[JACK-Client] All METADATA_ constants are available on Linux#13639
[JACK-Client] All METADATA_ constants are available on Linux#13639srittau merged 5 commits intopython:mainfrom
Conversation
|
Background: When I'm running stubtest locally on my Linux system, it complains that these constants are missing. And they can be imported. |
This comment has been minimized.
This comment has been minimized.
| @@ -0,0 +1,2 @@ | |||
| # The available constants differ based on the local environment. | |||
| jack\.METADATA_.* | |||
There was a problem hiding this comment.
If your local install now sees all the METADATA_ * constants. Wouldn't this now fail for you locally with an "unused allowlist entry" error?
I guess the idea is that it's fine to fail for you, as long as it passes on the CI.
Still I'm really curious about why it's different. My local environment aggrees with the CI
My setup:
There was a problem hiding this comment.
Wouldn't this now fail for you locally with an "unused allowlist entry" error?
AFAIK there are some heuristics that prevent that message for (some) regexps.
Edit: Here is a quote from https://mypy.readthedocs.io/en/latest/_sources/stubtest.rst.txt:
Note if an allowlist entry is a regex that matches the empty string,
stubtest will never consider it unused.
Edit 2: Of course that's not the case here, but there might be other variables that are not available instead? I'll test this some more in the next few days.
There was a problem hiding this comment.
I'm using Ubuntu 24.10, which has a slightly newer libjack-dev version (1:0.126.0-2build3). I assume that's the reason the constants are now present. But testing it again, I now get
note: unused allowlist entry jack\.METADATA_.*
I'll fix the allowlist.
|
According to mypy_primer, this change has no effect on the checked open source code. 🤖🎉 |
2 participants
No description provided.