Closed sisp closed 1 month ago
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 97.65%. Comparing base (
8839010
) to head (c52036d
). Report is 1 commits behind head on master.
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
I agree, not sure why Poetry must diverge from the standard. :shrug:
They designed their UX before it was standardized. They wanted the standard to reuse their dependency format, as they (well, let's be direct, Sébastien Eustace, mainly) believed it was best for both beginners and experienced users. It was decided otherwise, mainly because PEP 508 and PEP 440 are a pretty established thing, and it wasn't a good thing to start diverging from these standards. I think Sébastien was disappointed at first and camped his grounds for quite some time. Anyway, maintainers recently merged a PR to support PEP 621 :slightly_smiling_face:
Round 3 of the discussion, with links to round 1 and 2: https://discuss.python.org/t/pep-621-round-3/5472.
I've added an alternative version specifier for
types-psutil
which uses PEP440 and not Poetry syntax. The constraint is equivalent, but it also works around a Renovate bug that prevents Poetry lock file updates for dependencies constrained with*
.Renovate's implementation of Poetry's versioning scheme uses transformations to SemVer and NPM, as Poetry draws much inspiration from there, but PEP440 is a superset of SemVer, so Renovate's Poetry support has some limitations.
See also https://github.com/copier-org/copier/pull/1829#issuecomment-2430078530.