ``Categorical`` values in-place (:issue:`32831`)
+- Bug in :meth:`DataFrame.copy` _item_cache not invalidated after copy causes post-copy value updates to not be reflected (:issue:`31784`)
 
 Missing
 ^^^^^^^

pandas/core/generic.py

@@ -5665,6 +5665,7 @@ def copy(self: FrameOrSeries, deep: bool_t = True) -> FrameOrSeries:
         dtype: object
         """
         data = self._data.copy(deep=deep)
+        self._clear_item_cache()
         return self._constructor(data).__finalize__(self)
 
     def __copy__(self: FrameOrSeries, deep: bool_t = True) -> FrameOrSeries:

pandas/tests/frame/test_api.py

@@ -540,3 +540,21 @@ def test_attrs(self):
 
         result = df.rename(columns=str)
         assert result.attrs == {"version": 1}
+
+    def test_cache_on_copy(self):
+        # GH 31784 _item_cache not cleared on copy causes incorrect reads after updates
+        df = DataFrame({"a": [1]})
+
+        df["x"] = [0]
+        df["a"]
+
+        df.copy()
+
+        df["a"].values[0] = -1
+
+        tm.assert_frame_equal(df, DataFrame({"a": [-1], "x": [0]}))
+
+        df["y"] = [0]
+
+        assert df["a"].values[0] == -1
+        tm.assert_frame_equal(df, DataFrame({"a": [-1], "x": [0], "y": [0]}))