BUG: Prevent Unlimited Agg Recursion with Duplicate Col Names (#21066)

Author: WillAyd

doc/source/whatsnew/v0.23.1.txt

@@ -44,7 +44,10 @@ Documentation Changes
 Bug Fixes
 ~~~~~~~~~
 
--
+Groupby/Resample/Rolling
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+- Bug in :func:`DataFrame.agg` where applying multiple aggregation functions to a :class:`DataFrame` with duplicated column names would cause a stack overflow (:issue:`21063`)
 -
 
 Conversion

pandas/core/base.py

@@ -590,9 +590,10 @@ def _aggregate_multiple_funcs(self, arg, _level, _axis):
 
         # multiples
         else:
-            for col in obj:
+            for index, col in enumerate(obj):
                 try:
-                    colg = self._gotitem(col, ndim=1, subset=obj[col])
+                    colg = self._gotitem(col, ndim=1,
+                                         subset=obj.iloc[:, index])
                     results.append(colg.aggregate(arg))
                     keys.append(col)
                 except (TypeError, DataError):
@@ -675,7 +676,6 @@ def _gotitem(self, key, ndim, subset=None):
         subset : object, default None
             subset to act on
         """
-
         # create a new object to prevent aliasing
         if subset is None:
             subset = self.obj

pandas/core/frame.py

@@ -5731,7 +5731,12 @@ def diff(self, periods=1, axis=0):
     # ----------------------------------------------------------------------
     # Function application
 
-    def _gotitem(self, key, ndim, subset=None):
+    def _gotitem(self,
+                 key,           # type: Union[str, List[str]]
+                 ndim,          # type: int
+                 subset=None    # type: Union[Series, DataFrame, None]
+                 ):
+        # type: (...) -> Union[Series, DataFrame]
         """
         sub-classes to define
         return a sliced object
@@ -5746,9 +5751,11 @@ def _gotitem(self, key, ndim, subset=None):
         """
         if subset is None:
             subset = self
+        elif subset.ndim == 1:  # is Series
+            return subset
 
         # TODO: _shallow_copy(subset)?
-        return self[key]
+        return subset[key]
 
     _agg_doc = dedent("""
     The aggregation operations are always performed over an axis, either the

pandas/tests/frame/test_apply.py

@@ -554,6 +554,14 @@ def test_apply_non_numpy_dtype(self):
         result = df.apply(lambda x: x)
         assert_frame_equal(result, df)
 
+    def test_apply_dup_names_multi_agg(self):
+        # GH 21063
+        df = pd.DataFrame([[0, 1], [2, 3]], columns=['a', 'a'])
+        expected = pd.DataFrame([[0, 1]], columns=['a', 'a'], index=['min'])
+        result = df.agg(['min'])
+
+        tm.assert_frame_equal(result, expected)
+
 
 class TestInferOutputShape(object):
     # the user has supplied an opaque UDF where