Regression in to_excel when setting duplicate column names (pandas-de…

…v#39800)
znicholls · Feb 17, 2021 · a1173dc · a1173dc
1 parent d819f65
commit a1173dc
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 2 deletions.
diff --git a/doc/source/whatsnew/v1.2.3.rst b/doc/source/whatsnew/v1.2.3.rst
@@ -15,7 +15,7 @@ including other versions of pandas.
 Fixed regressions
 ~~~~~~~~~~~~~~~~~
 
--
+- Fixed regression in :func:`pandas.to_excel` raising ``KeyError`` when giving duplicate columns with ``columns`` attribute (:issue:`39695`)
 -
 
 .. ---------------------------------------------------------------------------

diff --git a/pandas/io/formats/excel.py b/pandas/io/formats/excel.py
@@ -475,7 +475,7 @@ def __init__(
             if not len(Index(cols).intersection(df.columns)):
                 raise KeyError("passes columns are not ALL present dataframe")
 
-            if len(Index(cols).intersection(df.columns)) != len(cols):
+            if len(Index(cols).intersection(df.columns)) != len(set(cols)):
                 # Deprecated in GH#17295, enforced in 1.0.0
                 raise KeyError("Not all names specified in 'columns' are found")
 

diff --git a/pandas/tests/io/excel/test_writers.py b/pandas/tests/io/excel/test_writers.py
@@ -1305,6 +1305,15 @@ def test_raise_when_saving_timezones(self, dtype, tz_aware_fixture, path):
         with pytest.raises(ValueError, match="Excel does not support"):
             df.to_excel(path)
 
+    def test_excel_duplicate_columns_with_names(self, path):
+        # GH#39695
+        df = DataFrame({"A": [0, 1], "B": [10, 11]})
+        df.to_excel(path, columns=["A", "B", "A"], index=False)
+
+        result = pd.read_excel(path)
+        expected = DataFrame([[0, 10, 0], [1, 11, 1]], columns=["A", "B", "A.1"])
+        tm.assert_frame_equal(result, expected)
+
 
 class TestExcelWriterEngineTests:
     @pytest.mark.parametrize(