Rebuild for python310

[regro-cf-autotick-bot]

This PR has been triggered in an effort to update python310.

Notes and instructions for merging this PR:

1. Please merge the PR only after the tests have passed.
2. Feel free to push to the bot's branch to update this PR if needed.

Please note that if you close this PR we presume that the feedstock has been rebuilt, so if you are going to perform the rebuild yourself don't close this PR until the your rebuild has been merged.

This package has the following downstream children:

And potentially more.

If this PR was opened in error or needs to be updated please add the bot-rerun label to this PR. The bot will close this PR and schedule another one. If you do not have permissions to add this label, you can use the phrase code>@<space/conda-forge-admin, please rerun bot in a PR comment to have the conda-forge-admin add it for you.

_{This PR was created by the regro-cf-autotick-bot. The regro-cf-autotick-bot is a service to automatically track the dependency graph, migrate packages, and propose package version updates for conda-forge. If you would like a local version of this bot, you might consider using rever. Rever is a tool for automating software releases and forms the backbone of the bot's conda-forge PRing capability. Rever is both conda (conda install -c conda-forge rever) and pip (pip install re-ver) installable. Finally, feel free to drop us a line if there are any issues! This PR was generated by https://github.com/regro/autotick-bot/actions/runs/1517808805, please use this URL for debugging}

[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.

[erykoff]

Any idea why the hough transform code is failing on win_64 pypy3.7?

[hmaarrfk]

honestly, no.

i kinda feel like fixing cpython 3.10 and merging since the pypy build already exists...

[erykoff]

Well other then that windows strangeness, the other problem is that the tests on the new cross compiled ppc64le are stalling out somehow. So the options are to either a) turn off tests for the cross-ppc64le or b) reverting to the native build on travis or c) try to debug those tests.

[hmaarrfk]

@conda-forge-admin please rerender

[hmaarrfk]

I missed the cross compilation. I disabled it for now. I don't want this recipe to be a trailblazer, but maybe we can ask for help in a bit. There used to be some strangeness before Azure was used to build windows. We can likely try to remove this in a future cleanup.

[hmaarrfk]

It might be that tests on QEMU are just flaky.

[hmaarrfk]

I tried cleaning up the recipe. So hopefully this will work. I deleted my own warning so :/

[erykoff]

Good news is that the travis build for ppc64le worked, and that removing those extra build steps and the scary warning seemed to make no difference. Bad news is that the windows pypy37 build is still failing some tests...

[github-actions[bot]]

Hi! This is the friendly conda-forge automerge bot!

I considered the following status checks when analyzing this PR:

• linter: passed
• travis: passed
• azure: failed

Thus the PR was not passing and not merged.

[hmaarrfk]

my guess is that it is related to the use of PyMem_Malloc but I'm not 100% sure.

@conda-forge/help-pypy are you able to help guess at any updates in cython or pypy that may have broken things here? scikit-image uses cython quite a bit in these tests

[erykoff]

Interesting, I didn't know cython was involved (and therefore I'm surprised that 3.10 is working). I ran into this problem recently that the pypi download of packages includes cython-generated .c files. And if these exist the setup script doesn't rerun cython, and thus you can get incompatibilities. The solution in that case (and might work here!) is to force remove any pre-cythonized .c files in the build script which forces the correct version of cython to run at build time.

See https://github.com/conda-forge/pytables-feedstock/pull/74/commits/e2037635d883790713ed59c151f3e1c785a73fd6 and https://github.com/conda-forge/pytables-feedstock/pull/74/commits/20e582263c22f961c81aaec137293437da403ed0 (but note that in this case be careful only to remove the cythonized files). Though in this case I don't know where the windows build script is coming from...

[jakirkham]

I believe PyMem_Malloc is defined for PyPy. For example see this commit ( https://github.com/cython/cython/commit/8657bc27cbf160689725d1a2e9de6833b4501098 )

[hmaarrfk]

I guess it is defined, i'm just trying to find something unique about that piece of code that is related to pypy :/

[github-actions[bot]]

Hi! This is the friendly conda-forge automerge bot!

Commits were made to this PR after the automerge label was added. For security reasons, I have disabled automerge by removing the automerge label. Please add the automerge label again (or ask a maintainer to do so) if you'd like to enable automerge again!

[hmaarrfk]

@jakirkham since we already have windows + pypy builds, do you think it would be ok to debug this after we merge?

[jakirkham]

since we already have windows + pypy builds, do you think it would be ok to debug this after we merge?

Sure. Would suggest raising an issue to track. This might be something that needs more investigation for the next scikit-image release

[mattip]

It might be that tests on QEMU are just flaky.

This rings true with the experiences from other feedstocks, like scipy-feedstock

Bad news is that the windows pypy37 build is still failing some tests...

Please don't delay the merge for PyPy + windows. It should first pass on the regular CI before trying to debug it here.

[hmaarrfk]

Ok we might need to restart CIs a little to get this to pass. We could otherwise skip tests on azure + cross compile, though that seems risky.