Closed lilyminium closed 3 weeks ago
Hello @lilyminium! Thanks for updating this PR. We checked the lines you've touched for PEPÂ 8 issues, and found:
There are currently no PEP 8 issues detected in this Pull Request. Cheers! :beers:
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 93.15%. Comparing base (
d7153b9
) to head (c64627e
). Report is 3 commits behind head on develop.
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
@lilyminium this PR is marked as draft - is it meant to be up for review or is there more work that needs to be done?
Failures aren't happening in other PRs - it looks like something that has been specifically introduced here.
Hmm. Yes that would be the case that multi-bonds can't be sketchily represented by digging into Topology internals, which downstream people may well be doing.
I can think of two solutions to this in this PR:
order
attribute (which TopologyObjects should have) -- a bit of work, may have time later today to push it through but not guaranteed. Also, even more changes in current behaviourPushed a commit that removes the "addition" fixes and preserves previous behaviour (should be easy to revert if discussion goes the other way)
I'm leaning towards the simpler fix of just ensuring that all identical bonds are deleted when deletion is caused, as it's less disruptive to current behaviour, and potentially fixing the addition of identical bonds in a separate issue/PR later, so I've updated the changelog too to reflect that -- should still be easy to revert if there's any disagreement
Going with just the bond deletion fix seems like a reasonable thing for this release.
Fixes #4762
Changes made in this Pull Request:
Changes Topology creation and addition to only accept a single identical bond/angle/dihedral/etc instead of multiple(see discussion in PR)PR Checklist
Developers certificate of origin
📚 Documentation preview 📚: https://mdanalysis--4763.org.readthedocs.build/en/4763/