TST: Remove some xfails in groupby tests #52065
Conversation
rhshadrach
added
Testing
|
|
||
| if groupby_func == "ngroup": | ||
| if groupby_func in ("cumcount", "corrwith", "ngroup"): |
There was a problem hiding this comment.
ive had these corrwith xfails on my radar for a bit. if we remove the xfail, should we also say "no" to #32293?
There was a problem hiding this comment.
I don't think so; replacing xfail with the attribute check means that this test will still start failing if they are ever added, but without the perf impact of using an xfail. And it is better behaved - the xfail can pass even if corrwith is added (if it doesn't happen to pass the rest of this test).
|
thanks @rhshadrach |
| if groupby_func in ("cumcount",): | ||
| msg = "DataFrame has no such method" | ||
| request.node.add_marker(pytest.mark.xfail(reason=msg)) | ||
| # first and last are entirely different between frame and groupby |
There was a problem hiding this comment.
For these cases where nothing is tested, I would prefer using pytest.skip(reason) just so it's clear that nothing is supposed to be tested
There was a problem hiding this comment.
Makes sense - I'll do a follow up.
doc/source/whatsnew/vX.X.X.rstfile if fixing a bug or adding a new feature.