[03:21] I <3 X-Launchpad-Subscription === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [13:12] hi danilos [13:12] bac, hi, how's it going? [13:12] i worked on bug 770245 yesterday but didn't really make much progress. what were your thoughts on how to solve the problem? [13:12] <_mup_> Bug #770245: Unsubscribing a team from a dupe doesn't work in advanced subscription overlay < https://launchpad.net/bugs/770245 > [13:16] bac, well, I was thinking of exporting on +subscribe form if a team-subscription is from a dupe somehow, and then using unsubscribeFromDupes webservice call; if that doesn't work, perhaps we can blindly call unsubscribeFromDupes call for *all* team subscriptions [13:17] bac, or, as the last resort, we can call both (and that might be necessary anyway if team is both directly subscribed and subscribed through dupe) [13:18] danilos: that seems to be most sane b/c that's what you want to do, right? unsubcribe, whether through this bug or through a dupe [13:18] bac, also, sorry about the confusing bug (i.e. not noting that the UI seems to work yet unsubscribing fails) :/ [13:19] np [13:19] bac, yeah, I am just wondering if unsubscribeFromDupes does that for you automatically [13:19] trying to figure out how to test this change has been vexing too [13:19] danilos: i'll check and make sure [13:20] bac, well, with the current code it's egregious, so I wouldn't mind if you don't test it at all, and I'll be providing tests as I go along [13:20] i recalled you mentioned passing in more info but it wasn't clear to me where to embed it [13:20] danilos: :/ [13:20] ok, thanks for the nudge [13:21] bac, right, I would have done it probably using more hidden input fields in the form, but perhaps doing both is best [13:21] == doing both API calls is best [13:22] which reminds me, I was supposed to be OCR :/ [13:22] heh [13:25] bac benji gmb danilos, kanban now-ish, call in 5, baby watch continually. [13:29] gmb, could you put the card for your bug on the board? I'm sorry we have to maintain everything twice [13:29] gary_poster: I meant to ask you about that. What do you want me to do with the existing card for that bug that's in Done-Done (I assume it was to track the triaging of the problem) [13:30] * gmb adds a card without the bug number for now anyway [13:30] gmb, just remove the number from the existing card [13:30] thank you [13:31] Done [13:32] good morning [13:33] morning Ursinha [13:33] danilos Skype? [13:33] gary_poster, yep, ready [13:34] gary_poster, didn't see the call invitation, sorry [13:36] np :-) [13:47] gmb, I've added some new unique indices for structural subscriptions, and IPersonSet.merge is complaining that we have not handled structuralsubscription.subscriber. Looking around, it doesn't appear that we merge structural subscriptions. Does that happen to ring a bell with any warnings, thoughts, or experiences, or should I just continue forging ahead blindly? [13:50] gary_poster: It rings a bell... I might have had to do something with it in the db patch branch that I did a few weeks back. [13:50] (i.e. I might have had to tell merge() to skip something) [13:50] Let me check... [13:51] gmb, I suspect that you told merge to skip.append(('bugsubscriptionfiltermute', 'person')) [13:51] gary_poster: Yes, that's it (just found it myself). [13:51] cool [13:52] but AFA you know, it is probably an oversight that we do not merge structural subscriptions, and I should do so? [13:52] Oh, and shouldn't we either delete or merge the bugsubscriptionfiltermute thing? [13:53] Well, AIUI, we shouldn't need to because the old Person becomes inactive after the merge, so it's just irrelevant data. [13:53] I don't know whether un-merging is possible, though - that's the only reason to keep such data around, really. [13:54] gary_poster: I don't know if not merging subscriptions is an oversight or whether it would cause horrible things to happen. TBH I'd try doing it and see what needs to be fixed to make it work. [13:54] gmb, fair enough, thanks === benji is now known as Guest41434 === benji____ is now known as benji [15:15] ok, I am off OCR, as soon as I put my name on, reviews have been coming in and I couldn't put a MP up myself :) [15:34] gary_poster: Can you remember whether, in the case of multiple filters, it's the most verbose or the least verbose that wins? [15:35] gmb, most always "wins"--it's a union of filters, so you can think of it as all of them "winning" [15:35] Hmm. [15:35] (that's the way we present it when we send emails with filter names--all matching names are included) [15:36] gary_poster: I'm just wondering, then, if maybe the bug isn't that we send out emails