You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
the grayskull website can be used to generate CRAN recipes as well
Presently the website asks for project name and version. Now that we support CRAN also, the website should first ask repository (CRAN or PyPI), then project name and version
The text was updated successfully, but these errors were encountered:
yeah, that needs some work, but it will not be my priority for now. I would rather focus on adding support for poetry and flint first, after that I can rework the webapi.
A semi-crazy direction to go would be getting the whole grayskull shooting match running in the browser under WASM,
and hosting on a simple static ReadTheDocs page or GitHub pages. This would shift much of the burden from a backend process to CDN-mirrored static assets.
Either pyodide or xeus-python and emscripten-forge could likely do the job.
The only missing dep from pyodide, at present is rapidfuzz, which appears to also be missing from emscripten-forge. Getting it into one of those channels probably wouldn't be a huge lift, but then maybe having rapidfuzz be hidden behind a conditional dependency, and having a fallback pure-python implementation would get it sorted faster.
At present, pyodide's solver already relies on the PyPI JSON API, so no worries there, but as the CRAN integration is scraping-based, it would likely cause more issues when actually running.
End Result
project name
andversion
. Now that we support CRAN also, the website should first askrepository
(CRAN or PyPI), thenproject name
andversion
The text was updated successfully, but these errors were encountered: