pmxbot IRC Log Viewer

[07:20:25] <gutsytechster> good morning everyone! :)

[07:26:02] <Adetomi> Good morning

[07:58:58] <gutsytechster> pradyunsg: I wonder, why do we have an empty scripttest.py. I see that we've imported the PipTestEnvironment within it. Couldn't we just define it in the scripttest.py?

[08:28:13] <travis-ci> pypa/pip#15231 (master - 69cc558 : Paul Moore): The build passed.

[08:28:13] <travis-ci> Change view : https://github.com/pypa/pip/compare/8c95de11ccb6...69cc55831d52

[08:28:13] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/666678220

[13:08:01] <sumanah> techalchemy: lemme know when you swing back into stuff for the day?

[14:34:30] <McSinyx[m]> pradyunsg, not specifically to you but to continue the (clumsy) discussions for the last few days, I've figured out the download architecture: https://gist.github.com/McSinyx/990a2d1ed39a3e0e884e116a2daf0ecb

[14:38:34] <McSinyx[m]> I don't think refactoring the calling stack would be too much of an issue (even with resolvelib _ methods which I assume can be touched freely), so the main concern is back to #825 with the UX

[14:40:35] <McSinyx[m]> however, I am confident that such task can be solved within 8 weeks, so I am asking if you are (or anyone here is) willing to mentor if I submit it as a GSoC project

[14:40:38] <McSinyx[m]> *the UX of displaying multiple downloads at the same time

[14:42:07] <McSinyx[m]> also to techalchemy, somehow my matrix client doesn't pull the complete irc log, thus I lost the few lines discussing the not-coupling from 2 days ago

[14:44:49] <gutsytechster> McSinyx[m]: This channel is logged. You may see the logs by typing !logs

[14:44:54] <gutsytechster> !logs

[14:44:54] <pmxbot> http://kafka.dcpython.org/channel/pypa-dev

[14:45:27] <McSinyx[m]> gutsytechster: that's why I know my client missed a few line d-:

[14:45:35] <McSinyx[m]> I trusted it

[14:45:48] <gutsytechster> Oh! :|

[14:46:40] <gutsytechster> devesh: o/

[14:46:54] <McSinyx[m]> thankfully no bigger misunderstanding was caused (-;

[14:49:53] <pradyunsg> gutsytechster: maybe? There might be a cyclic dependency on it, between scripttest.py and __init__.py, so that's why it might've been moved. It's certainly something worth cleaning up.

[14:58:44] <devesh> Hi @gutsytechster

[15:00:55] <McSinyx[m]> hi devesh, I don't think we need to use @ to mention others in IRC but I wonder if it works

[15:01:13] <McSinyx[m]> please try it on me

[15:01:28] <devesh> Hi McSinyx[m], I am new to IRC's

[15:01:39] <devesh> I thought using @ will give me autocomplete on the names

[15:01:46] <pradyunsg> devesh: gutsytechster went offline.

[15:02:08] <devesh> pradyunsg: got it

[15:02:34] <McSinyx[m]> pradyunsg: is my client again, it does not show anything, ugh, too modern

[15:03:28] <pradyunsg> McSinyx[m]: it might not be showing all those details? I'm pretty sure matrix has a whole bunch of configuration knobs.

[15:04:04] <McSinyx[m]> pradyunsg: yep is a client thing (nheko)

[15:04:48] <McSinyx[m]> I was using riot but electron is not very resource friendly to run 24/7

[15:23:46] <sumanah> techalchemy: I have an idea for when you're around

[15:31:49] <McSinyx[m]> pradyunsg: I've just updated the gist, I was incorrect to think the download got prepared and then triggered later

[16:27:49] <devesh> Hi all, is there a document somewhere on how AUTHORS.txt should be updated and with what details?

[16:28:00] <devesh> within the pip repo

[16:30:08] <devesh> or where can a guideline on how to update AUTHORS.txt be maintained, I was working on this issue: https://github.com/pypa/pip/issues/5979 which focuses on removing emails from AUTHORS.txt

[16:30:27] <devesh> And thought it will be a good idea to document it somewhere so that only names are used in future

[16:32:24] <pradyunsg> devesh: authors.txt is updated as part of our release process.

