Log file Viewer

[06:25:27] <pmxbot> jaraco pushed 5 commits to setuptools (https://bitbucket.org/pypa/setuptools/) :

[06:25:27] <pmxbot> Add tests capturing expected behavior, including failure to match expectation indicated in docstring.

[06:25:28] <pmxbot> Update docstring and test to match long-standing expectation in behavior.

[06:25:28] <pmxbot> Merge

[06:25:28] <pmxbot> Restore old behavior for calculating the base. Fixes failing test and fixes #425.

[06:25:28] <pmxbot> Another refactor of findall, this time separating the simple walk / join operation from the conditional relative path.

[07:20:12] <pmxbot> jaraco pushed 3 commits to setuptools (https://bitbucket.org/pypa/setuptools/) :

[07:20:12] <pmxbot> Update changelog

[07:20:12] <pmxbot> Bumped to 18.3.1 in preparation for next release.

[07:20:12] <pmxbot> Added tag 18.3.1 for changeset 7a705b610abb

[19:49:25] <ronny> dstufft: ping? btw, is it on anyones radar to remove the version string in the dist-info directory name?

[19:49:35] <dstufft> hi

[19:49:49] <dstufft> It's not on my radar, dunno about anyone elses

[19:50:28] <ronny> hmk

[19:51:09] <ronny> dstufft: its a clunky artifact of the wheel ormat that makes it necessary to *search* for the dist-info folder, even tho there are no multi-version installs

[19:51:37] <dstufft> sure

[19:51:58] <dstufft> dunno if we'll ever end up supporting MVI or not

[19:52:10] <dstufft> it's just not a high enough priority to land on my radar atm

[19:55:00] <ronny> dstufft: for mvi its actually more sane to just install a single wheel with no deps ino a target dir named after the versioned filename

[19:55:14] <dstufft> maybe

[19:55:18] <dstufft> I've given it near zero thought

[19:56:01] <ronny> dstufft: i have a rough plan for cache-able mvi on todo that would include distribution support and sane de-bundleing

[19:57:07] <ronny> problem is i work fulltime on something else ^^

[22:06:52] <xafer> hello, any comment on https://github.com/pypa/pip/pull/3080 ?

[22:08:59] <dstufft> xafer: seems fine, maybe called it cannonical_name or something?

[22:10:00] <xafer> that was its initial name, but we have a lot of canonical_name variable already, cf https://github.com/pypa/pip/pull/3080/files#diff-1c5c2a05a3e5ec92eed891d087d4b8b8L176

[22:12:05] <xafer> any suggestion for the variable names ? :-/

[22:13:03] <dstufft> hm

[22:13:05] <dstufft> maybe make it a verb

[22:13:15] <xafer> canonicalize ? :p

[22:13:24] <dstufft> canonicalize_name() ?

[22:13:42] <dstufft> I'm just throwing stuff out there

[22:13:53] <xafer> it seems to exist O_o https://en.wiktionary.org/wiki/canonicalize

[22:13:55] <dstufft> I think it's fine as is too, but it might be clearer with a slightly diff name

[22:14:57] <xafer> canonicalize_name seems fine

[22:15:56] <xafer> should I add tests for this function ? (certainly directly stolen from setuptools)

[22:16:16] <xafer> I didnt think it was worth it

[22:17:24] <dstufft> xafer: it's probably OK not to

[22:20:54] <xafer> k, while you're here any news on PyPI search ?

[22:21:19] <dstufft> yea everything is terrible and I hate computers :)

[22:21:37] <dstufft> probably I'm going to just write Warehouse's search tomorrow

[22:21:41] <dstufft> and switch PyPI over to use that

[22:21:55] <xafer> lol ok :)