Sorry, something went wrong.
|
All PRs right now have this problem: https://dev.azure.com/Python/cpython/_build/results?buildId=87889&view=logs&j=91c152bd-7320-5194-b252-1404e56e2478&t=c7e99cd8-4756-5292-d34b-246ff5fc615f Another examples: |
Sorry, something went wrong.
|
Windows build error on Azure Pipelines is tracked by: https://bugs.python.org/issue45220 |
Sorry, something went wrong.
|
Rebased, CI now passes. |
Sorry, something went wrong.
There was a problem hiding this comment.
LGTM. @pitrou: Would you mind to double check this multiprocessing test fix?
I'm kind of unhappy that you have to prepend "/". SharedMemory._prepend_leading_slash does a magic dance (add/remove "/" prefix) with SharedMemory(name) constructor parameter and in SharedMemory.name property. But I don't think that it's worth it to add something just for this very specific test. We can always revisit the code later if we start to have issue on some corner cases.
Sorry, something went wrong.
|
I'm kind of unhappy that you have to prepend "/". SharedMemory._prepend_leading_slash does a magic dance (add/remove "/" prefix) with SharedMemory(name) constructor parameter and in SharedMemory.name property Yes, me too. I've tried several other things like smm.name (was not producing the correct result by raising KeyError). After several attempts this was the simplest solution. And since we only run this part on posix, it should work correctly. |
Sorry, something went wrong.
|
Since the slash hack is tests-only, let's have it. We can fix the fix later if we find a way to do so. |
Sorry, something went wrong.
|
GH-28499 is a backport of this pull request to the 3.10 branch. |
Sorry, something went wrong.
|
GH-28500 is a backport of this pull request to the 3.9 branch. |
Sorry, something went wrong.
https://bugs.python.org/issue45209