Repository navigation
TST: Patch failing tests caused by mpl-dev - #11292
Conversation
|
The fixes needed suggest very much that But I'll admit to being out of my league here - hopefully, @astrofrog has a better idea of what is actually going on. |
|
I agree with @mhvk that xfailing the tests would be a better short term approach. But we really should just fix this properly - has anyone had a chance to bisect to figure out the MPL commit that broke things? |
|
Actually I think we should maybe not even xfail tests but rather just comment out the CI config temporarily if it becomes a problem. Is it really an issue since each job has its own status though? |
Well, I wouldn't care if it is just |
|
Okay, I see the discussions at matplotlib/matplotlib#19432 , so perhaps we just xfail them for mpldev for now. When they pass again, xfailed test would fail because of xfail strict, and we can remove the xfail. 💭 |
6ca1012 to
41ce59f
Compare
pllim
added
Upstream Action Required
mhvk
left a comment
mhvk
left a comment
There was a problem hiding this comment.
Yes, I think this is the best way forward. I do think we should try to keep all the jobs pass - it helps to just see at a glance whether things can be merged. And we will now get warning if things improve on the matplotlib side!
|
Thanks! Merging then.
I agree! The coordinates and mysterious table failures are another story though... 😹 |
Description
This pull request is to patch failing tests caused by dev version matplotlib. I have no idea why it is failing but I also don't want it to be causing failures in unrelated PRs. If there is a better way to fix this, please let me know.
Temporarily address #11267
Also see matplotlib/matplotlib#19419 for why my version check is so shady.