API: re-allow duplicate index level names

Author: jorisvandenbossche

pandas/core/indexes/multi.py

@@ -672,30 +672,18 @@ def _set_names(self, names, level=None, validate=True):
 
         if level is None:
             level = range(self.nlevels)
-            used = {}
         else:
             level = [self._get_level_number(l) for l in level]
-            used = {self.levels[l].name: l
-                    for l in set(range(self.nlevels)) - set(level)}
 
         # set the name
         for l, name in zip(level, names):
             if name is not None:
-
                 # GH 20527
                 # All items in 'names' need to be hashable:
                 if not is_hashable(name):
                     raise TypeError('{}.name must be a hashable type'
                                     .format(self.__class__.__name__))
-
-                if name in used:
-                    raise ValueError(
-                        'Duplicated level name: "{}", assigned to '
-                        'level {}, is already used for level '
-                        '{}.'.format(name, l, used[name]))
-
             self.levels[l].rename(name, inplace=True)
-            used[name] = l
 
     names = property(fset=_set_names, fget=_get_names,
                      doc="Names of levels in MultiIndex")

pandas/tests/frame/test_alter_axes.py

@@ -130,19 +130,24 @@ def test_set_index2(self):
         result = df.set_index(df.C)
         assert result.index.name == 'C'
 
-    @pytest.mark.parametrize('level', ['a', pd.Series(range(3), name='a')])
+    @pytest.mark.parametrize(
+        'level', ['a', pd.Series(range(0, 8, 2), name='a')])
     def test_set_index_duplicate_names(self, level):
-        # GH18872
+        # GH18872 - GH19029
         df = pd.DataFrame(np.arange(8).reshape(4, 2), columns=['a', 'b'])
 
         # Pass an existing level name:
         df.index.name = 'a'
-        pytest.raises(ValueError, df.set_index, level, append=True)
-        pytest.raises(ValueError, df.set_index, [level], append=True)
+        expected = pd.MultiIndex.from_tuples([(0, 0), (1, 2), (2, 4), (3, 6)],
+                                             names=['a', 'a'])
+        result = df.set_index(level, append=True)
+        tm.assert_index_equal(result.index, expected)
+        result = df.set_index([level], append=True)
+        tm.assert_index_equal(result.index, expected)
 
         # Pass twice the same level name:
         df.index.name = 'c'
-        pytest.raises(ValueError, df.set_index, [level, level])
+        # pytest.raises(ValueError, df.set_index, [level, level])
 
     def test_set_index_nonuniq(self):
         df = DataFrame({'A': ['foo', 'foo', 'foo', 'bar', 'bar'],

pandas/tests/groupby/test_categorical.py

@@ -553,14 +553,14 @@ def test_as_index():
         columns=['cat', 'A', 'B'])
     tm.assert_frame_equal(result, expected)
 
-    # another not in-axis grouper
-    s = Series(['a', 'b', 'b'], name='cat2')
+    # another not in-axis grouper (conflicting names in index)
+    s = Series(['a', 'b', 'b'], name='cat')
     result = df.groupby(['cat', s], as_index=False, observed=True).sum()
     tm.assert_frame_equal(result, expected)
 
-    # GH18872: conflicting names in desired index
-    with pytest.raises(ValueError):
-        df.groupby(['cat', s.rename('cat')], observed=True).sum()
+    # # GH18872: conflicting names in desired index
+    # with pytest.raises(ValueError):
+    #     df.groupby(['cat', s.rename('cat')], observed=True).sum()
 
     # is original index dropped?
     group_columns = ['cat', 'A']

pandas/tests/indexes/test_multi.py

@@ -656,22 +656,22 @@ def test_constructor_nonhashable_names(self):
         # With .set_names()
         tm.assert_raises_regex(TypeError, message, mi.set_names, names=renamed)
 
-    @pytest.mark.parametrize('names', [['a', 'b', 'a'], ['1', '1', '2'],
-                                       ['1', 'a', '1']])
-    def test_duplicate_level_names(self, names):
-        # GH18872
-        pytest.raises(ValueError, pd.MultiIndex.from_product,
-                      [[0, 1]] * 3, names=names)
-
-        # With .rename()
-        mi = pd.MultiIndex.from_product([[0, 1]] * 3)
-        tm.assert_raises_regex(ValueError, "Duplicated level name:",
-                               mi.rename, names)
-
-        # With .rename(., level=)
-        mi.rename(names[0], level=1, inplace=True)
-        tm.assert_raises_regex(ValueError, "Duplicated level name:",
-                               mi.rename, names[:2], level=[0, 2])
+    # @pytest.mark.parametrize('names', [['a', 'b', 'a'], ['1', '1', '2'],
+    #                                    ['1', 'a', '1']])
+    # def test_duplicate_level_names(self, names):
+    #     # GH18872
+    #     pytest.raises(ValueError, pd.MultiIndex.from_product,
+    #                   [[0, 1]] * 3, names=names)
+
+    #     # With .rename()
+    #     mi = pd.MultiIndex.from_product([[0, 1]] * 3)
+    #     tm.assert_raises_regex(ValueError, "Duplicated level name:",
+    #                            mi.rename, names)
+
+    #     # With .rename(., level=)
+    #     mi.rename(names[0], level=1, inplace=True)
+    #     tm.assert_raises_regex(ValueError, "Duplicated level name:",
+    #                            mi.rename, names[:2], level=[0, 2])
 
     def assert_multiindex_copied(self, copy, original):
         # Levels should be (at least, shallow copied)

pandas/tests/reshape/test_pivot.py

@@ -1729,10 +1729,17 @@ def test_crosstab_with_numpy_size(self):
         tm.assert_frame_equal(result, expected)
 
     def test_crosstab_dup_index_names(self):
-        # GH 13279, GH 18872
+        # GH 13279
         s = pd.Series(range(3), name='foo')
         pytest.raises(ValueError, pd.crosstab, s, s)
 
+        result = pd.crosstab(s, s)
+        expected_index = pd.Index(range(3), name='foo')
+        expected = pd.DataFrame(np.eye(3, dtype=np.int64),
+                                index=expected_index,
+                                columns=expected_index)
+        tm.assert_frame_equal(result, expected)
+
     @pytest.mark.parametrize("names", [['a', ('b', 'c')],
                                        [('a', 'b'), 'c']])
     def test_crosstab_tuple_name(self, names):