BUG: RollingGroupby respects __getitem__ #35513
Conversation
There was a problem hiding this comment.
Thanks @mroeschke lgtm pending green (have restarted the failing travis job, failure was unrelated)
doc/source/whatsnew/v1.1.1.rst
Outdated
| @@ -44,6 +44,10 @@ Bug fixes | |||
|
|
|||
| - | |||
|
|
|||
| **Groupby/resample/rolling** | |||
|
|
|||
| - Bug in :class:`pandas.core.groupby.RollingGroupby` where ``__getitem__`` would not select the specified column (:issue:`35486`) | |||
There was a problem hiding this comment.
maybe move to regression section above.
(this section is for bug fixes, we maybe should have a discussion in #35489 regarding the criteria for backporting bugfixes otherwise this section may end up empty)
doc/source/whatsnew/v1.1.1.rst
Outdated
| @@ -15,7 +15,7 @@ including other versions of pandas. | |||
| Fixed regressions | |||
| ~~~~~~~~~~~~~~~~~ | |||
|
|
|||
| - | |||
| - Bug in :class:`pandas.core.groupby.RollingGroupby` where ``__getitem__`` would not select the specified column (:issue:`35486`) | |||
There was a problem hiding this comment.
Minor nit but is there a way to make this more user-friendly? Fair to say something like Bug in ... where column selection was being ignored?
There was a problem hiding this comment.
Sure, ya that phrasing would be more user friendly
| df = DataFrame( | ||
| {"a": [1, 2, 3, 2], "b": [4.0, 2.0, 3.0, 1.0], "c": [10, 20, 30, 20]} | ||
| ) | ||
| result = df.groupby("a")[["b"]].rolling(2).max() |
There was a problem hiding this comment.
can you also test a df.groupby('a')['b'] as well
|
thanks for addressing @mroeschke |
|
@meeseeksdev backport to 1.1.x |
Co-authored-by: Matthew Roeschke <[email protected]>
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diff