on creation to people who don't want them (e.g. Gavin), but that we tell them that the email matched a filter that it didn't actually match. [15:36] Because I can't reproduce Gavin's problem. [15:37] (I thought I could but that was due to db cruft; make schema fixed that) [15:37] I'll try writing a test for that instead... [15:37] gmb, maybe. did you dupe *that* problem? [15:37] Not yet. [15:37] ack [15:37] Well, I might have, but I wasn't looking for that problem. [15:38] * gmb goes back to the codeface [15:38] :-) [15:55] hi danilos -- this seems to be a fix that works: http://pastebin.ubuntu.com/599818/ [16:14] bac, right, so you are chaining them both after all? looks good, r=me if you want to land it :) [16:15] danilos: cool. i'm submitting a MP now [16:16] bac, btw, what happens with the animation/progress UI? [16:16] danilos: works as expected [16:17] danilos: as far as i can tell [16:17] bac, right, just checking [16:17] danilos: i was pleasantly surprised. :) [16:17] heh, so am I :) [16:19] gary_poster, btw, I've put up one branch for review, but no reviewer anymore: https://code.launchpad.net/~danilo/launchpad/advanced-bug-subscription-cleanup/+merge/59235 [16:19] bac, benji, gary_poster: if any of you can take a look at the above MP, I'd appreciate it, though I plan to leave in a bit [16:20] danilos: i can [16:20] bac, cool, thanks [16:21] danilos, how soon before EoD? Already? I have a branch for review now too [16:21] and you would be a good fit, but not necessary [16:21] gary_poster, something like 5 minutes :) [16:21] danilos, sounds perfect! ;-) [16:22] gary_poster, heh, send it over then :) [16:23] danilos, heh, I was kidding, but you can take a look if you want. stub has already reviewed the DB parts (lines 1-364) and the rest are changes I did to make the tests pass [16:23] https://code.launchpad.net/~gary/launchpad/bug753000/+merge/59139 [16:23] danilos, take a glance and see if you want to review it, or even if you don't but you have some comments ;-) [16:24] oh, diff is not showing the permission changes yet,, [16:24] there they are [16:24] danilos: can you look at https://code.launchpad.net/~bac/launchpad/bug-770245/+merge/59239 ? no surprises [16:25] bac, done [16:25] great [16:25] danilos: i'm not that fast. yours will take a little longer. [16:26] I like "Tests: Hmm." [16:26] :( [16:27] bac, :( [16:28] gary_poster, your branch generally looks fine, but I think it needs to get a better look than 3 minutes I can spare :) [16:28] danilos, understood, thanks for the glance [16:28] have a great evening [16:29] gary_poster, especially because of the person-merging additions [16:29] yeah [16:29] thanks, enjoy the day everyone :) [16:29] bac benji or gmb, would any of you be willing and able to do a code review (db review already done by stub) of https://code.launchpad.net/~gary/launchpad/bug753000/+merge/59139 ? [16:30] gary_poster: Sure. [16:30] yay, thanks gmb. [16:35] gary_poster: I've had another thought about that there bug: https://bugs.launchpad.net/launchpad/+bug/720147/comments/4 [16:35] looking [16:37] ack gmb. I think that making the filter work on newly created bugs would be ideal, but if not, yeah, we can clarify the current behavior better. [16:38] Right. [16:39] gary_poster: Now that I know it's not a bug per se I might actually find life a bit less complicated. I'll let you know how it goes. [16:39] cool [16:41] gary_poster: You need to add some explanatory comments / docstrings to the tests you've added to test_person.py (or at least to the two assert*()s that you added). Other than that, r=me. [16:41] gmb, cool, makes sense. Thank you! [16:41] np [16:54] does anyone know if we can add a project group ourselves (on qastaging) or if we have to get a LOSA? [16:55] I don't know how to do it other than asking a LOSA and the links I found don't give me any different ideas [17:06] gary_poster: https://help.launchpad.net/ProjectGroups is the only thing I can find that's relevant. [17:06] I don't think staging's any different. [17:07] (Though obviously we can bypass the ask a question stage) [17:07] gmb, yeah, me too. thank you for looking. [17:25] benji, any news on your bug 739141? [17:25] <_mup_> Bug #739141: I get lp bug mail even though 'on own actions' turned off < https://launchpad.net/bugs/739141 > [17:27] gary_poster: I'll be filing a merge proposal after lunch [17:27] benji, wow! [17:27] :) [17:27] I'll be reading that one. I'm more than a little bit curious ;-) [18:29] gary_poster: unfortunately it's a little mundane: https://code.edge.launchpad.net/~benji/launchpad/bug-739141/+merge/59260 [18:31] benji, well, the change description sounds like it is a part of the story I am unfamiliar with, so it sounds at least a bit interesting. :-) Want me to review it? [18:32] eh, it's so small, I will. :-) [18:33] gary_poster: sure, review away [18:34] benji, r=me [18:34] cool [18:36] bac or benji, I need a pre-imp for bug 770387. I expect it to be brief. Are either of you up for it? [18:36] <_mup_> Bug #770387: After changing the subscription option for my user, there's no notification about the change made

