PMXBOT Log file Viewer

Help | Karma | Search:

#pypa logs for Tuesday the 26th of April, 2016

(Back to #pypa overview) (Back to channel listing) (Animate logs)
[03:39:58] <givemefive> hi is there a dh-virtualenv analog for rpm?
[10:26:34] <ionelmc> givemefive: not that i know of, it's not that hard to replicate
[10:27:03] <ionelmc> dh-virtualenv is a bit too complicated anyway
[10:27:30] <ionelmc> in the sense that deb packaging helpers are generally complex, lotsa magic
[11:23:51] <domenkozar> where's the email/reasoning behind the new pypi json api?
[11:25:24] <FRidh> domenkozar: the json api did not change. The url scheme changed
[11:25:54] <domenkozar> is there some documented reasoning for it?
[11:26:30] <FRidh> https://bitbucket.org/pypa/pypi/issues/438/backwards-compatible-un-hashed-package#comment-27230605
[11:30:43] <domenkozar> aha pypi.io has the mapping
[11:31:10] <dstufft> if someone wants to add it to legacy PyPI I'll review and merge, but I have very little interest in mucking around there :/
[11:31:21] <domenkozar> that's fine
[11:31:27] <domenkozar> we'll use the new api on nixos
[11:31:32] <domenkozar> and report issues if found
[11:32:43] <domenkozar> the alternative is https://github.com/triton/triton/commit/ffcbbf7211310ac3c3d69e52bf11ef2cba630612
[11:32:48] <domenkozar> but I really don't want to do this
[11:33:00] <dstufft> domenkozar: just make sure you follow redirects! (Not sure how nixos does things, but I know by default curl doesn't follow redirects)
[11:33:44] <domenkozar> sure, we do follow them because sourceforge
[11:35:09] <dstufft> meh, for some reason Fastly isn't caching the redirect
[11:35:13] <dstufft> weill have to figure out why
[11:38:28] <domenkozar> dstufft: quick q, did we get enough willpower start implementing static metadata alternative to setup.py?
[11:38:45] <domenkozar> I'm not following closely the packaging story lately
[11:43:49] <dstufft> domenkozar: Not specifically no
[11:45:36] <domenkozar> one day we'll have https://github.com/commercialhaskell/all-cabal-hashes/tree/hackage
[11:45:59] <domenkozar> but I'll probably have children before that day :)
[13:21:08] <givemefive> ionelmc ok thanks, I think I can figure something out similar, just wanted to make sure I wasn't replicating a tool
[13:22:19] <ionelmc> givemefive: last time i asked in #rhel the idea wasn't popular, and that's an understatement
[13:22:37] <givemefive> ionelmc haha, what was the reason
[13:22:45] <ionelmc> they were actually pretty offended that i didn't want to repackage every dependency lol
[13:23:00] <ionelmc> perish the thought
[13:23:22] <givemefive> well from a epel / fedora repo perspective I understand, but to deploy some code i wrote, NO lol
[13:25:56] <givemefive> although if this new pyp2rpm tool starts being reliable
[13:26:01] <givemefive> maybe not such a big deal
[13:58:51] <gchristensen> dstufft: may I PM?
[13:59:02] <dstufft> sure
[15:27:48] <asmacdo> !logs
[15:27:48] <pmxbot> http://chat-logs.dcpython.org/channel/pypa
[15:39:57] <asmacdo> fyi, ^ is a dead link