Skip to content
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

Replace nightly tests with pre-release tests #805

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Conversation

sethaxen
Copy link
Member

The nightly CI runs have failed persistently for 9 months. As a result, these are really not useful. More useful would be to test against pre-releases, and we should actually expect CI to pass for pre-releases. That's what this PR does.

@sethaxen
Copy link
Member Author

(On a personal note, since I created the JuliaNightly workflow, GitHub notifies me every single time it fails, which is at least daily, and I cannot unsubscribe from this without unsubscribing from the entire repo. Hopefully this improves things)

Copy link
Member

@devmotion devmotion left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think that's a good idea.

@sethaxen
Copy link
Member Author

There are some test failures on v1.11.0-rc3, specifically type-inference issues for the rrules of sortslices(::AbstractMatrix; dims::Integer) and norm(::AbstractMatrix, ::Real). If we merge this right now, CI will be marked as failing on main. Ideally we would fix these failures either here or shortly after merging, but after spending some time on this, it's not at all clear to me why these are no longer type-inferrable. Alternatively, we could just skip inferrability checks here.

@oxinabox
Copy link
Member

oxinabox commented Oct 8, 2024

Alternatively, we could just skip inferrability checks here.

For nightly I am pro this.
Type inference flicks around a fair bit on nightly.
CHAINRULES_TEST_INFERRED environment variable for this.

But idk for pre-release that is perhaps useful so we can complain upstream that someone broke inference and then either get it fixed before release.
Or disable it if they think it's ok

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants