BUG: get_indexer_not_unique inconsistent return types vs get_indexer p…

…andas-dev#16819
ri938 · Jul 4, 2017 · 27d126f · 27d126f
1 parent 565bf4c
commit 27d126f
Show file tree

Hide file tree

Showing 4 changed files with 13 additions and 4 deletions.
diff --git a/doc/source/whatsnew/v0.20.3.txt b/doc/source/whatsnew/v0.20.3.txt
@@ -38,7 +38,7 @@ Bug Fixes
 ~~~~~~~~~
 - Fixed issue with dataframe scatter plot for categorical data that reports incorrect column key not found when categorical data is used for plotting (:issue:`16199`)
 - Bug in reindexing on an empty CategoricalIndex (:issue:`16770`)
-
+- Bug in get_indexer_non_unique inconsistent return type with get_indexer (:issue:`16819`)
 
 
 Conversion

diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
@@ -2704,7 +2704,7 @@ def get_indexer_non_unique(self, target):
             tgt_values = target._values
 
         indexer, missing = self._engine.get_indexer_non_unique(tgt_values)
-        return Index(indexer), missing
+        return indexer, missing
 
     def get_indexer_for(self, target, **kwargs):
         """

diff --git a/pandas/tests/indexes/test_base.py b/pandas/tests/indexes/test_base.py
@@ -1131,6 +1131,16 @@ def test_get_indexer_strings(self):
         with pytest.raises(TypeError):
             idx.get_indexer(['a', 'b', 'c', 'd'], method='pad', tolerance=2)
 
+    def test_get_indexer_consistency(self):
+        # See GH 16819
+        for name, index in self.indices.items():
+            expected = index.get_indexer(index[0:2])
+            result, _ = index.get_indexer_non_unique(index[0:2])
+            assert isinstance(expected, np.ndarray)
+            assert expected.dtype == np.intp
+            assert isinstance(result, np.ndarray)
+            assert result.dtype == np.intp
+
     def test_get_loc(self):
         idx = pd.Index([0, 1, 2])
         all_methods = [None, 'pad', 'backfill', 'nearest']

diff --git a/pandas/tests/indexes/test_category.py b/pandas/tests/indexes/test_category.py
@@ -386,8 +386,7 @@ def test_reindexing(self):
             expected = oidx.get_indexer_non_unique(finder)[0]
 
             actual = ci.get_indexer(finder)
-            tm.assert_numpy_array_equal(
-                expected.values, actual, check_dtype=False)
+            tm.assert_numpy_array_equal(expected, actual, check_dtype=True)
 
     def test_reindex_dtype(self):
         c = CategoricalIndex(['a', 'b', 'c', 'a'])