[16:32:26] <pradyunsg> https://github.com/pypa/pip/blob/master/noxfile.py#L167

[16:33:11] <devesh> Aah, so the function needs to be updated in that case

[16:34:01] <pradyunsg> I'm not 100% sure what the concensus is on that issue, but I personally prefer to have a list (without emails) in a documentation page thanking contributors, like Rust has. :)

[16:34:46] <pradyunsg> Yes, that function needs to be updated. And we need to figure out what to do with our license (which is the tricky part, because legal).

[16:35:06] <devesh> Got it, I actually saw your comment in this, https://github.com/pypa/pip/pull/6046 which states a new PR is needed there, so thought to contribute

[16:35:11] <pradyunsg> Since the license references the Authors.txt file.

[16:35:56] <pradyunsg> Oh wait, you're right. We can just remove the emails.

[16:36:20] <pradyunsg> I thought this was about removing AUTHORS.txt. 🙃

[16:36:33] <pradyunsg> devesh: ^

[16:37:06] <pradyunsg> McSinyx[m]: I've not taken a look at the Gist yet, but I expect to be able to later tonight.

[16:37:37] <devesh> pradyunsg: No it's only about removing emails of contributors which is public info and can be used for spamming

[16:38:03] <pradyunsg> Yup. That's a definite thing we should do. :)

[16:38:58] <sumanah> renewing my ping techalchemy :-)

[16:39:57] <devesh> But now when I look at https://github.com/pypa/pip/blob/master/tools/automation/release/__init__.py#L53 ,I see that the emails will still be included

[16:40:08] <devesh> when this function is called at time of a new release, do we want that?

[16:41:02] <pradyunsg> devesh: don't remove the .mailmap file though, since I know that there's at least 1 person for whom removing it would result in their dead name being listed (they changed their name).

[16:41:33] <pradyunsg> devesh: you can edit the --format option, to change what we extract from git.

[16:42:22] <devesh> yessir, using `git log --use-mailmap --format='%aN'` instead of the old one

[16:42:54] <devesh> https://git-scm.com/docs/pretty-formats#Documentation/pretty-formats.txt-aN

[16:42:55] <pradyunsg> devesh: the file is updated to include any new contributors' names in AUTHORS.txt prior to making a release. We do want that to keep doing this.

[16:43:02] <McSinyx[m]> pradyunsg: the gist only demonstrates the triviallity of starting downloads in parallel, while it'd be nice for me to have you confirm that, the main concern is with UX and avoiding downloading the same file

[16:43:10] <pradyunsg> devesh: sounds about right.

[16:43:34] <McSinyx[m]> pradyunsg: so please save your time today and I'll send the proposal with clearer disscussion tomorrow

[16:43:43] <McSinyx[m]> goodnight everyone!

[16:43:52] <devesh> pradyunsg: True, but I see that the file on git is updated by the regular contributors only, so if it is documented somewhere that from now on,

[16:44:14] <devesh> only names need to be added, there shouldn't be issues in future I think

[16:46:47] <devesh> Ohh sorry, I mistook your message pradyunsg. Will any of my suggested changes affect that functionality which you stated in your second last message to me

[16:49:50] <devesh> 1. Remove the emails from AUTHORS.txt 2. Get rid of email formatting message 3. Include somewhere that we only add names of contributors while manually updating AUTHORS.txt on github

[17:01:25] <techalchemy> sumanah, sorry, meetings etc -- in a team meeting atm too

[17:02:01] <sumanah> techalchemy: understood. What time should I ping you again? like 2pm?

