ignore another big auto format

[mikemhenry]

Tips

• Comment "pre-commit.ci autofix" to have pre-commit.ci atomically format your PR.
  Since this will create a commit, it is best to make this comment when you are finished with your work.

Checklist

• Added a news entry

Developers certificate of origin

• I certify that this contribution is covered by the MIT License here and the Developer Certificate of Origin at https://developercertificate.org/.

[mikemhenry]

Okay this also tests that the PR template is now working! We can access the release template with a quary URL param, I don't know where we want to put a link to that, maybe on our docs or wiki? This can be future work.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.67%. Comparing base (3cea95b) to head (f07a9b1).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #437   +/-   ##
=======================================
  Coverage   98.67%   98.67%           
=======================================
  Files          36       36           
  Lines        2117     2117           
=======================================
  Hits         2089     2089           
  Misses         28       28           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mikemhenry]

Also I turned on "only squash commits" via the web UI since we talked about doing that but it never happened.

@ianmkenney can you check to see if when you run pre-commit locally now it doesn't try and change anything?

[ianmkenney]

    cd `mktemp -d`
    git clone https://github.com/openfreeenergy/gufe && cd gufe
    micromamba create -p ../env/ python=3.11 pre-commit --yes
    micromamba -p ../env/ run pre-commit run --all-files

shows that no files were reformatted. Do we know why it took multiple calls of the bot to fix things? That sounds like a bug.

[mikemhenry]

Do we know why it took multiple calls of the bot to fix things? That sounds like a bug.

I looked into it and looking at the logs (below) it had to do with an interaction where pyupgrade made a change (converted a .format into a f-string) that black didn't like (different line wrapping style), so black had to run again on those changes. For the last run nothing changed.

I will check to see if pyupgrade has any config options to play nicer with black.

[dotsdl]

@mikemhenry in discussion with @ianmkenney, can we just drop pyupgrade if it's fighting with black? We want to completely avoid nondeterminism in this system.

[mikemhenry]

Sure I can update this PR with that change. Just a few notes:

• Things are deterministic, even if the hooks were done in a different order, the results would be the same.
• pyupgrade I think would be worth running when we bump the oldest python we support to a newer version, but that can be done manually in a separate PR as needed.

[dotsdl]

@mikemhenry works for me! We also have agreement from @atravitz, @jthorton, and @ianmkenney!

[mikemhenry]

Okay this should be good to go now!

[ianmkenney]

lgtm, thanks @mikemhenry!

[jthorton]

LGTM!