Skip to content

use pixi, add types, python 3.9-3.14#162

Merged
tonyfast merged 125 commits intodeathbeds:mainfrom
bollwyvl:p13xi
Feb 26, 2026
Merged

use pixi, add types, python 3.9-3.14#162
tonyfast merged 125 commits intodeathbeds:mainfrom
bollwyvl:p13xi

Conversation

@bollwyvl
Copy link
Contributor

@bollwyvl bollwyvl commented Apr 4, 2025

references

changes

  • use pixi
    • add pixi.toml
    • check with mypy
    • add test_repo.py to validate deps, versions, etc. between (pyproject|pixi).toml, (.readthedocs|test).yml
  • update packaging metadata
    • add trove classifiers
  • update ci
    • test
    • release
  • src
    • add py.typed
    • update types to match upstreams where possible
    • update CLI parsing with -- to allow using flags "owned" by importnb (-h)
    • utils
      • add initial importnb.utils.coverage plugin
      • require explicitly enabling pytest plugin with -p=importnb.utils.pytest_importnb
  • update docs
    • README.md
      • note coverage and pytest integration techniques
      • fix (and test) basic loader info
    • add CHANGELOG.md
    • build and host conda packages along with docs
    • add jupyterlite
      • use a custom pyodide-lock.json
      • add interesting demo examples
        • pytest and coverage
        • sphinx and autodoc
        • ipywidgets
        • ???
      • test examples for restart-and-run-all in browser
    • use mkdocstrings effectively

@bollwyvl bollwyvl changed the title use pixi, add types, python 3.9-3.13 use pixi, add types, python 3.9-3.14 Oct 9, 2025
@bollwyvl bollwyvl closed this Oct 29, 2025
@bollwyvl bollwyvl reopened this Oct 29, 2025
@tonyfast
Copy link
Member

lets merge this an work with it. this way we don't get bit rot again.

this is amazing. hopefully, we can sort out the async and mypy gear soon.

@tonyfast tonyfast merged commit 9f3aff6 into deathbeds:main Feb 26, 2026
27 checks passed
@bollwyvl
Copy link
Contributor Author

huzzah! 🍻

bit rot

welp, things can rot on main as well :P

async

i think we're good now; have been working with some downstreams against it, and haven't made anything explode. now... sounds like 3.15 might bring some new challenges. i'll see if there's an easy way to start kicking that particular tire... worst-case, we have uv gin one up against pypi deps with no pins. but certainly not blocking an a0 release of what's in the can.

mypy

yeah, well, we have #168... but not holding my breath. maybe there would be a way with the (much slower) py3-none-any.whl which doesn't use mypyc...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants