BUG: Handle variables named 'name' in get_dummies, #12180

doc/source/whatsnew/v0.18.0.txt

@@ -521,6 +521,7 @@ Bug Fixes
 - Bug in ``read_csv`` when reading from a ``StringIO`` in threads (:issue:`11790`)
 - Bug in not treating ``NaT`` as a missing value in datetimelikes when factorizing & with ``Categoricals`` (:issue:`12077`)
 - Bug in getitem when the values of a ``Series`` were tz-aware (:issue:`12089`)
+- Bug in ``Series.str.get_dummies`` when one of the variables was 'name' (:issue:`12180`)
 
 
 

pandas/core/strings.py

@@ -1105,9 +1105,11 @@ def _wrap_result(self, result, use_codes=True, name=None):
 
         if not hasattr(result, 'ndim'):
             return result
-        name = name or getattr(result, 'name', None) or self._orig.name
 
         if result.ndim == 1:
+            # Wait until we are sure result is a Series or Index before
+            # checking attributes (GH 12180)
+            name = name or getattr(result, 'name', None) or self._orig.name
             if isinstance(self._orig, Index):
                 # if result is a boolean np.array, return the np.array
                 # instead of wrapping it into a boolean Index (GH 8875)

pandas/tests/test_strings.py

@@ -812,6 +812,14 @@ def test_get_dummies(self):
             idx = Index(['a|b', 'a|c', 'b|c'])
             idx.str.get_dummies('|')
 
+        # GH 12180
+        # Dummies named 'name' should work as expected
+        s = Series(['a', 'b,name', 'b'])
+        result = s.str.get_dummies(',')
+        expected = DataFrame([[1, 0, 0], [0, 1, 1], [0, 1, 0]],
+                             columns=['a', 'b', 'name'])
+        tm.assert_frame_equal(result, expected)
+
     def test_join(self):
         values = Series(['a_b_c', 'c_d_e', np.nan, 'f_g_h'])
         result = values.str.split('_').str.join('_')