FIX-#4577: Set attribute of Modin dataframe to updated value (#4588)

Signed-off-by: Karthik Velayutham <vkarthik@ponder.io>

modin/pandas/dataframe.py

@@ -2499,6 +2499,10 @@ def __setattr__(self, key, value):
             pass
         elif key in self and key not in dir(self):
             self.__setitem__(key, value)
+            # Note: return immediately so we don't keep this `key` as dataframe state.
+            # `__getattr__` will return the columns not present in `dir(self)`, so we do not need
+            # to manually track this state in the `dir`.
+            return
         elif isinstance(value, pandas.Series):
             warnings.warn(
                 "Modin doesn't allow columns to be created via a new attribute name - see "

modin/pandas/test/dataframe/test_iter.py

@@ -248,7 +248,8 @@ def test_inplace_series_ops(data):
         df_equals(modin_df, pandas_df)
 
 
-def test___setattr__():
+# Note: Tests setting an attribute that is not an existing column label
+def test___setattr__not_column():
     pandas_df = pandas.DataFrame([1, 2, 3])
     modin_df = pd.DataFrame([1, 2, 3])
 
@@ -257,6 +258,37 @@ def test___setattr__():
 
     df_equals(modin_df, pandas_df)
 
+    # While `new_col` is not a column of the dataframe,
+    # it should be accessible with __getattr__.
+    assert modin_df.new_col == pandas_df.new_col
+
+
+def test___setattr__mutating_column():
+    # Use case from issue #4577
+    pandas_df = pandas.DataFrame([[1]], columns=["col0"])
+    modin_df = pd.DataFrame([[1]], columns=["col0"])
+
+    # Replacing a column with a list should mutate the column in place.
+    pandas_df.col0 = [3]
+    modin_df.col0 = [3]
+
+    df_equals(modin_df, pandas_df)
+    # Check that the col0 attribute reflects the value update.
+    df_equals(modin_df.col0, pandas_df.col0)
+
+    pandas_df.col0 = pd.Series([5])
+    modin_df.col0 = pd.Series([5])
+
+    # Check that the col0 attribute reflects this update
+    df_equals(modin_df, pandas_df)
+
+    pandas_df.loc[0, "col0"] = 4
+    modin_df.loc[0, "col0"] = 4
+
+    # Check that the col0 attribute reflects update via loc
+    df_equals(modin_df, pandas_df)
+    assert modin_df.col0.equals(modin_df["col0"])
+
 
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
 def test_isin(data):