Fix left join turning into outer join #19624
Conversation
pandas/tests/frame/test_join.py
Outdated
| @@ -31,6 +31,11 @@ def right(): | |||
| return DataFrame({'b': [300, 100, 200]}, index=[3, 1, 2]) | |||
|
|
|||
|
|
|||
| @pytest.fixture | |||
| def right_non_unique(): | |||
There was a problem hiding this comment.
Why'd you make this a fixture? If it's only used in one place then I'd prefer it be defined where it's used.
There was a problem hiding this comment.
Well, the below test is just the bare minimum, I think test_join needs to cover many more situations. But I will pull it into the method for now.
pandas/tests/frame/test_join.py
Outdated
|
|
||
| def test_join_left_sequence_non_unique_index(left, right, right_non_unique): | ||
| # left join sequence of dataframes with non-unique indices (issue #19607) | ||
| joined = left.join([right_non_unique], how='left') |
There was a problem hiding this comment.
Could you also build the expected output and tm.assert_frame_equal against that?
There was a problem hiding this comment.
That should definitely be possible
Codecov Report
@@ Coverage Diff @@
## master #19624 +/- ##
==========================================
+ Coverage 91.59% 91.62% +0.02%
==========================================
Files 150 150
Lines 48795 48798 +3
==========================================
+ Hits 44696 44711 +15
+ Misses 4099 4087 -12
Continue to review full report at Codecov.
|
git diff upstream/master -u -- "*.py" | flake8 --diff