pipwin
pipwin copied to clipboard
lepisma
[Urgent] Service will be discontinued before July 2022
On the website of Christoph Gohlke's it is written:
Funding for the Laboratory for Fluorescence Dynamics has ceased. This service will be discontinued before July 2022.
I guess this means, we are in trouble.
There is also a stackoverflow question about this: https://stackoverflow.com/questions/72581592/what-to-do-when-gohlkes-python-wheel-service-shuts-down
Maybe @cgohlke could provide some information.
Options I see:
- porting workflows to github, host wheels in a repo and build them with Github actions
- create a new website
Wow, this is urgent. I suppose these have always been unofficial binaries that people take for granted. Porting the wheels isn't a long term solution since it'll freeze the builds in time. The best is if packages start emphasizing the need for windows wheels on pip (e.g. like numpy provides).
By porting I didn't mean "copying". I really meant a workflow checking for new releases and building wheels with the toolchains of Mr. Gohlke on github actions.
But for the long run, it could be good to have a crash at some point to make developers aware to provide good wheels for their packages. Especially gdal and similar. Since there are tools like cibuildwheel, this is more easy than before.
Also conda is a great alternative on windows.
Follwing this discussion: https://www.reddit.com/r/Python/comments/vcaibq/christoph_gohlkes_windows_wheels_site_is_shutting/
There seems to be a copy: http://eturnbull.ca/pythonlibs/
@MuellerSeb So now both the original page and the copy you mentioned is down. Do you know of any other copies out there?
Unless there is an alternative external resource, it seems that it might be time to archive this repo.
The wheels are now provided via some github repositories as stated here: https://www.cgohlke.com/
On the bright side, if this package were to be rewritten, the GitHub REST API for Releases is pretty good and won't need to "mock out a little javascript environment within which to run Gohlke's obfuscation code".
I agree with @mwtoews that it's probably time to archive this repo.
While I am happy for someone else to take this forward in the way pointed above (GitHub releases), I believe it will be better to do that in a separate package.
I am archiving this tool.