[17:02:11] <techalchemy> sumanah, should be

[17:02:12] <sumanah> (I think you are on Eastern Time)

[17:02:13] <sumanah> ok

[18:18:56] <sumanah> hiya techalchemy

[18:56:12] <techalchemy> sumanah, apologies, my computer is massively overheating so i'm trying to get things to a state where i can reboot but it's taking me a bit, just wrapping up

[18:56:34] <sumanah> techalchemy: understood! best wishes

[18:56:41] <techalchemy> should be 5 mins

[19:20:58] <sumanah> techalchemy: cool

[19:25:52] <techalchemy> jeez, that was not 5 mins

[19:26:18] <techalchemy> no idea why things were so broken, hopefully turning it off and on again fixed it

[19:29:32] <sumanah> techalchemy: https://pad.sfconservancy.org/p/help-pipenv-release-2020-03-07

[19:34:22] <techalchemy> thanks sumanah looking ot over

[19:35:08] <techalchemy> pushing tags to requirementslib and reviewing pythonfinder again -- the only change i wanted to make to pythonfinder was to add some new api endpoints and deprecate some old ones, just because the naming is confusing

[19:37:00] <sumanah> techalchemy: right

[19:37:16] <sumanah> techalchemy: makes sense. is it necessary to do now? does it block pipenv?

[19:38:53] <techalchemy> sumanah, i don't think so, it's more for reducing developer confusion since the naming is so unclear

[19:39:08] <sumanah> techalchemy: ok. so should it wait till next month?

[19:42:19] <techalchemy> yeah i'm thinking so

[19:42:35] <techalchemy> i'm just going to release these both, just reading over the changes to see if i need to make minor releases or not

[19:42:51] <sumanah> techalchemy: totally. Good idea

[19:46:49] <tos9> now that PyCon is officially off -- has there been any discussion to have a packaging summit anyhow remotely?

[19:47:47] <sumanah> tos9: good question. toad_polo pradyunsg ^

[19:49:27] <toad_polo> No.

[19:51:02] <toad_polo> I think we'll make the announcement soon, but almost none of the planning or work we did transfers to a virtual summit.

[19:51:39] <toad_polo> And there's no reason to believe that for example the day and time is still good for anyone.

[19:52:18] <sumanah> Thanks for the candor toad_polo

[19:52:55] <toad_polo> To me, the main reason for doing the summit at PyCon was that basically everyone's going to be in the same place at the same time in the same time zone.

[19:53:41] <toad_polo> It might make sense to start setting up some per-issue group calls to try and resolve individual issues that are outstanding and can't or shouldn't wait for next year.

[19:54:00] <sumanah> I agree with that toad_polo

[19:54:27] <sumanah> the pip resolver team was talking last week about what we would like to do to replace the relationship-building and info-sharing we were going to do at PyCon https://wiki.python.org/psf/PackagingWG/2020-03-19-pip

[19:54:36] <techalchemy> i am 10000x better at focusing on calls than forum discussions so am in favor with that

[19:54:41] <sumanah> and we also started talking about getting some small group calls together

[19:55:02] <sumanah> (scroll down to "PyCon and COVID-19")

[19:58:31] <sumanah> toad_polo: I may ping you (or Pradyun may) to start developing some lists of people to corral into some calls

[19:58:56] <sumanah> techalchemy: I see you pushed a requirementslib release. Does there need to be a new pythonfinder release?

[19:58:56] <toad_polo> We'll probably just make the topic lists public.

[19:59:02] <techalchemy> sumanah, done already

[19:59:25] <toad_polo> The only reason they weren't in the first place was to get the voting started for all of them at the same time.

[19:59:44] <toad_polo> Might be best to self-organize around topics.

[19:59:48] <sumanah> right

[20:00:00] <sumanah> techalchemy: great! so could you look at https://pad.sfconservancy.org/p/help-pipenv-release-2020-03-07 and we'll see whether you could send it today?

[20:08:57] <sumanah> techalchemy: ok, looks like it's nearly done - check my latest edit to the 2nd footnote

[20:10:17] <techalchemy> +1 thank you

[20:21:54] <sumanah> hitting refresh in my mail client :-)

[20:26:08] <techalchemy> sumanah, i made some tweaks as i was getting ready to send

[20:26:22] <techalchemy> updated the doc if you want to reread otherwise i'll just send

[20:26:28] <sumanah> techalchemy: looking now

[20:28:32] <sumanah> techalchemy: fixed a number, added a link and some punctuation, taking 1 last look

[20:28:45] <techalchemy> saw those

[20:29:22] <sumanah> techalchemy: I say it's good, ship it!

