Open hartwork opened 4 days ago
Looks good to me. Thanks for picking it up! Do you think a unit test could be added to verify that the reproducer is fixed and does not regress in the future?
Looks good to me. Thanks for picking it up! Do you think a unit test could be added to verify that the reproducer is fixed and does not regress in the future?
@a-gardner1 that's an interesting idea. I'm happy to add a test based on that reproducer once it's the last thing considered blocking a merge.
I'm happy to rebase off of it after that one has been merged.
Done.
@bwendling any chance you could review the fix this coming weak? The bug is quite a bummer and I'm happy to make adjustments where needed. Thank you! :pray:
@bwendling many thanks for the review and merge! Is there a chance for a new release v0.40.3 include this pull request to get the fixes to the users? Please let me know if you could use any help on that end :pray:
Fixes #1164 Fixes #1204
I have flaky pre-commit CI caused by Yapf in multiple repositories by now and it was about time to start doing something about it… I'm grateful for the great analysis by both @a-gardner1 and @whlook.
~The pull request sits on top of #1242 for the moment purely for a chance at a green CI: I'm happy to rebase off of it after that one has been merged.~
For a convenient reproducer of the race condition based on earlier work by @whlook please see https://github.com/google/yapf/issues/1204#issuecomment-2391417931 .
If this could be reviewed and merged soon-ish, that would rock the house. Thanks in advance! :pray:
CC @bwendling @whlook @a-gardner1 @kamahen