testing build for python 3.10 and 3.11

[williampiat3]

Remove condition for skipping builds based on Python version.

Checklist

• Used a personal fork of the feedstock to propose changes
• 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 @conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
• Ensured the license file is being packaged.

[conda-forge-admin]

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/recipe.yaml) and found it was in an excellent condition.

[williampiat3]

@conda-forge-admin, please rerender

[williampiat3]

@akhmerov, just testing if a python 3.10 and 3.11 build is possible for python-mumps 0.0.6 as I integrated it in this PR:
conda-forge/muscat-split-feedstock#63

[akhmerov]

I try to follow spec-0, which recommends dropping py<3.12. Furthermore, 3.10 and 3.11 don't receive bugfixes, only security fixes. Do you have a an explicit need to support these versions?

[williampiat3]

I try to follow spec-0, which recommends dropping py<3.12. Furthermore, 3.10 and 3.11 don't receive bugfixes, only security fixes. Do you have a an explicit need to support these versions?

We don't have an explicit need, we currently support python >=3.10 with an internal mumps wrapper but dropping support for 3.10 and 3.11 is not off the table.

[akhmerov]

Let's live in the happy spec-0 world 🥳

[williampiat3]

So it appears to work like this. We will discuss with me colleagues the next course of actions I see 3 options for now:

• Having a python-mumps 0.0.6 for python 3.10 and it would only be possible if you agree to it
• Dropping our support for python 3.10 (3.11 seems to work with a previous version of python-mumps) to comply with spec-0
• Dropping python-mumps dependency (after all, the wrapper I made in our package is actually to keep the preexisting functionnality of MUMPS but python-mumps could be used directly without this thin wrapper)

I'll keep you posted, I will push for option 2 and 3 so this PR might not be needed

Thank you for your quick replies! as always, much apreciated!

[akhmerov]

I'm open to supporting older versions, I think it comes with actual costs. At the same time, in particular in conda world, the upgrade cycle is faster than in e.g. linux distros. So if backwards compat is a concern, it should likely go to the main package.

[fbordeu]

I understand the Idea of spec-0 but I have a different interpretation. The release Pipelines of conda forge are easy to test. So why don't we keep "old" python (this python are supported) version as long as they pass the tests? And drop the support if they fail.

[akhmerov]

I am fine with this approach. To confirm: if there's a new release and older Python versions break, I will drop them until a fix is available.

@williampiat3 thank you for the contribution, merging.

[williampiat3]

Thank you @akhmerov, much apreciated!