Conversation
|
why did CI not run? |
|
ok here we go |
|
According to CI this actually failed to be resolvable. |
|
StructArrays v0.6.17 with compat for Adapt v4 has been released |
|
What exactly holds this back? |
|
Many tests failed when last I ran them. Perhaps this PR just needs to be rebased on master, if those are fixed in the meantime? |
|
The tests are unrelated to this patch. I've run the test suite on main and on main with Adapt 4 and the exact same tests fail. This could probably be merged and hopefully released soon as it's blocking a few downstream PRs. |
|
The same tests as fail on CI, or different ones? On CI it's all about one / zero, for which there were various recent PRs. But I don't have bandwith to dig into this. Am a bit reluctant to release things with tests failing, regardless of assurances that it's no worse than before. |
|
Understandable. I reran tests for both versions on Macos and Linux (4 runs total) and they're all erroring the same as the CI for Julia 1.10 on x64. The one / zero errors were fixed by #773. A rebase of this PR should show the same 1 failure and 8 errors as #773 While those were running, I also fixed new CI failures in a separate PR (#776) that should be a quick review. |
1f2531a to
f2fbf81
Compare
|
Passing on 1.6 x64. x86 timed out And the remaining failures on 1.10 are preexisting. (and have fixes) |
This pull request changes the compat entry for the
Adaptpackage from3.4.0to3.4.0, 4.This keeps the compat entries for earlier versions.
Note: I have not tested your package with this new compat entry.
It is your responsibility to make sure that your package tests pass before you merge this pull request.
Edit: closes #775