TST: MultiIndex.get_indexer with na/missing #48877
Conversation
lukemanley
added
Bug
Missing-data
pandas/core/indexes/base.py
Outdated
| @@ -4028,6 +4028,11 @@ def get_indexer( | |||
| target, method=method, limit=limit, tolerance=tolerance | |||
| ) | |||
|
|
|||
| if self._is_multi and self._contains_na_any_level: | |||
There was a problem hiding this comment.
Id prefer fixing this on the engine level and not special casing
There was a problem hiding this comment.
ok, i'll take another look
There was a problem hiding this comment.
This is probably non trivial, not sure what's wrong there
There was a problem hiding this comment.
not seeing an obvious way to fix this on the engine level. If you look a few lines above this, categorical is already special cased to handle nans in a similar way. Are you ok with special casing for now?
|
Could you rebase? Hopefully, all tests should pass now. Also, would be good to add the issues to my whatsnew note. wdyt? |
lukemanley
changed the title
Updated this PR to add tests only and close linked issues. If you think any of these tests are redundant, I'm happy to remove some (all reported for the same underlying bug). |
doc/source/whatsnew/v2.0.0.rstfile if fixing a bug or adding a new feature.Update: #49442 has been merged to fix the underlying bug here. This PR has been updated to add tests for closing the linked issues which are all related to the same underlying bug in #49442.