BUG: Bug in xs ignored droplevel #37776
Conversation
phofl
added
Indexing
pandas/core/generic.py
Outdated
| return self[key] | ||
| if drop_level: | ||
| return self[key] | ||
| return self[[key]] |
There was a problem hiding this comment.
PITA but this causes us to make a copy instead of a view. can we re-use any of the code in the rest of the method to keep this consistent?
There was a problem hiding this comment.
Would make this pretty complex probably, but will have a look tomorrow
There was a problem hiding this comment.
This should work, but got a bit more complex. Using iloc now to select the data if droplevel=False
| @@ -3749,8 +3752,13 @@ class animal locomotion | |||
| ) | |||
|
|
|||
| else: | |||
There was a problem hiding this comment.
can you make this part of the higher level if/else
pandas/core/generic.py
Outdated
| result.index = new_index | ||
| else: | ||
| result = ( | ||
| self.iloc[:, loc] if not is_scalar(loc) else self.iloc[:, [loc]] |
There was a problem hiding this comment.
worth making is_scalar another case in the eliif (I think it will read simpler that way)
� Conflicts: � doc/source/whatsnew/v1.2.0.rst
|
thanks @phofl |
| @@ -3747,7 +3750,10 @@ class animal locomotion | |||
| name=self.index[loc], | |||
| dtype=new_values.dtype, | |||
| ) | |||
|
|
|||
| elif is_scalar(loc): | |||
| result = self.iloc[:, [loc]] | |||
There was a problem hiding this comment.
Until we implement #36195, this returns a copy and not a view. In followup can you make this self.iloc[:, slice(loc, loc + 1)], probably need to test that we're getting a view
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diff