PMXBOT Log file Viewer

Help | Karma | Search:

#pypa-dev logs for Friday the 24th of April, 2020

(Back to #pypa-dev overview) (Back to channel listing) (Animate logs)
[15:48:32] <toad_polo> Have we discussed removing the option to unpublish things now that yank exists?
[17:02:03] <sumanah> toad_polo: I don't think we have but I'm unfamiliar with the "unpublish" option -- this is something on PyPI?
[17:03:06] <toad_polo> sumanah: I mean, you can remove files.
[17:04:17] <toad_polo> It breaks anyone relying on that package / version.
[17:04:56] <sumanah> toad_polo: ahhh ok. I understand, thanks. So -- just as a description of current capability -- right now on PyPI I see that, for a release on a project I own, I can (a) yank the release, (b) delete the release, or (c) delete one of the files associated with the release
[17:05:22] <sumanah> and you're talking about (b) and (c) right?
[17:06:54] <toad_polo> Yeah. At least b, probably c as well.
[17:09:14] <sumanah> so in the UI (screenshot in https://twitter.com/di_codes/status/1253166902437871616 - I think you have seen this but including the link for others here) as di_codes says:
[17:09:16] <sumanah> "In the maintainer's UI, when managing a release, you're presented with an option to yank a release (which is reversible) vs. deleting it (which is permanent).
[17:09:16] <sumanah> Yanking is strongly encouraged over deleting!"
[17:10:06] <sumanah> as I understand it, deleting as an option is available in case it's not just "this is deprecated" but "yow this is actively so bad that we would prefer to even break pins that ask for it"
[17:11:32] <sumanah> toad_polo: sorry, this is a longwinded way of saying that I don't know of any _fresh_ discussions on the point you raise, but there were probably discussions if this in the Warehouse issues about yanking
[17:13:55] <toad_polo> I'll throw an issue out there this weekend, just wanted to check if anyone knew offhand that it's a hard no.
[17:15:03] <toad_polo> sumanah: By the way, a few days ago you wanted to have a 5 minute chat, is that still relevant? I'm in a meeting now until 13:30 but I can do a quick VC after if you want.
[17:15:14] <sumanah> toad_polo: it's not but thank you!
[17:15:56] <toad_polo> Sorry I wasn't available at that time.
[17:25:26] <sumanah> It all worked out :-)
[18:53:08] <dhellmann> prometheanfire : o/
[18:53:17] <prometheanfire> dhellmann: o/
[18:53:45] <prometheanfire> so, is this resolver for testing the generation of lower-constraints
[18:53:46] <dhellmann> prometheanfire : the person I wanted to introduce you to is Paul Moore, but I don't see a nick here that's obviously him
[18:53:52] <dhellmann> well, sort of
[18:54:04] <dhellmann> the lower constraints thing may eventually be a new feature of the new resolver
[18:54:28] <dhellmann> https://github.com/pypa/pip/issues/8085 is the initial attempt at implementing that
[18:54:38] <dhellmann> the testing is for the new resolver more generally
[18:55:23] <prometheanfire> that's still a good thing :D
[18:55:40] <dhellmann> yes, especially since openstack presents a nicely complicated test case :-)
[18:57:57] <dhellmann> I think based on something I saw that Paul mostly works on pip Mon-Wed, so it might be next week before he's here in channel.
[19:03:14] <prometheanfire> once there's something we can test with (we'll need to get tox to install something special I imagine) then I'm down for it
[19:03:37] <dhellmann> the job would need to install pip from source, I think, unless there's a pre-release version we could use
[19:03:50] <dhellmann> this might even be something to just do as a 1-off
[22:41:42] <travis-ci> pypa/twine#1408 (master - 2c30b1e : Ian Stapleton Cordasco): The build passed.
[22:41:42] <travis-ci> Change view : https://github.com/pypa/twine/compare/d7377572a1ae93846dbc809b65b70665f88412db...2c30b1e9c7936a6edd2e54c44852b58461fbc771
[22:41:42] <travis-ci> Build details : https://travis-ci.org/pypa/twine/builds/679239930
[23:31:53] <travis-ci> pypa/pip#15882 (master - 6a7bf94 : Pradyun Gedam): The build passed.
[23:31:53] <travis-ci> Change view : https://github.com/pypa/pip/compare/daff81124ab7...6a7bf9477681
[23:31:53] <travis-ci> Build details : https://travis-ci.org/pypa/pip/builds/679249065