[21:29] <barry> @cachedproperty [21:29] <barry> def page_title(self): [21:29] <barry> """The page title, constructed from the reversed breadcrumbs.""" [21:29] <barry> return COLON.join( [21:29] <barry> breadcrumb.text for breadcrumb in reversed(self.items)) [21:29] <barry> [21:30] <barry> salgado: but the title says "+review : Mozilla Firefox : Mark Shuttleworth" [21:30] <barry> salgado: i can't figure out why the breadcrumbs doesn't have mark as the first component but the page_title does [21:30] <barry> salgado: does this ring any bells? [21:31] <barry> salgado: i'm looking at launchpad-hierarchy.pt and can't see where that might be conditional, and i'm using the same Hierarchy.items property (i think anyway) [21:31] <barry> salgado: first question i guess is which would you consider right? [21:31] <barry> salgado: i.e. maybe the in-page breadcrumbs really should be mark >> firefox >> +review [21:32] <barry> salgado: oh, and this page: https://launchpad.dev/~name16/+karma [21:32] <salgado> barry, that page uses a custom hierarchy view (see BranchHierarchy in lib/lp/code/browser/branch.py), and I guess you're using the standard Hierarchy view to generate the title [21:32] <barry> salgado: has Foo Bar >> +karma as the breadcrumbs and +karma : Foo Bar as the title [21:32] <barry> salgado: gah! that must be it [21:33] <salgado> the breadcrumbs on that page are broken, though. someone has to fix that custom Hierarchy view [21:33] <barry> salgado: thanks. i'd seen that but didn't but 2 and 2 together [21:34] <barry> salgado: i'm not going to worry about that :) when they fix that, both will be fixed. i just need to add another adapter for @@+page-title [21:34] <barry> salgado: let me try that... thanks! [21:34] <salgado> barry, you're welcome. :) [21:36] <salgado> barry, I don't know how your adapter works, but you might want to use getMultiAdapter((obj, request), name='+hierarchy') to get the correct Hierarchy view instead of hardcoding the standard one there and redefining it for the objects that use a custom Hierarchy view [21:39] <barry> salgado: yeah, dang. just adding the browser:page to configure.zcml did not work [21:39] <mwhudson> morning [21:40] <barry> salgado: i didn't add an adapter, i just added a page_title property to Hierarchy, and then added the browser:page zcml [21:40] <thumper> morning [21:40] <thumper> can someone explain to my why the build has been broken for over 10 hours? [21:40] <mwhudson> gary_poster: did you see my buildbot merge request? [21:40] <barry> salgado: i would have thought BranchHierarchy.objects would be enough to make it work [21:41] <gary_poster> mwhudson: probably. on call, one sec [21:41] <mwhudson> ok [21:42] <mwhudson> thumper: "psycopg2.OperationalError: could not write to file "base/195483/2667": No space left on device" [21:42] <mwhudson> slaves seem to be running out of space!? [21:42] <barry> salgado: ah, there are more browser:pages to add... [21:42] <salgado> barry, it does, but I think the problem is that the Hierarchy view is hardcoded somewhere, so it's being used to generate the title of that page instead of BranchHierarchy [21:44] <gary_poster> mwhudson, yes, I got it all nice and queued up for me to read this morning and have not looked at it since. :-P looking now [21:45] <mwhudson> gary_poster: :) [21:45] <barry> salgado: joy! [21:45] <gary_poster> I also looked at the draft last night [21:47] <thumper> mwhudson: oh arse [21:48] <thumper> barry: on the new active reviews page we have approved merges shown too [21:48] <thumper> barry: you have some old ones showing [21:48] <gary_poster> mwhudson: fwiw, I contributed some bzr bits to the "contrib" directory (or equivalent, I forget the exact name) of twisted. They included a bzr push plugin, and a poller that at one time bore some resemblance to the one we are using. IIRC, neither of those had tests either. [21:48] <thumper> barry: which I think are just artifacts due to looms [21:48] <gary_poster> It might be nice in a theoretical sort of way to sometime get to the point that we are using the contributed versions for some of these things--so that our changes continue to enrich upstream, for instance. Probably obviously, I didn't do that because of ease and expedience, and that may continue to be a driving factor. [21:48] <thumper> barry: can you mark them merged if they are actually merged? [21:49] <gary_poster> mwhudson: s/twisted/buildbot/ [21:49] <gary_poster> that is, I contributed to buildbot, as you'd expect [21:49] <mwhudson> gary_poster: yes, that's a good goal [21:50] <mwhudson> gary_poster: our testfix mode voodoo seems a little special cased to us, perhaps, but maybe not... [21:50] <gary_poster> ah yes [21:50] <mwhudson> gary_poster: