PMXBOT Log file Viewer

Help | Karma | Search:

#pypa-dev logs for Monday the 19th of January, 2015

(Back to #pypa-dev overview) (Back to channel listing) (Animate logs)
[01:13:23] <pmxbot> jaraco pushed 7 commits to setuptools (https://bitbucket.org/pypa/setuptools/) :
[01:13:23] <pmxbot> Extract messages as class attributes for nicer indentation.
[01:13:23] <pmxbot> Extract variable for bdist_wininst command
[01:13:23] <pmxbot> Extract variable for exec_param
[01:13:23] <pmxbot> Defer resolution of the CommandSpec and do it exactly once.
[01:13:23] <pmxbot> Extract writer resolution as a variable
[01:13:23] <pmxbot> Allow the CommandSpec class to be resolved by the writer.
[01:13:23] <pmxbot> Adding test capturing failure where sys.executable loses backslashes on Windows. Ref #331.
[01:29:59] <pmxbot> jaraco pushed 3 commits to setuptools (https://bitbucket.org/pypa/setuptools/) :
[01:29:59] <pmxbot> Use the command spec as resolved by the best ScriptWriter.
[01:29:59] <pmxbot> Also use command_spec_class in ScriptWriter.
[01:29:59] <pmxbot> Use non-posix semantics in a WindowsCommandSpec and use that class in the Windows script writers. Fixes #331.
[01:37:27] <pmxbot> jaraco pushed 3 commits to setuptools (https://bitbucket.org/pypa/setuptools/) :
[01:37:27] <pmxbot> Update changelog
[01:37:27] <pmxbot> Added tag 12.0.2 for changeset a177ea34bf81
[01:37:27] <pmxbot> Bumped to 12.0.3 in preparation for next release.
[01:51:04] <pmxbot> jaraco pushed 4 commits to setuptools (https://bitbucket.org/pypa/setuptools/) :
[01:51:04] <pmxbot> Correct command reference.
[01:51:04] <pmxbot> Update changelog
[01:51:04] <pmxbot> Added tag 12.0.3 for changeset bf8c5bcacd49
[01:51:04] <pmxbot> Bumped to 12.0.4 in preparation for next release.
[17:08:09] <ionelmc> msabramo: have you seen dstufft "readme" project?
[17:08:30] <msabramo> ionelmc: yes and I've sent a few PRs as well
[17:08:39] <msabramo> it's a good idea
[17:08:41] <ionelmc> msabramo: couldn't that be used on pypi?
[17:08:54] <msabramo> I think dstufft had the same idea
[17:09:00] <ionelmc> the old pypi, not warehouse
[17:09:04] <msabramo> I remember working on it even
[17:09:25] <ionelmc> what's holding its use?
[17:09:28] <msabramo> https://bitbucket.org/pypa/pypi/pull-request/60/fix-bb-214-make-rst-rendering-not-fail-for/diff#comment-4374437
[17:10:28] <msabramo> Hmmm I thought I send a PR to make PyPI use readme
[17:10:35] <msabramo> maybe I never finished it
[17:12:55] <sigmavirus24> msabramo: I think the plan was for dstufft to see how to make it work with current PyPI because it might not work perfectly
[17:13:33] <dstufft> it shouldn't be hard to do, I just haven't had time. I'm switching teams at work so I'm tying up loose ends heh
[17:14:44] <msabramo> it looks like I have some unchecked-in code sitting on my machine that was an attempt at making pypi use readme
[17:14:49] <msabramo> I suspect that it doesn't work
[17:16:55] <msabramo> let me see if there's anything useful in there
[17:28:48] <msabramo> IIRC, the blocker was that there were no tests so hard to know that I'm not breaking current behavior
[17:29:13] <ionelmc> test in production
[17:29:22] <ionelmc> :p
[17:29:25] <msabramo> that was probably what made me open https://bitbucket.org/pypa/pypi/pull-request/58/add-tests-test_description_utilspy/diff
[17:31:25] <msabramo> I think I might've also made some changes to "readme" to make it accommodate current PyPI behavior
[17:31:45] <ionelmc> dstufft: it's more efficient to boss people around to do what you don't have time for :-)
[17:32:53] <msabramo> ah yes, my changes to make pypa/pypi use pypa/readme required https://github.com/pypa/readme/pull/15
[17:33:35] <dstufft> why did it require that
[17:33:51] <dstufft> is it just so it doesn't have the rel="nofollow"?
[17:33:57] <msabramo> yeah it looks like it
[17:34:03] <msabramo> which maybe is fine actually
[17:34:14] <dstufft> yea I'm fine with it getting rel="nofollow"
[17:34:23] <dstufft> readme needs to handle SVGs better though
[17:34:48] <msabramo> but my idea was to write tests for pypa/pypi that exactly characterized the current behavior, then make it use pypa/readme and the tests should still pass, meaning I didn't change anything functionally
[17:35:13] <msabramo> then we could take out the linkify stuff and update the tests to expect rel="nofollow"
[17:35:24] <dstufft> https://github.com/github/markup/blob/master/lib/github/commands/rest2html#L105-L125
[17:35:56] <msabramo> https://bitbucket.org/pypa/pypi/pull-request/58/add-tests-test_description_utilspy/diff is not expecting the rel="nofollow" to be there
[17:36:19] <msabramo> cuz it's just making assertions about the current PyPI behavior
[17:36:32] <msabramo> a baseline if you will
[17:36:33] <dstufft> current PyPI behavior is probably wrong in that regards
[17:36:57] <msabramo> might be wrong but it is what it is and I feel like I should use that as the starting point
[17:37:22] <msabramo> because my first priority is to not break PyPI or to not even change the behavior of PyPI at all inadvertently
[17:38:11] <msabramo> I guess I could send one PR that adds the "readme" stuff and updates the tests if that is preferable
[17:38:44] <msabramo> then it wouldn't rely on the pypa/readme change which is nice
[17:38:54] <dstufft> i'd rather not have readme be configurable at all, it should be super simple and do one thing and only one thing
[17:40:28] <msabramo> ok
[17:40:49] <msabramo> you should add a comment and close https://github.com/pypa/readme/pull/15 then
[18:02:14] <msabramo> dstufft, ionelmc: I just opened https://bitbucket.org/pypa/pypi/pull-request/64/use-pypa-readme-to-render-rest/diff
[19:08:36] <ionelmc> what the hell is this lib/python2.7/site-packages/matplotlib-1.3.1-py2.7-nspkg.pth ?!
[19:19:01] <sigmavirus24> ionelmc: looks like the file that creates a package namespace. Why are you surprised by this?
[19:19:11] <sigmavirus24> Also why so excited over something so normal?
[19:21:15] <ionelmc> sigmavirus24: it's reflexes, i made the mistake of looking inside the file :-)
[19:21:32] <sigmavirus24> ionelmc: keep calm
[19:21:34] <sigmavirus24> python on