[20:29:28] <techalchemy> nice

[20:29:32] <sumanah> If you want to say what version # it would be, go ahead

[20:29:50] <techalchemy> ugh thats a whole other thing i dont even want to worry about atm

[20:29:59] <sumanah> ok sure :-) no prob

[20:30:01] <techalchemy> pipenv semver uses the specific release date

[20:30:02] <techalchemy> lol

[20:30:14] <techalchemy> 2020.3.25 if i released today

[20:30:16] <sumanah> right

[20:30:32] <sumanah> techalchemy: ok, so in that case, yeah, ship the email, IMO

[20:30:42] <techalchemy> terrible decision that one

[20:31:45] <techalchemy> thanks for the help, sending now

[20:34:14] <sumanah> And it's up! https://groups.google.com/d/msgid/pypa-dev/7e77dccc-df48-4e54-1dd1-44caf21bde68%40danryan.co you're welcome techalchemy

[20:35:12] <sumanah> and at https://mail.python.org/archives/list/distutils-sig@python.org/thread/EY2KKSWVCYBVO2RB6OMD7TFOWO35B2WG/

[20:35:17] <sumanah> I will go around linking it now

[20:35:40] <sumanah> techalchemy: do you have time now to work on one of the next outstanding blockers? like making a vistir release?

[20:35:50] <techalchemy> does vistir need a release? hrm

[20:35:51] <techalchemy> let me see

[20:36:16] <sumanah> I put it in the list earlier this month just because it was one of the ones you mentioned ..... if you don't need a new release, awesome

[20:36:23] <techalchemy> if there are issues in vistir that could take ages to fix, that library is a huge pain

[20:39:29] <sumanah> ah

[20:40:03] <sumanah> techalchemy: so I presume that what you need to do is do the other merge -- the vendor updates -- in pipenv, and then check whether pipenv master passes tests ok given the current release of vistir?

[20:40:37] <techalchemy> there is a 0% chance of it passing, azure takes eternity too

[20:40:47] <techalchemy> i'll spin up a ramdisk on my own machine and run tests though

[20:41:10] <techalchemy> vistir seems ok

[20:51:04] <sumanah> techalchemy: ok - what do you think is a good next step, then?

[20:56:57] <techalchemy> sumanah, i really want to simplify releasing

[20:57:01] <sumanah> techalchemy: I know you do

[20:57:04] <sumanah> techalchemy: but

[20:57:10] <sumanah> techalchemy: you already know what I am gonna say right? :-)

[20:57:21] <techalchemy> well, i mean even current comaintainers dont know how to release

[20:57:31] <techalchemy> it's the actual reason there isn't another release yet

[20:57:36] <sumanah> so I think maybe some light estimating would be good.....

[20:57:51] <techalchemy> oh yeah that's a good idea

[20:57:59] <techalchemy> sumanah, have you ever used zenhub?

[20:58:00] <sumanah> how long would it take you to release using the current "process" and how long would it take you to do the automating?

[20:58:06] <sumanah> techalchemy: I don't think I have

[20:58:21] <techalchemy> oh i'm not going to automate before i release *NOW*

[20:58:45] <techalchemy> i mean, i think i'm doing this release by hand, i would have to do it by hand and also automate it at the same time

[20:58:50] <sumanah> techalchemy: ah, ok, cool..... so .... yeah I get it

[20:58:53] <techalchemy> it would like triple the amount of time it'd take

[20:59:08] <sumanah> is there any kind of release checklist? if not, it sounds like you will need to make one

[20:59:17] <sumanah> just taking notes as you do the current release

[21:00:26] <sumanah> I made a super long one for Twine once https://github.com/pypa/twine/pull/314 before jaraco streamlined it with a bunch of automation https://github.com/pypa/twine/commits/master/docs/contributing.rst

[21:01:04] <techalchemy> yeah there is no checklist, also this part has to be done each time where things are vendored and patched

[21:01:07] <techalchemy> it's basically a nightmare

[21:01:21] <sumanah> techalchemy: we can start a checklist now and at least it'll be something

[21:01:34] <techalchemy> i did start one

[21:01:38] <sumanah> oh cool, where?

