BUG: allow duplicate column names if they are not merged upon, pandas-dev#10639

Author: jdrudolph

doc/source/whatsnew/v0.17.0.txt

@@ -108,6 +108,8 @@ Other enhancements
 - ``regex`` argument to ``DataFrame.filter`` now handles numeric column names instead of raising ``ValueError`` (:issue:`10384`).
 - ``pd.read_stata`` will now read Stata 118 type files. (:issue:`9882`)
 
+- ``pd.merge`` will now allow duplicate column names if they are not merged upon (:issue:`10639`).
+
 .. _whatsnew_0170.api:
 
 .. _whatsnew_0170.api_breaking:

pandas/tools/merge.py

@@ -402,19 +402,14 @@ def _validate_specification(self):
                 if self.left_on is None:
                     raise MergeError('Must pass left_on or left_index=True')
             else:
-                if not self.left.columns.is_unique:
-                    raise MergeError("Left data columns not unique: %s"
-                                     % repr(self.left.columns))
-
-                if not self.right.columns.is_unique:
-                    raise MergeError("Right data columns not unique: %s"
-                                     % repr(self.right.columns))
-
                 # use the common columns
                 common_cols = self.left.columns.intersection(
                     self.right.columns)
                 if len(common_cols) == 0:
                     raise MergeError('No common columns to perform merge on')
+                if not common_cols.is_unique:
+                    raise MergeError("Data columns not unique: %s"
+                                     % repr(common_cols))
                 self.left_on = self.right_on = common_cols
         elif self.on is not None:
             if self.left_on is not None or self.right_on is not None:

pandas/tools/tests/test_merge.py

@@ -843,7 +843,6 @@ def test_join_append_timedeltas(self):
         assert_frame_equal(result, expected)
 
     def test_overlapping_columns_error_message(self):
-        # #2649
         df = DataFrame({'key': [1, 2, 3],
                         'v1': [4, 5, 6],
                         'v2': [7, 8, 9]})
@@ -853,7 +852,16 @@ def test_overlapping_columns_error_message(self):
 
         df.columns = ['key', 'foo', 'foo']
         df2.columns = ['key', 'bar', 'bar']
-
+        expected = DataFrame({'key': [1, 2, 3],
+                         'v1': [4, 5, 6],
+                         'v2': [7, 8, 9],
+                         'v3': [4, 5, 6],
+                         'v4': [7, 8, 9]})
+        expected.columns = ['key', 'foo', 'foo', 'bar', 'bar']
+        assert_frame_equal(merge(df, df2), expected)
+
+        # #2649, #10639
+        df2.columns = ['key1', 'foo', 'foo']
         self.assertRaises(ValueError, merge, df, df2)
 
 def _check_merge(x, y):