Skip to content

Conversation

@manzt
Copy link
Collaborator

@manzt manzt commented Oct 13, 2025

Original patch in jcrist#852.

@ionelmc
Copy link

ionelmc commented Oct 15, 2025

@manzt I would like to suggest that renaming everything to msgspec_m seems counterproductive if the goal of this fork is to be a drop-in, bugfix-only project. Unless you want to do more than just bugfixes. If there's a choice to add from __future__ import annotations all over the place or replace all the imports in one's project I think people will find the former simpler.

@manzt
Copy link
Collaborator Author

manzt commented Oct 15, 2025

@ionelmc thanks for the suggestion. I've made #7 reverting the rename (and providing some rationale for why I chose to rename in the first place).

@ionelmc
Copy link

ionelmc commented Oct 15, 2025

I kinda understand where that is coming from - I guess you have a dependency that depends on msgspec, and your main project or other deps want to depend on msgspec-m. Sadly there's no pretty solution, and Obsoletes-Dist/Provides-Dist metadata is not supported in any tool (https://packaging.python.org/en/latest/specifications/core-metadata/#obsoletes-dist-multiple-use) but having msgspec and msgspec_m trying to interoperate in a single codebase ain't it (is it even possible? wouldn't all the utilities be broken).

The way I see it the path of least friction is to just get dependencies to also depend on msgspec-m.

@ofek
Copy link

ofek commented Oct 19, 2025

I became a co-maintainer and will try to merge everything that is required for supporting 3.14 in the next few days.

@manzt
Copy link
Collaborator Author

manzt commented Oct 19, 2025

fantastic news, thank you for shepherding @ofek

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants