Drop python 3.6 and 3.7 which are past end of life upstream

[Pierre-Sassoulas]

Cherry-picked from #8, I'm doing things progressively so it's reviewable.

Closes #11

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (48212e2) 94.16% compared to head (f6e9b9e) 93.97%.

Additional details and impacted files

@@ Coverage Diff @@ ## master #23 +/- ## ========================================== - Coverage 94.16% 93.97% -0.19%  ========================================== Files 18 18 Lines 548 548 Branches 106 106 ========================================== - Hits 516 515 -1  Misses 23 23 - Partials 9 10 +1 

Flag	Coverage Δ
3.10	93.97% <ø> (ø)
3.11	93.97% <ø> (ø)
3.6	?
3.7	?
3.8	93.97% <ø> (ø)
3.9	93.97% <ø> (ø)
macos-latest	93.97% <ø> (ø)
ubuntu-20.04	?
ubuntu-latest	93.97% <ø> (ø)
windows-latest	93.97% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[Pierre-Sassoulas]

@stdedos I think it could be merged even with the negative coverage change.

[Pierre-Sassoulas]

@stdedos could you review this, please :) ? The CI is red because of an indirect coverage changes that I think we can ignore.

[stdedos]

@stdedos could you review this, please :) ? The CI is red because of an indirect coverage changes that I think we can ignore.

Yeah, this seems to be the problem:

https://app.codecov.io/gh/pylint-dev/pylint-pytest/pull/23/indirect-changes

I'd be nice to figure out why is this happening, but it is definitely not a priority - given the benefits of this PR.

"I am not reviewing this" since it's not scheduled to go in yet.
Otherwise all looks golden to me and it is approved 💚

---

Plan is:

• New bugfix on Monday 13.11
• Merge 🐛 Ignore collection failures in non-tests #15
• Merge whatever else minor comes in (like Windows Artifacts have incorrect Slugification #25, A collection of minor improvements to tests #26)
• New bugfix on Monday 20.11
• Questions
• Start https://github.com/pylint-dev/pylint-pytest/milestone/2
• New b/rc asap

Question:

1. Can Add python 3.12 support in CI and metadata #24 become thinner? I could consider it for v1
2. Can Support pylint v3 #12 be fixed for v1? I lost track on what was the pylint issue 😅

[Pierre-Sassoulas]

Can #24 become thinner? I could consider it for v1

We can remove the drop of python 3.6 / 3.7 (probably)

Can #12 be fixed for v1? I lost track on what was the pylint issue 😅

Need to support 3.12 maybe something else

[stdedos]

Can #24 become thinner? I could consider it for v1

We can remove the drop of python 3.6 / 3.7 (probably)

I'd consider it for v1 - but it doesn't have to wait "the usual" 1w-period.

So next week we release master, and a thin #24 afterwards

[Pierre-Sassoulas]

Hey, I tried dropping the removal of python 3.6 / 3.7 and simply adding python 3.12, but this is non trivial (https://github.com/pylint-dev/pylint-pytest/actions/runs/7010519218/job/19071209667?pr=24). It's been nearly two months since the release of python 3.12 / pylint 3.0 and python 3.6 / 3.7 have been EOL for a long time, maybe we could just drop EOL interpreters and release 2.X serie, what do you think ?

[Pierre-Sassoulas]

See this issue in vscode/pylint : microsoft/vscode-pylint#460 for example

[stdedos]

Thank you for trying ❤️

Sadly, my OSS time has been near-zero these days.
I will make it my first priority to merge these.

[Pierre-Sassoulas]

No problem. This one is ready the coverage drop is acceptable imo.

[stdedos]

Must be merged after #27, if it becomes ready for a Monday release

[stdedos]

I took the liberty to prepare this for tomorrow 🙃

[Pierre-Sassoulas]

Thank you @stdedos :)

[stdedos]

Thank you @Pierre-Sassoulas for all your contributions and patience 😁 🙏