Merge pull request #2965 from hayd/iss2843-left-join-on-na

BUG check both left and right indexers for None (fix 2843)

Author: hayd

pandas/tools/merge.py

@@ -209,23 +209,26 @@ def _maybe_add_join_keys(self, result, left_indexer, right_indexer):
             if name in result:
                 key_col = result[name]
 
-                if name in self.left and left_indexer is not None:
-                    na_indexer = (left_indexer == -1).nonzero()[0]
-                    if len(na_indexer) == 0:
-                        continue
-
-                    right_na_indexer = right_indexer.take(na_indexer)
-                    key_col.put(
-                        na_indexer, com.take_1d(self.right_join_keys[i],
-                                                right_na_indexer))
-                elif name in self.right and right_indexer is not None:
-                    na_indexer = (right_indexer == -1).nonzero()[0]
-                    if len(na_indexer) == 0:
-                        continue
-
-                    left_na_indexer = left_indexer.take(na_indexer)
-                    key_col.put(na_indexer, com.take_1d(self.left_join_keys[i],
-                                                        left_na_indexer))
+                if left_indexer is not None and right_indexer is not None:
+
+                    if name in self.left:
+                        na_indexer = (left_indexer == -1).nonzero()[0]
+                        if len(na_indexer) == 0:
+                            continue
+
+                        right_na_indexer = right_indexer.take(na_indexer)
+                        key_col.put(
+                            na_indexer, com.take_1d(self.right_join_keys[i],
+                                                    right_na_indexer))
+                    elif name in self.right:
+                        na_indexer = (right_indexer == -1).nonzero()[0]
+                        if len(na_indexer) == 0:
+                            continue
+
+                        left_na_indexer = left_indexer.take(na_indexer)
+                        key_col.put(na_indexer, com.take_1d(self.left_join_keys[i],
+                                                            left_na_indexer))
+
             elif left_indexer is not None:
                 if name is None:
                     name = 'key_%d' % i

pandas/tools/tests/test_merge.py

@@ -717,6 +717,24 @@ def test_merge_nosort(self):
 
         self.assert_((df.var3.unique() == result.var3.unique()).all())
 
+    def test_merge_nan_right(self):
+        df1 = DataFrame({"i1" : [0, 1], "i2" : [0, 1]})
+        df2 = DataFrame({"i1" : [0], "i3" : [0]})
+        result = df1.join(df2, on="i1", rsuffix="_")
+        expected = DataFrame({'i1': {0: 0.0, 1: 1}, 'i2': {0: 0, 1: 1},
+                              'i1_': {0: 0, 1: np.nan}, 'i3': {0: 0.0, 1: np.nan},
+                               None: {0: 0, 1: 0}}).set_index(None).reset_index()[['i1', 'i2', 'i1_', 'i3']]
+        assert_frame_equal(result, expected, check_dtype=False)
+
+        df1 = DataFrame({"i1" : [0, 1], "i2" : [0.5, 1.5]})
+        df2 = DataFrame({"i1" : [0], "i3" : [0.7]})
+        result = df1.join(df2, rsuffix="_", on='i1')
+        expected = DataFrame({'i1': {0: 0, 1: 1}, 'i1_': {0: 0.0, 1: nan},
+                              'i2': {0: 0.5, 1: 1.5}, 'i3': {0: 0.69999999999999996,
+                              1: nan}})[['i1', 'i2', 'i1_', 'i3']]
+        assert_frame_equal(result, expected)
+
+
     def test_overlapping_columns_error_message(self):
         # #2649
         df = DataFrame({'key': [1, 2, 3],