-
Notifications
You must be signed in to change notification settings - Fork 732
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
uv >=4.8 cache error output #7479
Comments
Can you point to the uv invocation you're using? cc @BurntSushi |
Here's the relevant job from my GitHub Action
|
Do you see the same errors if you avoid using the shared GitHub Actions cache? Like with |
The issue initially showed up on 4.10, I deleted all of my caches and then tried a few different versions to narrow it down between 4.7 and 4.8. I'll set enable-cache to false and report back. |
It does still output that error with enable-cache = false on 4.8 but not 4.7 |
Okay thank you! If you run |
Thanks. I can replicate this just by cloning and running |
That does also happen for me. |
I can't replicate this as soon as I switch to a local uv build, even at the same commit. |
It looks like that job is running |
I think when you run |
Yep, updating the uv requirement in |
(I do think there may still be a bug here, maybe we forgot to bump a cache version somewhere between 0.3.1 and 0.4.10.) |
Actually, perhaps our behavior is correct... The main issue is we included a change in the cache such that 0.4.7 entries can be read by 0.4.8, but not the other way around. |
I think my suggestion would be to pin the uv version within your |
That sounds good, thanks for looking into it! |
UV 4.8 and above is outputting an error that is breaking tests that expect specific console output when caching is enabled. 4.7 works fine.
The error is:
error: Failed to build:
curses-menu @ file:///home/runner/work/curses-menu/curses-menu
Caused by: Failed to deserialize cache entry Caused by: wrong msgpack marker FixArray(2)'which comes from this run.
It doesn't seem to be causing any issues running most of my tests, only the ones that are using pyte to test the visual output of my code, which leads me to believe the error itself might not be causing the break, but that it might be being output somewhere it shouldn't be (or I'm reading output from somewhere I shouldn't be).
Do I just need to suppress this error somehow or is this an actual bug?
The text was updated successfully, but these errors were encountered: