[00:41] wgrant: StevenK: local search removal was "inadvertent" http://www.androidcentral.com/international-galaxy-s3-local-search-removal-inadvertent-says-samsung-patch-coming [00:43] Hah [00:44] at least the UK courts have some sense. i can't wait to see those Apple "We are sorry" ads they will be forced to run [00:48] :) [00:49] wallyworld: Have you read The Rainmaker by John Grisham? [00:50] StevenK: no. what;s it about? [00:50] A lawyer, effectively. [00:50] of course, stuoid question :-) [00:52] wallyworld: I can just draw a few parallels between the court battle in The Rainmaker and Samsung v Apple in the US and the UK [00:53] i for the life of me will never understand how Apple has got away with all their shit in the US. i mean wtf. how did things get so broken [00:53] they must have photos of judges or something [00:54] Or the judges are Apple fans. [00:54] "The American company MUST win, the Koreans obviously stole everything." etc [00:55] yeah :-( But their Korean CEO wasn't quoted as saying "Good artists copy, great artists steal" or whatever it was [00:57] Oh my god, can I delete lib/lp/bugs/doc/bugnotification-sending.txt for being utterly disgusting? [01:01] only if you convert to unit tests :-P [01:01] wallyworld: Read through it and then say that with a straight face. [01:02] but then how could I push your buttons [01:07] StevenK: I think that was the first LP test I touched [01:07] So it has a special place in my heart. [01:07] Kill. [01:08] wgrant: I've stopped it failing at least. [01:08] 12 files changed, 21 insertions(+), 299 deletions(-) [01:09] Yeah [01:09] I changed most of that test [01:09] Well, reordered [01:09] Since my first LP change caused the order to no longer be undefined [01:09] 1407 lib/lp/bugs/doc/bugnotification-sending.txt [01:09] Holy crap [01:09] $ bzr cdiff -c8976 'lib/lp/bugs/doc/bugnotification-sending.txt' | wc -l [01:09] 1060 [01:10] wgrant: So I can bzr rm it from the tree for making peoples lives miserable? [01:11] Heh [01:21] wallyworld, wgrant: https://code.launchpad.net/~stevenk/launchpad/no-implicit-maintainer-sub/+merge/116991 [01:22] looking [01:22] StevenK: +21? What is this treachery? [01:23] I've got a locksmith turning up in about five [01:23] StevenK: couldn't you open the handcuffs? you still attached to the bed? [01:23] wallyworld: Hah [01:24] wallyworld: The deadbolt on the front door stopped working on Tuesday. The bolt is fully retracted and no amount of fiddling has made it come out. [01:24] StevenK: 521 -Foo Bar and Sample Person got subscribed to the bug. Foo Bar got [01:24] 522 -subscribed "implicitly", because he's the product owner. [01:24] * wallyworld bites his tongue [01:24] 523 +Foo Bar got subscribed to the bug. [01:24] StevenK: Was the original text the wrong way around? [01:24] wgrant: I have NFI [01:24] I made a guess [01:25] wallyworld: Oh? [01:25] The code changes look correct, but that test change seems the wrong way around. I guess if it passes it doesn't matter... [01:25] StevenK: don't worry. my mind is full of things which are best left unsaid [01:25] Is foo.bar Foo Bar or Sample Person? [01:26] should that line say "Sample Person ..." [01:26] since it previously said Foo Bar was the one implicitly subscribed [01:26] StevenK: Hm. I am confuse [01:27] StevenK: I think that code might be separate... [01:28] StevenK: Observe that the same logic is present in get_also_notified_subscribers [01:28] Maybe that's just to add the reason. [01:29] if pillar.owner in also_notified_subscribers: [01:29] Yeah [01:29] So kill that block [01:29] And addRegistrant [01:29] Right [01:29] And addRegistrant's tests [01:29] MOAR KILLING [01:30] def addMaintainer(self, person): [01:30] """Registers a maintainer of a bugtask's pillar of this bug.""" [01:30] Looks suspicious as well, but I don't think it's actually used [01:30] wtf [01:31] wtf [01:31] wtf [01:31] lib/lp/bugs/subscribers/bug.py [01:31] search for addMaintainer [01:31] I've often wondered why that strangeness happened [01:31] Now I know [01:32] I'll a card for reworking that [01:32] Or a bug [01:32] It's separate code from what you're working on [01:32] But should be reexamined in light of private structsubs [01:32] * StevenK watches the four bugnotificationrecipients.txt fall over in a heap of fail [01:33] wallyworld, StevenK: Do you see any reason to keep the maintainer/bugsupervisor section of add_bug_change_notifications? I'm pretty sure it becomes irrelevant once structsubs work [01:35] wgrant: i think it can go [01:35] StevenK: addSecurityContact isn't used or tested [01:35] addBugSupervisor is used, but only in the condemned code like addMaintainer is [01:37] wgrant: I'm to kill addMaintainer, or is that for later? [01:37] StevenK: This branch is big enough. Kill addSecurityContact since it's unused, and we'll investigate addMaintainer/addBugSupervisor's single callsite later [01:37] "later" may mean after lunch [01:38] wgrant: add{Registrant,SecurityContact}()'s killing has been pushed [01:38] BugNotificationRecipients is going to be like 2 lines by the time you're done [01:39] This sounds like an excellent plan [01:42] wgrant, wallyworld: Diff updated [01:42] StevenK: Thanks [01:42] I've verified that the suspicious changes were just a bug in the old test. [01:43] r=me [01:43] This is my surprised face. [01:44] wgrant: no-qa or is there a bug for this? [01:44] Now now [01:44] Doctests are people too [01:44] this? :O [01:44] StevenK: There may not be a bug, but there should be [01:44] Let me file one and create a card [01:44] ie. if there isn't, file [01:46] I thought there was one, but I can't find it. So file a new one :) [01:46] StevenK: Bug #483521 is relevant to the next phase of this [01:46] <_mup_> Bug #483521: Allow the project owner to appoint other people as the bug supervisor < https://launchpad.net/bugs/483521 > [01:48] Bug #113825 could also be [01:48] <_mup_> Bug #113825: Null bug supervisor is confusing

< https://launchpad.net/bugs/113825 > [01:49] StevenK: Sort of [01:49] But not directly. [01:49] * StevenK files a new bug [01:52] wgrant: Bug #1029725 [01:52] <_mup_> Bug #1029725: Pillar owners should not be implicity notified about bugs < https://launchpad.net/bugs/1029725 > [01:52] wgrant: as far as i can tell, there's no api to get a bug by id honouring the privacy rules. i can add a bug_id parameter to BugTaskSearchParams. do you have an issue with this? [01:52] StevenK: Bug #1029724 [01:52] <_mup_> Bug #1029724: Retargeting a private bug notifies maintainer and bug supervisor without subscriptions < https://launchpad.net/bugs/1029724 > [01:53] wallyworld: bug= should work [01:53] It's meant to cope with an ID too, I'm pretty sure [01:53] ok, i didn't realise that supported taking an id [01:53] will check [01:54] standard_args = { | self.distroseries = distroseries [01:54] BugTaskFlat.bug: params.bug, [01:54] wgrant: Hm. [01:54] It'll compare it with the Storm Reference [01:54] wgrant: Both of them? [01:54] Which should work for an ID [01:54] StevenK: ? [01:54] wgrant: Oh, is the 724 the upcoming work? [01:54] StevenK: Yeah, 724 is the next bit [01:55] 724 is about removing the implicit, hardcoded, un-opt-outable subscription [01:55] 521 is part of removing the automatic structural subscription given to bugsupervisors [01:55] * StevenK notes Coding has been wallyworld'd [01:56] StevenK: i have a branch which fixes 4 bugs [01:56] Hm [01:56] I think that's all of it [01:57] After fixing those two, you'll only be notified about a project's bugs if you ask to be notified about a project's bugs :) [01:57] A revolutionary concept, I know [01:57] wgrant: Tossing it at ec2. [01:58] Great. [01:58] * wgrant wanders off for a while. [02:42] wgrant: Can haz pre-impl? Or a review? [02:42] As for preimp... you basically just need to copy the status/importance stuff [02:42] It's identical [02:44] wgrant: Where does that get stuffed into the DB? [02:44] TABLE "bugsubscriptionfilterstatus" CONSTRAINT "bugsubscriptionfilterstatus_filter_fkey" FOREIGN KEY (filter) REFERENCES bugsubscriptionfilter(id) [02:45] Right [02:51] Rargh, it's a new table, not even a column :-( [02:51] Right [02:51] Since it's many-to-many [03:51] wgrant: Uhhhh. lib/lp/archiveuploader/tests/nascentupload-closing-bugs.txt fails with the death, murder and destruction we did this morning. [03:52] StevenK: Failure is expected, but in what manner? [03:52] wgrant: http://pastebin.ubuntu.com/1113149/ [03:53] StevenK: Right, is there anything odd in the slightest about that? [03:53] Seems completely expected [03:53] You removed a notification mechanism [03:53] And now some notifications aren't happening. [04:36] wgrant: StevenK: https://code.launchpad.net/~lifeless/python-oops-tools/report-recipient/+merge/116999 if you want less mail. [04:40] lifeless: r=me [05:00] wgrant: Review? [05:02] StevenK: I'm trying to break it [05:03] It's not going very well [05:03] Hahaha [05:04] I guess this means I am smart enough to write it [05:06] I need more monitors [05:08] Hah [05:09] wgrant: I have the DB patch up [05:10] StevenK: Is it as immense as I expect? [05:10] +24 [05:11] Looks good [05:19] StevenK: How's no-implicit-maintainer-sub doing in EC2? [05:19] no-implicit-maintainer-sub => devel [FAILED] (up for 3:19:34) i-96b9b5ee [05:20] nascentupload-closing-bugs.txt is the only failure [05:20] Excellent [05:21] So I guess I can lp-land it in about an hour [05:42] StevenK: r not quite equals me [05:42] wgrant: Oh? Why? [05:46] wgrant: assignees behaves differently depending if there is a bugtask or not, I wanted to support both. [05:47] StevenK: Which bit's that about? [05:47] Why have you split the subquery out into a separate local? Might as well inline it, or at least not shadow a builtin. Also, did you consider inlining the subsequent filter into load_people's where argument? [05:48] I considered the Or/True thing for whatever I rename forbidden_recipients_filter to, so I'll do that. [05:53] Ah [06:30] wgrant: no-implicit-maintainer-sub is in r15699 [06:30] So I saw [06:31] lifeless, hi [06:31] hi all [06:31] Hi poolie. [06:31] well done on the privacy and other recent changes [06:35] hey poolie [06:35] poolie: hi [06:35] poolie: O hai [06:37] * StevenK steals r15700 [07:45] good morning [07:51] hey adeuring [07:51] hi czajkowski! [08:08] gary_poster, benji (others welcome): we're looking at switching our dependency management to buildout and were wondering if you could help us get started. === adeuring changed the topic of #launchpad-dev to: http://dev.launchpad.net/ | On call reviewer: adeuring | Firefighting: - | Critical bugs: 4.0*10^2 [09:54] jcsackett: ping me when you're online please :) [10:58] jml: I am about to be AFK for an hour and then I have calls and interviews for much of the day (similar for Gary) but I would love to help you with buildout if I can. [10:58] benji: thanks! [10:59] benji: https://code.launchpad.net/~james-w/pkgme-service-python/buildout/+merge/116980 has a summary of where we're at [10:59] cool, I'll take a look when I have a spare moment; is there anything in particular you need added next or are you looking more for suggestions with what you have? [11:01] benji: wondering if there are known workarounds for https://bugs.launchpad.net/zc.buildout/+bug/1029715 and possibly some direction in fixing it. [11:01] <_mup_> Bug #1029715: bootstrap.py fails if there is a system-wide install of the desired buildout version < https://launchpad.net/bugs/1029715 > [11:01] benji: also confirmation of our approach in sharing eggs [11:02] sounds good, I'll take a look when I can [11:02] benji: and of course we'd welcome any feedback from you & gary since you've got way more experience with this and will probably have suggestions on things we wouldn't think to ask about. [11:03] benji: thanks. [11:59] How can I configure launchpad to send notifications(email)? Any hints? [12:03] jml, hi. I will be more than happy to leave that to benji's capable hands, but also will be happy to help later if needed. === nigelbabu is now known as nigelb [14:29] jcsackett: can you investigate why comment hiding is broken on a bug https://bugs.launchpad.net/launchpad/+bug/1029668 [14:29] <_mup_> Bug #1029668: Hide Comment works sometimes, doesn't work other times

< https://launchpad.net/bugs/1029668 > [14:30] sinzui: looking. [14:32] jcsackett: maybe comment hiding does not work with the comments that are loaded in the sceond phase [14:33] sinzui: i think it has been fixed by my last comment js branch, which has landed. [14:33] sinzui: open the same bug on qastaging and see if you can reproduce. i cannot, though i can on production. [14:34] jcsackett: the low number comments can be hidden, but I cannot hide the high number comments. I think the division is the comments that are loaded by the timer are inserted into the dom, but they are not wired for comments [14:34] sinzui: right, and that isn't an issue in the refactor i did, as it just puts a delegate listener on the container for all comments. [14:36] jcsackett: I agree that delegates fixed the issue. [14:36] sinzui: fantastic. [14:36] jcsackett: claim the bug, mark it fix committed, and a card in the ready lane [14:36] will do. [14:40] jcsackett: ignore ping that was the bug I was looking for you for. [14:44] adeuring: Could you please review https://code.launchpad.net/~abentley/launchpad/re-roll-r15692/+merge/117070 ? [14:44] abentley: sure [14:50] abentley: r=me [14:51] adeuring: Thanks! [15:11] benji: sorry I go by the online calendar :) [15:12] czajkowski: as you should; I am sorry I didn't get my leave in earlier [15:12] sinzui: http://people.canonical.com/~curtis/lp-milestone/report.html is in the iso-8859-1 encoding, but contains what appears to be a utf-8-encoded character after "In progress and fix committed", etc. [15:12] that sucks. [15:12] I will fix that today [15:13] thank you abentley [15:13] sinzui: np [15:13] benji: no bother I try and sent out it out late UK time on a Friday === ev_ is now known as ev [18:10] jcsackett: I like your proposal to keep the bug info in one column. I think the misalignment of ids is acceptable and will not happen often. You may have noticed that when we moved the related bug to the MP summary, we got phantom space caused by attempts to handle unwanted table columns...your code might be reusable to fix the MP page [18:24] sinzui: awesome, thanks. [19:05] sinzui: do you have time to look at the finished branch for me? https://code.launchpad.net/~jcsackett/launchpad/misaligned-details/+merge/117118 [19:05] I will do that now [19:05] thanks. [19:07] jcsackett: I don't think class="amount" is needed [19:07] sinzui: no? i'll remove it and check how it looks. [19:07] i think that was just cribbed from other layouts. [19:08] jcsackett: yes, it means right align the number in the table column [19:08] is class="icon left" really doing something? I think it want to float an icon, but we are using a sprite [19:09] oh, jcsackett you may need to add action-icon to the sprite so that older browsers see it [19:10] sinzui: not sure. it was "icon right" which had the sprite sort of misaligned with the edge of the table. [19:10] "icon left" fixed that. [19:10] Older browser do not show empty spans, so we need the action-icon, and that turns the span into an inline-block which will change the float behaviour. [19:12] jcsackett: okay, I understand. we might need it, check that it is still needed after action-icon. I don't think anything should be floated, but we might have to if something else is making other parts float [19:14] sinzui: dig. [19:14] jcsackett: the sprite could also be on the bug id so we do not need the action-icon and we have less markup to manage [19:14] sinzui: ah, nice idea. [19:14] oh, and that makes the table item rules from bugs and branch almost the same [19:19] sinzui: hm. adding sprite onto the bug id span and removing the enclosing span ends up with everything floating a bit too far right from the edge of the table. [19:19] getting you a screenshot. [19:20] sinzui: http://people.canonical.com/~jc/images/misaligned-with-new-markup.png [19:22] jcsackett: with the removal of 3 columns, we have ambigous colspans being used by bugs and branches. We just want one column for the td [19:32] jcsackett: when I hack the html in https://qastaging.launchpad.net/launchpad/+sharing/abentley to do what what you tried, I see the bug sprites finally line up with the branch sprites. [19:35] jcsackett: I see the issue, it is not just the colspan, it is the extra cols in colgroups [19:36] we do not need the first two with your change [19:36] sinzui: ok. so i'll remove the extra col stuff and ping you when i've got that all working. [20:26] sinzui: fixes pushed up. sorry for the delay, my znc server has crashed. [20:26] np [20:28] jcsackett: move {{bug_id}} from amount into the sprite [20:28] then delete the amount span [20:28] and most importantly, does it look good after you make the change? [20:29] sinzui: it looked perfect post change. http://people.canonical.com/~jc/images/simplified-markup.png [20:29] fab. [20:29] Then r=me