-
Notifications
You must be signed in to change notification settings - Fork 190
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
Fix tests on Julia nightly #863
Conversation
The tests for specific exception types in `sem` are relying on the exceptions thrown from deeper in the call stack, outside of this package's control. We do still expect these cases to throw, but since we can't actually control the type of the exception, we can instead test that any exception is thrown. The test of the number of allocations from calling `mad` on a generator fails because there are nearly 10x more allocations than the upper bound used in the test. We can mark this test broken with Julia v1.10, and the test will fail if the allocations for this case are improved, be it upstream or in the implementation of `mad` (if applicable).
Simply going from Documenter 0.24 to the latest, 0.27, fixes `makedocs` for me locally. This should fix the documentation build on CI.
This is so weird. I can reproduce it locally only if there is an existing
|
It seems the Documenter issue was fixed upstream (but is not available in 1.9 yet), so this shouldn't hold back the PR, should it? |
It sounds like the upstream change that fixed the issue with Documenter will not go into 1.9 (JuliaLang/julia#49975 (comment)).
Yeah I suppose not. I had forgotten about this PR but I believe at the time I figured I'd tackle CI issues together but the Documenter thing turned out to really be A Thing. |
The tests for specific exception types in
sem
are relying on the exceptions thrown from deeper in the call stack, outside of this package's control. We do still expect these cases to throw, but since we can't actually control the type of the exception, we can instead test that any exception is thrown.The test of the number of allocations from calling
mad
on a generator fails because there are nearly 10x more allocations than the upper bound used in the test. We can mark this test broken with Julia v1.10, and the test will fail if the allocations for this case are improved, be it upstream or in the implementation ofmad
(if applicable).