-
Notifications
You must be signed in to change notification settings - Fork 6.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Please don't push spuriously numbered git tags #7650
Comments
CC @pmeier any idea what these tags are? |
They are coming a mechanism that is used in PyTorch core to conditionally run CI jobs on PRs. Without going into too much details, whenever you put a That being said, there are two things wrong here:
For now we can safely delete these tags. In the future, if move forward with more conditional CI, e.g. #7507, the conda-forge bot might need to be fixed. |
Deletion mechanism seems to work. I've removed the label from #5009 and the tag is gone now: https://github.com/pytorch/vision/releases/tag/ciflow%2Fdefault%2F5009 |
I've removed the label from all the open PRs that still had them. All corresponding tags should be gone now. |
Closing this for now, since as of now there are no more of the tags in question. However, they might get a comeback if we need this for our CI infrastructure. @h-vetinari I'll give you a ping in that case. |
Hi!
I help package various pytorch (and many other) packages in conda-forge. We rely on an automated bot to scrape the various repositories to check if new versions have been published, and tags like
ciflow/default/5009
make this much more difficult, because the bot now thinks that a version "5009" of torchvision has been published, and then fails with:Of course, one could argue that the bot should be(come) smarter, but OTOH, that tag looks pretty spurious AFAICT. Is it really necessary to publish such tags on the main repo? Could you remove them and avoid that going forward? That would be amazing (otherwise we potentially fall far behind with updates, because it takes a human to figure out why the bot has stopped).
CC @NicolasHug
The text was updated successfully, but these errors were encountered: