BUG-24212 clarify test

pandas/core/reshape/merge.py

@@ -777,6 +777,7 @@ def _get_join_info(self):
  and not isinstance(self.left.index, MultiIndex)):
  # if values missing (-1) from right index,
  # take from left index instead
+ print(right_indexer)
  join_list = join_index.to_numpy()
  absent = right_indexer == -1
  join_list[absent] = self.left.index.to_numpy()[absent]

pandas/tests/reshape/merge/test_merge.py

@@ -908,6 +908,25 @@ def test_merge_two_empty_df_no_division_error(self):
  with np.errstate(divide='raise'):
  merge(a, a, on=('a', 'b'))
 
+ def test_merge_on_index_with_more_values(self):
+ # GH 24212
+ # pd.merge gets [-1, -1, 0, 1] as right_indexer, ensure that -1 is
+ # interpreted as a missing value instead of the last element
+ df1 = pd.DataFrame([[1, 2], [2, 4], [3, 6], [4, 8]],
+ columns=['a', 'b'])
+ df2 = pd.DataFrame([[3, 30], [4, 40]],
+ columns=['a', 'c'])
+ df1.set_index('a', drop=False, inplace=True)
+ df2.set_index('a', inplace=True)
+ result = pd.merge(df1, df2, left_index=True, right_on='a', how='left')
+ expected = pd.DataFrame([[1, 2, np.nan],
+ [2, 4, np.nan],
+ [3, 6, 30.0],
+ [4, 8, 40.0]],
+ columns=['a', 'b', 'c'])
+ expected.set_index('a', drop=False, inplace=True)
+ assert_frame_equal(result, expected)
+
 
 def _check_merge(x, y):
  for how in ['inner', 'left', 'outer']:
@@ -1102,17 +1121,6 @@ def test_merge_incompat_dtypes_error(self, df1_vals, df2_vals):
  with pytest.raises(ValueError, match=msg):
  pd.merge(df2, df1, on=['A'])
 
- def test_merge_on_index_with_more_values(self):
- # GH 24212
- df1 = pd.DataFrame([[1, 2], [2, 4], [3, 6], [4, 8]],
- columns=['a', 'b'])
- df2 = pd.DataFrame([[3, 30], [4, 40]],
- columns=['a', 'c'])
- df1.set_index('a', drop=False, inplace=True)
- df2.set_index('a', inplace=True)
- result = pd.merge(df1, df2, left_index=True, right_on='a', how='left')
- assert 1 in result.index
-
 
 @pytest.fixture
 def left():