BUG: DataFrame.at with CategoricalIndex #41846
Conversation
jbrockmendel
commented
Jun 7, 2021
•
jbrockmendel
commented
- closes #xxxx
- tests added / passed
- Ensure all linting tests pass, see here for how to run them
- whatsnew entry
gfyoung
added
Bug
Indexing
|
release note? |
|
whatsnew added + green |
| col = self.columns.get_loc(col) | ||
| index = self.index.get_loc(index) | ||
| return self._get_value(index, col, takeable=True) | ||
| except AttributeError: |
There was a problem hiding this comment.
once this is merged ill push my branch that implements IntervalTree.get_loc, getting rid of the AttributeError
|
|
||
| Notes | ||
| ----- | ||
| Assumes that index and columns both have ax._index_as_unique; |
There was a problem hiding this comment.
don't we have this as a base Index property? (or it can raise NotImplementedError), e.g. dont' like the AttributeError checking here.
There was a problem hiding this comment.
ill rephrase, it isn't that they have the attribute, it is that this is only called if self.index._index_as_unique and self.columns._index_as_unique
|
This is for 1.3? |
|
don't add things to 1.3 |
|
This has a release note in 1.3.0 whatsnew. if not 1.3 we need to request it moved. any suggestions how to track this. |
|
i undestand and it prob will be merged but let's look first |
|
its a bugfix so i wrote the note for 1.3, wont be the end of the world if it has to wait for the next one |
|
should i move release note to 1.4 file? |
not yet. I think Jeff just didn't want stuff added to the 1.3 milestone for now. |
|
if the RC is put off until tomorrow, can this go in? it is a blocker for #41956 |
|
Doesn't look like there's a fix for the wheel issue. So may release what we
have.
…On Sat, 12 Jun 2021, 22:52 jbrockmendel, ***@***.***> wrote:
if the RC is put off until tomorrow, can this go in? it is a blocker for
#41956 <#41956>
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#41846 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADEMUXI6G5LUOT2QSZAPY5LTSPJKVANCNFSM46G2R5QA>
.
|
tbc. I should have said we are not necessarily putting off the rc, and that if the wheel issue is not fixed quickly, we could release rc0 and then follow-up with a rc1. So the same backport rules apply. We have a rc0, but may not advertise the fact. So the rc is not put-off, it has effectively happened. |
|
@meeseeksdev backport 1.3.x |
|
Something went wrong ... Please have a look at my logs. |
Co-authored-by: jbrockmendel <[email protected]>
