gh-140042: Removing unsafe call to sqlite3_shutdown#141690
gh-140042: Removing unsafe call to sqlite3_shutdown#141690encukou merged 3 commits intopython:mainfrom
Conversation
|
All commit authors signed the Contributor License Agreement. |
|
Most changes to Python require a NEWS entry. Add one using the blurb_it web app or the blurb command-line tool. If this change has little impact on Python users, wait for a maintainer to apply the |
encukou
left a comment
There was a problem hiding this comment.
Thank you.
I don't see a good way to add a test here -- we'd need to inject a failure in the module initialization logic, and then watch undefined behaviour.
If we want to clean up properly, we should call sqlite3_shutdown in e.g. module_free. That's not the case here -- sqlite3 leaks in the “happy” path. Also leaking on error isn't a big issue.
(Presumably we leak a fixed amount of resources per process, so adding proper calls to sqlite3_shutdown isn't a big priority.)
@erlend-aasland, do you want to take a look?
I'll merge around Thursday if there are no objections.
|
Thanks @prithviraj-chaudhuri for the PR, and @encukou for merging it 🌮🎉.. I'm working now to backport this PR to: 3.13, 3.14. |
|
GH-141793 is a backport of this pull request to the 3.14 branch. |
|
Sorry, @prithviraj-chaudhuri and @encukou, I could not cleanly backport this to |
This commit addresses this issue #140042 (comment).
It removes the sqlite3_shutdown call that could cause closing connections for sqlite when used with multiple sub interpreters.