pmxbot IRC Log Viewer

[00:53:28] <pradyunsg> 4 now...

[01:05:58] <techalchemy> yeah i just installed pytest-timeout so uh

[01:06:04] <techalchemy> pushed that up

[01:06:35] <techalchemy> still trying to get a windows vm

[01:06:45] <pradyunsg> techalchemy: run it once w/ --collect-only? To see that it's able to collect tests? Maybe 'pipenv run' is failing?

[01:07:18] <techalchemy> it was before, if it doesn't collect tests it tells you no tests ran

[01:07:58] <pradyunsg> yea, but given that there's no output... it'd help to know that succeeds.

[01:16:15] <mbacchi> !logs

[01:16:15] <pmxbot> http://kafka.dcpython.org/channel/pypa-dev

[01:21:49] <techalchemy> i seriously have no idea

[01:21:57] <techalchemy> im going to find something to put windows on i guess

[01:22:04] <pradyunsg> hehe

[01:22:11] <pradyunsg> I'mma sleep now.

[01:33:55] <mbacchi> I'm trying to get back into working on https://github.com/pypa/warehouse/pull/4038, and after merging master into my branch, and cleaning up conflicts, I see the following error: `RuntimeError: Cannot use localizer without has_translations=True`. Since I'm basing my `credits.html` page on `base.html` it tries localizing anytime I load the page. How

[01:33:56] <mbacchi> can I work around this in my dev env?

[05:15:45] <devesh> Hi folks, I was looking at the list of PRs in pip, and found some of them were approved but never merged.

[05:15:54] <devesh> e.g. https://github.com/pypa/pip/pull/7780

[05:16:20] <devesh> Is it fine to comment on such PRs and intimate the maintainers so that take a look again at it and possibly merge it?

[09:17:05] <cjc7373> Hey guys, I was applying for GSoC pip project, I tried to email to pradyunsg+gsoc@gmail.com, but didn't get any reply. Any suggestions? Even we've met the application deadline, should I keep working on the proposal?

[10:49:52] <travis-ci> pypa/pip#15411 (master - ac26f1b : Paul Moore): The build passed.

[10:49:52] <travis-ci> Change view : https://github.com/pypa/pip/compare/4b159c5420b0...ac26f1bc3339

[10:49:52] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/670073620

[10:51:19] <travis-ci> pypa/pip#15412 (master - 2a1f3c2 : Paul Moore): The build passed.

[10:51:19] <travis-ci> Change view : https://github.com/pypa/pip/compare/ac26f1bc3339...2a1f3c22a626

[10:51:19] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/670073949

[11:35:10] <sumanah> cjc7373: hi

[11:35:19] <sumanah> I saw your question via http://kafka.dcpython.org/day/pypa-dev/2020-04-02

[11:35:30] <sumanah> pradyunsg might be at dinner right now

[11:36:01] <sumanah> I'll let him answer you

[11:36:16] <sumanah> just wanted to let you know people were not deliberately ignoring you

[11:44:40] <cjc7373> sumanah thanks! I know the community is always helpful :)

[11:49:52] <sumanah> :-)

[14:09:42] <travis-ci> pypa/pip#15434 (master - c88fa39 : Paul Moore): The build passed.

[14:09:42] <travis-ci> Change view : https://github.com/pypa/pip/compare/f6457b7bb22a...c88fa395a369

[14:09:42] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/670150771

[16:44:06] <tos9> bleh new virtualenv broke my virtualenv hack

[16:44:32] <tos9> guess that'll teach me :/

[16:56:49] <gutsytechster> reading logs is so much insightful sometimes. :)

[17:06:21] <travis-ci> pypa/packaging#885 (master - a9f31ee : Dustin Ingram): The build passed.

[17:06:21] <travis-ci> Change view : https://github.com/pypa/packaging/compare/ed0a58f0dacf...a9f31ee35b6d

[17:06:21] <travis-ci> Build details : https://travis-ci.org/pypa/packaging/builds/670234529

[17:17:18] <sumanah> techalchemy: I'm in the middle of a few things but I am here to talk to in case you are stuck

[18:34:49] <devesh> I was wondering what will adding a new warning message will fall on in terms of NEWS file extension, will it be a bugfix?

[18:34:57] <devesh> issue in question: https://github.com/pypa/pip/issues/7430

[18:35:40] <pradyunsg> .bugfix or .feature -- both work IMO.

