Rebuild for python312

[regro-cf-autotick-bot]

This PR has been triggered in an effort to update python312.

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.

---

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 @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. Feel free to drop us a line if there are any issues! This PR was generated by https://github.com/regro/cf-scripts/actions/runs/6336145624, please use this URL for debugging.}

[conda-forge-webservices]

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.

[hmaarrfk]

@h-vetinari can you summarize the findings of:
numpy/numpy#24660

why is it that I'm still seeing this issue?

[hmaarrfk]

@conda-forge-admin please restart cis

[h-vetinari]

can you summarize the findings of: numpy/numpy#24660

It was a meson config issue, but unrelated to 3.12

why is it that I'm still seeing this issue?

I don't think it's the same issue. That issue is about broken tests, you have a missing symbol:

import: 'numpy_allocator'
Traceback (most recent call last):
  File "/home/conda/feedstock_root/build_artifacts/numpy-allocator_1696291268092/test_tmp/run_test.py", line 2, in <module>
    import numpy_allocator
ImportError: $PREFIX/lib/python3.12/site-packages/numpy_allocator.cpython-312-x86_64-linux-gnu.so: undefined symbol: PyDataMem_SetHandler

[hmaarrfk]

Sorry i thought i saw PyDataMem_SetHandler in that issue so i got excited. I'll dig a little more locally.

[eliaskoromilas]

It looks like there is an issue when numpy>=1.25.0 during build:

# works
docker run --entrypoint=sh --interactive --rm --tty python:3.11 -c "git clone https://github.com/inaccel/numpy-allocator && sed -i 's/1.22.0/1.24.4/' numpy-allocator/pyproject.toml && sed -i 's/>=1.22.0/==1.24.4/' numpy-allocator/setup.py && pip install ./numpy-allocator && python -c 'import numpy_allocator'"

# doesn't work
docker run --entrypoint=sh --interactive --rm --tty python:3.11 -c "git clone https://github.com/inaccel/numpy-allocator && sed -i 's/1.22.0/1.25.0/' numpy-allocator/pyproject.toml && sed -i 's/>=1.22.0/==1.25.0/' numpy-allocator/setup.py && pip install ./numpy-allocator && python -c 'import numpy_allocator'"
# ImportError: /usr/local/lib/python3.11/site-packages/numpy_allocator.cpython-311-x86_64-linux-gnu.so: undefined symbol: PyDataMem_SetHandler

[eliaskoromilas]

Currently, I set numpy==1.22.0 in the build system (which is the min with custom allocator support), but is incompatible with Python 3.12.

[hmaarrfk]

@conda-forge-admin please rerender

[hmaarrfk]

I think all the pins were "correct", but lets try again.

[eliaskoromilas]

Now every build will fail!

[eliaskoromilas]

Oh, it looks like it sticks to numpy<=1.23 in python<=3.11 builds, so they won't fail.

[hmaarrfk]

condaforge tries to build with the oldest "supported" simply across the ecosystem.

it has recently been increased to at least 1.22 across the board