=== Moot2 is now known as MootBot [03:00] hello everybody and welcome to this weeks asiapac reviewers meeting [03:00] who's here today? [03:00] hi! [03:00] hi! [03:00] wow, i am _actually here_! [03:01] :) [03:01] also, I really need to make my IRC ident like yours barry [03:01] hi [03:01] jml: use your time machine and sign up at freenode 12 years ago :) [03:01] hmm, mine used to be like barry's [03:02] barry: I'm dead certain you weren't working at Canonical 12 years ago. [03:02] barry: and it was called openprojects back then anyway :) [03:03] oh that! ask kiko i think :) [03:03] anyway... [03:03] == Agenda == [03:03] * Roll call [03:03] * Next meeting [03:03] * Action items [03:03] * Queue status [03:03] * Mentoring update [03:03] * Review process [03:03] * Help people learn how big branches can be split up (BjornT) [03:03] * (Julian) Since I seem to be finding it hard to get my Soyuz comrades to follow our own informal coding standards, when reviewing Soyuz code please make sure you don't let code of the form: ` if archive.purpose == ArchivePurpose.PPA:` land, instead it should be the simpler: `if archive.is_ppa:` which not only encapsulates the decision in IArchive, it should remove an import of the DBEnum. [03:03] * Next meeting [03:03] week += 1? [03:03] +1 [03:04] yeha [03:04] I mean, r=jml [03:04] * thumper ment yeah [03:04] :) [03:04] not cowboy mode [03:04] sure [03:04] thumper: that would be yee haw [03:04] cool [03:04] * Action items [03:05] actually, i forgot to remove thumper's ai, so there really is nothing [03:05] hooray [03:05] * Queue status [03:05] 3 on PR [03:06] * jml looks [03:06] 6 pinks (tho we won't count stubs branch) [03:06] jml: your branch got rejected? [03:07] * mwhudson tries to remember if jml asked him to look at the stacking puller [03:07] barry: because the bzr branch it depends on is unfetchable. [03:07] jml: ! [03:07] barry: I need to fix it up. [03:07] I've had other things to do, I'm afraid. [03:07] ah well [03:08] * barry didn't get to nearly as many branches as he'd hoped today [03:08] it's not an immediate concern because we aren't going to land it until the bazaar changes we need are in trunk. [03:08] the review process doesn't have a clear place for this sort of thing. [03:08] jml: not a WIP? [03:09] barry: the Launchpad part of the code is done. [03:09] jml: no biggie if the branch'll be ready soon. if not, just remove it from PR and add it back when you're ready [03:10] ok. [03:10] anything else on the queue? [03:11] I'm not sure I understand the question [03:11] * thumper doesn't look very often any more [03:11] er, sorry. any other queue related comments? [03:11] no :) [03:12] i'll skip mentoring [03:12] good good. [03:12] i don't think i have anything else to say about bjorn's item, except that i think he was going to take it to the ml [03:12] * (Julian) Since I seem to be finding it hard to get my Soyuz comrades to follow our own informal coding standards, when reviewing Soyuz code please make sure you don't let code of the form: ` if archive.purpose == ArchivePurpose.PPA:` land, instead it should be the simpler: `if archive.is_ppa:` which not only encapsulates the decision in IArchive, it should remove an import of the DBEnum. [03:13] bigjools was going to start a page for product-specific coding guidelines [03:13] hmm. [03:13] e.g. what should soyuz devs and reviewers look for [03:13] etc. [03:13] hah [03:13] I'm thinking of a bzrlp specific guidelines [03:13] * use unit tests [03:13] :-) [03:14] barry: I'm not sure that this is such a good idea. [03:14] jml: why not? [03:14] thumper: as opposed to doctests you mean? [03:14] * are they really unit-y? no? go back and try again. [03:14] mwhudson: yeah [03:14] barry: a couple of reasons [03:14] 'read xunit test patterns' [03:15] barry: the example that Julian gives could probably be enforced in code, rather than by review, by making 'purpose' a private attribute [03:15] or protected or however zope spells it. [03:16] barry: also, it feels kind of a roundabout way of getting the soyuz team clear on their own abstractions. [03:17] jml: it's also to get reviewers who aren't as familiar with the internal conventions, to know what to look for [03:17] barry: I'm not strongly opposed to the idea, but... [03:18] jml: i think what you're saying is that there should be one obvious way to do it, even if you're not dutch [03:18] barry: yeah, that's kind of what I'm saying. [03:18] it's a good point [03:18] barry: in fact, that's very much what I'm saying :) [03:19] * barry chants the zop [03:20] barry: I think there's a stronger case for having review guidelines split by service types rather than by team. [03:20] jml: i'll communicate that on to the ameus [03:20] jml: 'service types' ? [03:20] jml: what do you mean? [03:20] barry: webapp, package builder, codehosting etc. [03:20] because they are actually quite different areas and hard to navigate if you aren't familiar [03:21] but then the guidelines wouldn't be so much "Use the provided interface" as brief tours [03:21] hard too to remember if you're not deep in it every day [03:21] so maybe forget I said anything. [03:22] or rather, I think it's a good idea, but it's heading off topic [03:22] well, i agree that it would be good to have better roadmaps for functional areas. this would especially help new devs [03:22] the sort of thing that fits on to one A3 page. [03:23] yeah [03:23] anyway, that's all i have for today. anything more from y'all? [03:23] * jml thinks [03:24] barry: nope. [03:24] alrighty then, have a great week. g'night :) [03:25] g'night bazza [03:25] barry: g'night. [03:25] mwhudson: heh. === mwhudson__ is now known as mwhudson === cprov is now known as cprov-lunch === cprov-lunch is now known as cprov === salgado-afk is now known as salgado === mrevell is now known as mrevell-lunch === mrevell-lunch is now known as mrevell === salgado is now known as salgado-lunch === cprov is now known as cprov-out === salgado-lunch is now known as salgado [21:19] yay! [21:19] Mootbot works! === salgado is now known as salgado-afk === mwhudson__ is now known as mwhudson