[18:36:22] <devesh> ok

[18:36:42] <pradyunsg> cjc7373: hi o/

[18:37:38] <devesh> btw, I was wondering if the bugs at Print Better Error Messages milestone qualify as "good first issues"

[18:38:21] <pradyunsg> devesh: not necessarily. Quite a few of them are fairly involved fixes, that require some amount of understanding of the codebase.

[18:39:03] <devesh> True, btw are there good videos, maybe in a pycon which go into pip's architecture

[18:39:23] <devesh> I do see that there's a architecture doc which is starting to form in pip.pypa.io

[18:39:37] <sumanah> hi devesh - so you can go to pyvideo.org and look around for videos about pip

[18:39:56] <devesh> wow, there's a pyvideo.org

[18:39:57] <sumanah> I don't think there are any that deep dive into pip's architecture though

[18:40:38] <sumanah> devesh: there is a danger that videos are, of course, outdated

[18:41:21] <devesh> So the codebase has drastically changed in the past few years?

[18:41:45] <pradyunsg> devesh: yep.

[18:42:06] <devesh> ohh

[18:43:26] <devesh> The way I understand, there is "_internal" with all the core logic, and "_vendor" for third party dependencies

[18:43:48] <sumanah> hey devesh have you already read the architecture documentation?

[18:44:20] <devesh> hi sumanah: I haven't but I was going to :)

[18:44:44] <devesh> sorry I have a habit of going into tangents sometime

[18:44:49] <sumanah> ok devesh -- so, just in case you were about to ask a question, it's really good to read docs BEFORE you ask questions that are probably answered in the docs.

[18:45:13] <devesh> yep, I understood that when you asked me :)

[18:45:16] <sumanah> https://www.harihareswara.net/sumana/2016/10/12/0 We have to make a balance between person-to-person discussion and documentation that everyone can read, so we save time answering common questions but also get everyone the personal help they need.

[18:45:32] <sumanah> ok great, glad we are on the same page!

[18:45:57] <devesh> That page I think you already gave me, that's a great starter doc

[18:46:23] <sumanah> ok great :-)

[18:46:26] <pradyunsg> cjc7373: I saw your email, however it wasn't possible to provide feedback on your proposal so close to the deadline. We'll be reviewing your proposal as submitted on the GSoC portal.

[18:46:52] <sumanah> back in ~10-20 min

[19:01:42] <devesh> btw, I'll probably re-ask the question I asked in the morning

[19:01:58] <devesh> I was looking at the list of PRs in pip, and found some of them were approved but never merged. e.g. github.com/pypa/pip/pull/7780 . Is it fine to comment on such PRs and intimate the maintainers so that take a look again at it and possibly merge it?

[19:02:50] <sumanah> here's a question: what would your goal be?\

[19:02:57] <sumanah> in doing that?

[19:03:51] <devesh> I was thinking that apart from looking at pr's and submitting fixes, it might be a good idea to get some of the muck out

[19:04:17] <sumanah> sorry - could you explain that a bit more?

[19:04:18] <devesh> issues which are no longer reproducible since they are fixed, pr's which are lying around approved, but were missed to be merged

[19:04:44] <sumanah> ok -- so -- help pip reduce backlog

[19:04:53] <devesh> I have actually looked at some of the old issues, found out they were fixed and informed the folks

[19:04:57] <sumanah> great!

[19:05:05] <devesh> sumanah: yes reducing backlog

[19:05:26] <sumanah> so your idea is: maintainers may have missed that these PRs were ready to be merged, and so you would systematically go through and leave a comment on ones that were already approved and seem merge-ready

[19:05:30] <devesh> I think that folks here are already burdened with so much, why not help them in some way

[19:05:36] <sumanah> devesh: totally a great motivation

[19:06:10] <sumanah> I suggest you try and think of it from the point of view of one of the pip maintainers. They get a lot of notifications. What would help them KNOW that those PRs are still mergeable?

[19:06:30] <devesh> I haven't thought of systematically going through, but maybe a few every day

[19:07:08] <sumanah> or: could you perhaps make a special GitHub PRs search, using search syntax, for open PRs that are approved, and see how many there are?

[19:07:08] <devesh> sumanah: Maybe if they are approved by someone, or maybe I try to pull in the patch, try it out and verify that it indeed fixes things

[19:07:09] <sumanah> and see whether there are any patterns among them?