[21:01:49] <sumanah> is it in ZenHub? maybe that's why you mentioned it?

[21:03:08] <techalchemy> no, i mentioned zenhub because it integrates with github but allows for some project management stuff like epics, plus has effort estimates

[21:03:44] <techalchemy> i think 1,2,3,5,8,13,20,40 are the allowable effort levels on an issue

[21:03:57] <sumanah> techalchemy: right, makes sense

[21:04:13] <sumanah> techalchemy: so where is the release checklist you started?

[21:06:18] <techalchemy> it's not exactly a checklist :p

[21:06:29] <sumanah> techalchemy: it's ok. what is it?

[21:06:47] <sumanah> even non-sequential scribbles are a start and we can go from there

[21:06:50] <techalchemy> markdown summary

[21:07:05] <techalchemy> i got to the part where you update vendored dependencies and it required lots of text

[21:07:11] <sumanah> right

[21:07:30] <sumanah> sometimes when I am writing down a procedure I get bogged down .... I end up going depth-first instead of breadth-first

[21:07:45] <sumanah> that's ok, we can just come back to it and give it another go. techalchemy may I see it? :-)

[21:08:25] <techalchemy> https://demo.codimd.org/wY3hW3f2SuGXBfd0ZsHsag?both

[21:08:37] <techalchemy> i was writing as i was doing

[21:12:19] <sumanah> cool

[21:12:41] <sumanah> ok so techalchemy how much more time would you like to work on pipenv today?

[21:24:03] <sumanah> techalchemy: it's ok if the answer is "zero" :-)

[21:24:17] <techalchemy> sorry i'm updating vendored dependencies as we speak

[21:24:18] <techalchemy> lol

[21:24:49] <sumanah> techalchemy: go you! cool

[21:25:00] <sumanah> that's great, no need for apologies :-)

[21:50:00] <sumanah> techalchemy: wanna stop for the day in 10 min?

[21:51:27] <techalchemy> sure, i managed to get this to actually re-vendor things and reapply patches

[21:51:31] <techalchemy> so it's almost a good stopping point

[21:51:44] <techalchemy> tests are slow so i'll probably kick them off shortly

[21:52:15] <techalchemy> but yeah i will revisit tomorrow

[21:56:40] <sumanah> techalchemy: ok!

[21:56:49] <sumanah> techalchemy: what's a good time for us to talk tomorrow?

[21:56:56] <sumanah> also I can do a call with you if that makes things easier

[21:57:42] <techalchemy> either way is ok with me, i have meetings in the morning scattered from 8 through 1230 so prob sometime in the afternoon

[21:58:18] <sumanah> techalchemy: ok, will ping you around 2pm ET

[21:58:20] <sumanah> have a good night

[21:58:25] <techalchemy> you too, thanks again!

[22:24:28] <travis-ci> pypa/twine#1343 (master - f7402e0 : Brian Rutledge): The build passed.

[22:24:28] <travis-ci> Change view : https://github.com/pypa/twine/compare/5482d6877df0cd90545c4708e5abe13b359f8a08...f7402e0d2e1c1e6ecde61dc8bcba515b807a48a0

[22:24:28] <travis-ci> Build details : https://travis-ci.org/pypa/twine/builds/667011203

[22:26:50] <travis-ci> pypa/pip#15243 (master - 69cc558 : Paul Moore): The build passed.

[22:26:50] <travis-ci> Change view : https://github.com/pypa/pip/compare/4d0a55ee6f7d7458a4f8e5688782f3a907c76057...69cc55831d529aaec13e0fb42c763ed103edd457

[22:26:50] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/667003137

[22:36:10] <travis-ci> pypa/packaging#873 (fix/282 - c5591ce : Dustin Ingram): The build passed.

[22:36:10] <travis-ci> Change view : https://github.com/pypa/packaging/compare/45ee2ab0295a^...c5591ce0a025

[22:36:10] <travis-ci> Build details : https://travis-ci.org/pypa/packaging/builds/667014723

Log file Viewer

Help | Karma | Search:

#pypa-dev logs for Wednesday the 25th of March, 2020