py-matplotlib: add v3.6.0, new backends#32697
Conversation
| conditional("gtk4cairo", when="@3.5:"), | ||
| # Qt | ||
| conditional("qtagg", when="@3.5:"), | ||
| conditional("qtcairo", when="@3.5:"), |
There was a problem hiding this comment.
From what I can tell, the intention of having qtagg and qtcairo is to have a single backend for all Qt versions. The user then specifies the QT_API environment variable in order to specify which backend they want. See matplotlib/matplotlib#19255. However, we have separate packages for py-pyqt4/5/6 and py-pyside4/5/6, so there's no way to have a single dependency. Instead, I'm hardcoding to use py-pyqt6, but that's not really the intent of this backend.
|
@spackbot fix style |
|
Let me see if I can fix that for you! |
|
I was able to run spack style --fix==> Running style checks on spack
selected: isort, mypy, black, flake8
==> Modified files
var/spack/repos/builtin/packages/py-matplotlib/package.py
==> Running isort checks
isort checks were clean
==> Running mypy checks
Success: no issues found in 558 source files
mypy checks were clean
==> Running black checks
reformatted var/spack/repos/builtin/packages/py-matplotlib/package.py
All done! ✨ 🍰 ✨
1 file reformatted.
black checks were clean
==> Running flake8 checks
flake8 checks were clean
==> spack style checks were clean
I've updated the branch with isort fixes. |
|
@pradyunsg I'm seeing the following error when building: My best guess is that |
|
@scottwittenburg I'm trying to reproduce the CI build failure locally but it seems like the Python that gets pulled down from the binary cache is broken: $ /home/software/spack/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spa/linux-ubuntu18.04-x86_64/gcc-7.5.0/python-3.7.13-2rasf254brdwtgeudnsupharkvcceker/bin/python
qemu-x86_64: Could not open '/lib64/ld-linux-x86-64.so.2': No such file or directoryIs this a known issue? |
No idea. That's likely related to how matplotlib is being generated/built, and that's something I'm not particularly familiar with. I won't have more time to look into this until next week, since I'm currently swamped between talk prep and a couple of things I'd like to get into the next pip release. |
168ef0b to
a1c21b4
Compare
|
Reported the build issue upstream: matplotlib/matplotlib#23997 |
tldahlgren
left a comment
There was a problem hiding this comment.
LGTM. Confirmed both new sha256.
https://github.com/matplotlib/matplotlib/releases/tag/v3.6.0
These backends are starting to get pretty unwieldy. I'm open to suggestions for how to best support this. The problem is that each backend has multiple possible dependencies. Currently, we're hard-coding to use a specific dependency, but it would be great if we didn't have to do this. There's also been a lot of changes in backend support over the years, and I'm not sure which we actually want to support.