[19:07:12] <sumanah> YES

[19:07:17] <sumanah> try it out, test it, verify it

[19:07:40] <sumanah> a "I have tested this" comment is so much more valuable than only saying "I noticed this is not merged yet"

[19:08:22] <devesh> True, someone approving it a few months back doesn't break down to it possibly being merged

[19:08:27] <devesh> unless someone verifies it

[19:08:43] <sumanah> devesh: if you are going to do some manual testing, you might enjoy skimming https://thebraidytester.com/downloads/YouAreNotDoneYet.pdf

[19:09:02] <sumanah> which is a list of different kinds of scenarios to test when you are trying to check whether software is buggy

[19:09:18] <devesh> skimming 32 pages :) will do

[19:09:31] <devesh> btw is that endeavour of mine a good idea?

[19:09:53] <sumanah> devesh: I don't know.

[19:10:08] <sumanah> devesh: I think that if you are willing to go through PRs and ..... back in 4 min

[19:10:21] <devesh> okay we will carry on then

[19:11:19] <sumanah> devesh: ok, back sooner than I thought. If you are willing to go through approved-but-not-merged PRs and locally test them, then report "I tested this and it still works" etc., that is valuable and a good diea

[19:11:20] <sumanah> idea

[19:11:53] <sumanah> devesh: but also: I do suggest you run that search and find out how many PRs are in that state.

[19:12:32] <sumanah> You said "folks here are already burdened with so much, why not help them in some way" -- there are many valuable things you can do in triage, writing documentation, writing blog posts, answering questions on Stack Overflow, and so on. You're not lacking in valuable things to do.

[19:13:12] <devesh> Stack overflow is something I have done, but have reduced lately

[19:13:21] <devesh> I have a gold badge in python :)

[19:13:48] <sumanah> it would be great for you to keep on answering questions specifically about pip -- and keep watching that area because when we start letting people test the new resolver in several weeks, lots of people will have new questions

[19:13:55] <devesh> But yes, you are right sumanah, Open Source is much more than writing code and committing pr's

[19:15:08] <sumanah> yup. In fact there is, in my opinion, an oversupply of people who only want to write code, compared to an undersupply of the people who have interest and skill in the other stuff.... and then the people who make decisions that decide _what code actually gets deployed and used_ are the people who can write, etc.

[19:15:31] <sumanah> who can write in English, who are in touch with what users need via customer support, who write grant proposals and business plans, etc.

[19:15:44] <sumanah> (I'm speaking of open source specifically)

[19:17:00] <sumanah> devesh: I need to go pay attention to prioritizing some issues now. hope this has been helpful! Talk with you again soon

[19:17:16] <devesh> True, as an aside, I saw 4 PRs which are in "approved but not merged state", and I also understand that approvals from people without write access don't count

[19:17:23] <devesh> No worries sumanah, talk to you later

[19:17:27] <sumanah> :-)

[19:17:46] <sumanah> (don't "count" -- they're useful! they are useful but not sufficient.)

[19:20:25] <devesh> true that :)

[19:37:11] <travis-ci> pypa/pip#15447 (master - 87aaae0 : Paul Moore): The build passed.

[19:37:11] <travis-ci> Change view : https://github.com/pypa/pip/compare/6086f71cde81...87aaae0e0848

[19:37:11] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/670282116

[20:29:11] <travis-ci> pypa/packaging#887 (master - 9731c00 : Devesh Kumar Singh): The build passed.

[20:29:11] <travis-ci> Change view : https://github.com/pypa/packaging/compare/a9f31ee35b6d...9731c00837b8

[20:29:11] <travis-ci> Build details : https://travis-ci.org/pypa/packaging/builds/670304138

[20:51:40] <devesh> Aaand github is down

[22:38:22] <travis-ci> pypa/pip#15448 (master - 87aaae0 : Paul Moore): The build passed.

[22:38:22] <travis-ci> Change view : https://github.com/pypa/pip/compare/7511737ee79275c429bda210a1041448d2e59db4...87aaae0e0848cbef0edbc9cc7bcef9594f1a3a14

[22:38:22] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/670333835

[22:40:55] <travis-ci> pypa/twine#1351 (master - f7402e0 : Brian Rutledge): The build passed.

[22:40:55] <travis-ci> Change view : https://github.com/pypa/twine/compare/5482d6877df0cd90545c4708e5abe13b359f8a08...f7402e0d2e1c1e6ecde61dc8bcba515b807a48a0

[22:40:55] <travis-ci> Build details : https://travis-ci.org/pypa/twine/builds/670340587

Log file Viewer

Help | Karma | Search:

#pypa-dev logs for Thursday the 2nd of April, 2020