Rebuild python39 0 2 he4191d

[chrisburr]

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.

Closes #54

[conda-forge-linter]

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]

Why is there so much extra stuff in the diff, like changes to the travis file? Also, currently, pybind11 segfaults (rarely on Linux, more often on macOS) with python 3.9.0 (working on solving this soon, PR being made to CPython, hoping for 3.9.1).

Next version in a few days should be PyPy and Python 3.9.x friendly.

[chrisburr]

Why is there so much extra stuff in the diff, like changes to the travis file?

That comes from @conda-forge-admin, please rerender or conda smithy rerender. Everything outside of recipe/ (and conda-forge.yml) is automatically generated and is an implementation detail of how conda-forge works.

[henryiii]

No, I mean in the gh2146.diff file that you added that has changes to pybind11's Travis file, which doesn't affect conda-forge at all. It should go away in the near future anyway, but I noticed all the odd changes.

[chrisburr]

Ah apologies! It's just the contents of pybind/pybind11#2146. It contains a little more than strictly necessary but I guess it was easier to grab with https://github.com/pybind/pybind11/pull/2146.patch when it was originally applied in #46.

(I don't know any details about the patch itself, it was already in master when the Python 3.9 migration PR was created.)

[tdegeus]

I'm experiencing the segmentation fault on macOS as well : conda-forge/python-goosefem-feedstock#7 . I'll guess I'll just have to wait for the new version and retry ?

[henryiii]

Assuming that's Python 3.9.0? That will be fixed in Python 3.9.1. The 2.6.0 version of pybind11 will probably also try to trade the segfault for a leak if you are using exactly 3.9.0, see pybind/pybind11#2576.