Log file Viewer

[00:22:48] <fellowlinguist> greets :)

[00:30:10] <dstufft> hello!

[04:34:55] <Ivo> qwcode: best thing Ive found was 0install when it was python

[07:45:18] <kr> hi

[07:45:35] <kr> I'd like to ask about the status of twine

[07:46:21] <kr> and - if it would be possible to extract the upload functionality into a library

[11:13:05] <dstufft> kr: the status is I haven't had a lot of time to work on it :(

[12:31:10] <Ivo> dstufft: you're asking now if we should go back on our instance that pip complain about using insecure transports?

[12:32:23] <dstufft> Ivo: yea, I added that, and we've had a number of complaints about it warning people about insecure transport when they are on a secure network

[12:32:48] <dstufft> the most obvious of those is localhost (in all it's various incantations), but there's also "I'm only on the LAN so a http url is fine"

[12:33:19] <Ivo> dstufft: I still think a good solution is an option to specify DNs pip won't complain about

[12:33:59] <Ivo> so for instance populate it with localhost by default

[12:34:20] <Ivo> ppl can populate it with others if they want o use http on others

[12:34:48] <dstufft> yea that's not a terrible idea, and what i'd do if we kept that

[12:34:59] <dstufft> I was just wondering if the warning really is doing much at all

[12:40:08] <Ivo> dstufft: read http://blogs.computerworld.com/privacy/23698/google-customers-you-will-use-https again, do you really want to become that guy

[12:41:19] <Ivo> roight, brb off into windows land

[12:59:11] <kr> dstufft: I'll do a cleanup/refactor on twine/upload.py if you do not mind

[13:00:08] <dstufft> kr: Not at all! Especially if you write tests at the same time :) (I wrote twine over a few hours on a weekend to solve a problem and never got around to writing any :/) but fwiw make sure it supports 2.6+ and 3.2+

[13:05:00] <kr> do you need to support 2.6 nor 3.2? I can not check those, only 2.7 and 3.3

[13:05:34] <kr> nor can I check it on windows & mac

[13:05:47] <dstufft> I can check it on 2.6 and 3.2 if you make a PR

[17:00:10] <dstufft> I'm goig to cut a 1.5.5 release with some fixes in it, https://github.com/pypa/pip/issues?milestone=9&state=open is what i'm planning, if anyone else has something they want add it to the milestone

[17:14:46] <qwcode> dstufft, 1.6?

[17:15:04] <dstufft> qwcode: no, just 1.5.5

[17:15:17] <dstufft> fix a few bugs that are affecting people

[17:16:15] <qwcode> dstufft, I was just casually considering doing 1.6. except for dropping dependency links, 1.6 has small things so far

[17:17:28] <dstufft> qwcode: we could backport some stuff but i think we should wait to actually release a 1.6

[17:18:19] <qwcode> dstufft, it's just come up before whether a ~3 month minor cycle might work, vs ~ 6months

[17:19:51] <qwcode> the argument for 6months I guess is that our releases can cause so much grief unfortunately as it's a bedrock tool

[17:20:04] <dstufft> qwcode: ah, well I dunno.

[17:20:10] <dstufft> hopefully we don't cause much grief anymore

[17:20:17] <dstufft> but there are still people trying to catch up to 1.5

[17:21:11] <qwcode> so why are you thinking to wait to release 1.6?

[17:22:08] <dstufft> I dunno, just felt like there wasn't enough stuff to justify a 1.6

[17:22:30] <dstufft> I'm not particularly against it majorly tbh, I just didn't think there was much stuff done to make one

[17:22:50] <qwcode> oh, because the minor increment traditionally means more features for pip

[17:23:12] <qwcode> yea, it's just 1.5.X is a pain

[17:24:47] <dstufft> i'm perfectly happy to do a 1.6 instead

[17:26:50] <qwcode> I don't know for sure. just thinking out loud. maybe a pypa-dev discussion. @jezdez may have a solid opinion on this.

[17:30:04] <dstufft> qwcode: made an email thing

[18:37:19] <dstufft> pfmoore: ping

[22:12:35] <pmxbot> jaraco pushed 1 commit to setuptools (https://bitbucket.org/pypa/setuptools/) :

[22:12:35] <pmxbot> Apply patch based on patch in 2008 by Klaus Zimmerman. Fixes #176.

[22:14:26] <pmxbot> jaraco pushed 2 commits to setuptools (https://bitbucket.org/pypa/setuptools/) :

[22:14:26] <pmxbot> Re-ran egg_info

[22:14:26] <pmxbot> Removed notice about PEP 420 package finder. Discussion in the ticket (Ref #97) indicates that the implementation is currently inadequate.