implement test for GH pandas-dev#58446

pandas/tests/groupby/test_groupby.py

@@ -2954,3 +2954,38 @@ def test_groupby_dropna_with_nunique_unique():
     )
 
     tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("as_index_flag", [False])
+def test_groupby_agg_namedagg_with_duplicate_columns(as_index_flag):
+    # GH #58446
+    df = DataFrame(
+        {
+            "col1": [2, 1, 1, 0, 2, 0],
+            "col2": [4, 5, 36, 7, 4, 5],
+            "col3": [3.1, 8.0, 12, 10, 4, 1.1],
+            "col4": [17, 3, 16, 15, 5, 6],
+            "col5": [-1, 3, -1, 3, -2, -1],
+        }
+    )
+
+    result = df.groupby(by=["col1", "col1", "col2"], as_index=as_index_flag).agg(
+        new_col=pd.NamedAgg(column="col1", aggfunc="min"),
+        new_col1=pd.NamedAgg(column="col1", aggfunc="max"),
+        new_col2=pd.NamedAgg(column="col2", aggfunc="count"),
+    )
+
+    expected = DataFrame(
+        {
+            "col1": [0, 0, 1, 1, 2],
+            "col2": [5, 7, 5, 36, 4],
+            "new_col": [0, 0, 1, 1, 2],
+            "new_col1": [0, 0, 1, 1, 2],
+            "new_col2": [1, 1, 1, 1, 2],
+        }
+    )
+
+    if not as_index_flag:
+        expected.reset_index(drop=True, inplace=True)
+
+    tm.assert_frame_equal(result, expected)