< https://launchpad.net/bugs/770387 > [18:36] gary_poster: I'm available [18:36] cool thanks [18:37] benji, mumble or skype? [18:37] skype [19:49] bac or benji, review of smallest branch evah, plz? https://code.launchpad.net/~gary/launchpad/bug770387/+merge/59269 [19:49] gary_poster: I can [19:49] thank you [19:50] gary_poster: approved [19:50] thanks benji :-) [19:50] np [19:50] I know, it was a pretty ambitious branch [20:07] gary_poster: would you object to me fixing bug 771335 by taking the bug name out of the form title? [20:08] <_mup_> Bug #771335: Title of add subscription widget can wrap if long. < https://launchpad.net/bugs/771335 > [20:09] benji, I would this week. If we don't have a good answer from Huw by next week, I'm alright with various hacks (I Think bac proposed one too, a few weeks ago) [20:09] So, IOW, I'm OK with you doing that, say, Tues or Wed of next week if there's not a better aswer by then [20:09] benji: huw said he is looking at it [20:09] right [20:09] and it is a generic LP problem [20:10] ok, good [20:10] i just forwarded his email [20:17] gary_poster: one of your branches is one away from being the release blocker! [20:17] quick! to the qastaging! [20:19] bac thanks for heads up... /me thought he had done everything [20:20] gary_poster: yeah, i don't look at that page often enough === gary_poster changed the topic of #launchpad-yellow to: https://dev.launchpad.net/yellow | http://launchpad.leankitkanban.com/Boards/Show/14028610 | http://irclogs.ubuntu.com/ | https://devpad.canonical.com/~lpqateam/qa_reports/deployment-stable.html [20:35] * gary_poster back in a bit [20:46] gary_poster: I may have duplicated some work, are you working on bug 771232? [20:46] <_mup_> Bug #771232: Edit structural subscription overlay is headed "Add a mail subscription for..."

< https://launchpad.net/bugs/771232 > [20:46] oh, gary if AFK [20:47] s/if/is/ [21:30] benji, yo. "a bit" turned out to be longer than I expected [21:30] that's cool [21:30] benji, yeah I was just starting on that. If you've already fixed it then that's great :-) [21:31] it seemed like the problem was we were creating a new form overlay each time [21:31] without deleting the old one sufficiently [21:31] I started, but have thrown it away, so it's all yours. [21:31] benji, so should you claim the bug, and I'll move on to another? [21:31] oh ok [22:18] gary_poster: remind me how you do python-like string interpolation in YUI [22:19] bac, I think you mean Y.substitute [22:20] bac, subscription.js has an advanced usage [22:20] yes, i think that is is [22:20] is it [22:20] thanks, i looked about but didn't see it [22:21] gary_poster: would you think this is overkill? http://pastebin.ubuntu.com/600017/ [22:23] bac, I guess that's alright. outerHTML is the only thing that makes me even slightly unhappy, really, but I see why you did it. FWIW, you have a typo in that line. var header = Y.Node.create('

or use

like benji does. [22:24] benji is just lazy. i like that. [22:24] :-) [22:25] gary_poster: what would you do to get rid of the outerHTML? [22:26] bac I didn't suggest anything because I didn't have any immediate counter suggestions, but I'll think for a sec... [22:26] gary_poster: i figured this was better than using Y.substitute...but not strongly [22:26] bac, did you try just passing it the node itself? Sometimes YUI is OK with stuff like that [22:27] That would be prettier [22:28] i did not [22:28] I just looked through the code a bit [22:29] indeed that works fine [22:29] thanks [22:29] awesome [22:29] np [22:29] i guess i'm shellshocked from seeing "YUI_3_object_br_549" rendered [22:29] heh === Ursinha is now known as Ursinha-afk [22:31] where does "heh" fit in with "ha" and "LOL"? [22:31] is it at all related to "meh" [22:32] I always ask myself the same question [23:04] For me, LOL is louder than heh [23:04] I never use ha, but if I did, it might be sarcastic [23:04] heh is a quieter laugh [23:05] bac, if you have a few more minutes, then https://code.launchpad.net/~gary/launchpad/bug771232/+merge/59302 could use a review; if not, no problem; we can do it tomorrow [23:06] * gary_poster departs for dinner and crying children, though will be back later