Remove setuptools constraint

conda-forge / numba-feedstock

A conda-smithy repository for numba.

BSD 3-Clause "New" or "Revised" License

0 stars 27 forks source link

Remove setuptools constraint #106

Closed gmarkall closed 2 years ago

gmarkall commented 2 years ago

Based on and closes #105 with the setuptools constraint removed from setup.py in an attempt to make pip check succeed. I expect a further patch to some tests may be required.

Closes #104 too.

Checklist

[X] Used a personal fork of the feedstock to propose changes
[X] Bumped the build number (if the version is unchanged)
[ ] Reset the build number to 0 (if the version changed)
[ ] Re-rendered with the latest conda-smithy (Use the phrase code>@<space/conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
[X] Ensured the license file is being packaged.

conda-forge-linter commented 2 years ago

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

henryiii commented 2 years ago

Why would tests need patching? Everything should be working again after the rollback in setuptools 65.0.2 or 65.1. I could be wrong, of course, but I'd expect it to work.

gmarkall commented 2 years ago

Why would tests need patching? Everything should be working again after the rollback in setuptools 65.0.2 or 65.1. I could be wrong, of course, but I'd expect it to work.

There are some tests that ensure that Numba import doesn't produce any warnings. I was concerned that the deprecation warning that suggests using setuptools < 60 might be produced by NumPy and cause this test to fail. Though, looking at the set of tests that run, we don't have a test with NumPy 1.23 in the list of configurations, so maybe we won't run into this issue :crossed_fingers:

henryiii commented 2 years ago

Shouldn't we have a test with NumPy 1.23? :D

(not in this PR, but sounds like something that needs to be added then)

gmarkall commented 2 years ago

Yes, we should test with NumPy 1.23 - I was thinking of following up with that later once the dust has settled here.

jakirkham commented 2 years ago

cc @xhochy

gmarkall commented 2 years ago

Seems that the PPC builds are a bit close to the time limit and some of them timed out for this PR - do we just re-run tests when this happens?

jakirkham commented 2 years ago

Yep have restarted those jobs

jakirkham commented 2 years ago

Thanks all! 🙏

Just a note, we are still seeing ppc64le builds hanging on main ( https://github.com/conda-forge/numba-feedstock/issues/102 ).