in any case, it needs testing in our environment before we do that :) [21:50] <gary_poster> agree [21:51] <barry> thumper: will do [21:52] <mwhudson> gary_poster: also, does this out of disk space error make any sense to you> [21:52] <mwhudson> ? [21:52] <gary_poster> mwhudson: "Carry on" link: +1 :-) [21:52] <thumper> ta [21:52] <mwhudson> gary_poster: yeah, that's hideous :) [21:55] <gary_poster> mwhudson: don't see out of disk space error--maybe started with a slash and was swallowed by IRC gods? Looked on lpbuildbot and didn't see anything pertinent at a quick glance. We have had things like that before. They usually seemed to be spurious random EC2 things. Sometimes we did identify some way that we were not cleaning up our logs, but most of the time it seemed unlikely that we were at fault [21:57] <barry> thumper: done [21:57] <thumper> barry: awesome, thanks [21:57] <thumper> barry: in which case you can reply to my email to encourage the others :) [21:58] <barry> :) [22:02] <gary_poster> mwhudson: r=gary. ship it. === salgado is now known as salgado-afk [22:11] <thumper> barry: easy _and_ fun? :-) === henrique is now known as henrique-out [22:13] <gary_poster> mwhudson: thumper may have already spoken to you. the next buildbot I think we will want is a JS buildbot, running once daily like the bzr buildbot. flacoste is the keeper of the keys of knowledge here, while mars is out. Please ping me about it when you have a moment, or go to thumper or flacoste. Note that the js tests are currently failing; having a reminder of this is the point. [22:14] <mwhudson> gary_poster: sounds like an excuse to tidy up master.cfg [22:14] <gary_poster> mwhudson: +1 [22:16] <thumper> mwhudson: skype? [22:16] <thumper> rockstar: skype? [22:17] <mwhudson> thumper: yeah, skype works better when it's running [22:20] <kfogel> jml: my other call tomorrow got moved; I can (& would like to) do our call earlier if possible. How much earlier is good for you? === matsubara is now known as matsubara-afk [22:30] <barry> thumper: okay. easy :) === henrique-out is now known as henrique [22:35] <thumper> sinzui: IRevisionCache(self.context).getRevisions.order_by(revision_date).config(limit=5) === cprov is now known as cprov-afk [22:39] <thumper> sinzui: getUtility(IAllBranches).getMergeProposalsForReviewer(person) [22:41] <mwhudson> gary_poster: in general, shouldn't it be possible for buildbot-poll to be replaced with something that listens to a statustarget we add to buildbot? [22:41] * mwhudson dislikes polling [22:58] <mwhudson> flacoste: you here? [22:58] <flacoste> mwhudson: still [22:58] <mwhudson> flacoste: do you want to talk about a windmill buildbot? [22:58] <flacoste> mwhudson: sure, i'll forward you the email from mars on this [22:59] <mwhudson> flacoste: how much longer are you going to be around for? [22:59] <mwhudson> thanks === thumper is now known as thumper-afk [22:59] <flacoste> mwhudson: i need to leave for an external dinner very soon [22:59] <flacoste> so not much actually [22:59] <mwhudson> flacoste: ok [23:00] <flacoste> mwhudson: i'll pop-in when i'm back later though [23:00] <mwhudson> flacoste: so if the mail doesn't make enough sense, i'll talk to you next week about it i guess [23:00] <mwhudson> flacoste: oh, ok [23:01] <flacoste> mwhudson: it's actually three emails [23:01] <flacoste> all sent [23:02] <mwhudson> flacoste: thanks [23:03] <flacoste> ok, i need to run [23:03] <mwhudson> bye for now [23:30] * mwhudson decamping to a cafe in town [23:33] <bac> hi sinzui [23:33] <bac> sinzui: i'm blocked on this test: http://pastebin.ubuntu.com/264677/ [23:34] <bac> sinzui: in the 'Adding a download file' i expected to get some view.errors since some required input is missing. i don't care to demonstrate uploading a file, just checking label and page_title, but i'm confused why the error processing isn't working [23:36] <bac> sinzui: the branch with the broken test is at lp:~bac/launchpad/bug-422974-productrelease [23:36] <mwhudson> biab [23:36] * bac -> dinner [23:42] <wgrant> I still can't access launchpad's +activereviews.

[00:00] cprov, wuu Code team [00:02] wgrant: beuno-on-vacatio and barry may be able to explain. I think some more work is needed. Barry has bravely undertaken the issue [00:03] sinzui: What do you think I should do next for this? flacoste and I conjectured that the appropriate thing to do next would be for me to commit one more revision to the branch, releasing and tagging the new version, upload the source to my PPA, and then a ~launchpad member could pull/push my branch into trunk and copy the source+binaries to ~launchpad/ppa [00:04] maxb: I will leave that to foundations who makes the deps. [00:04] maxb: I do not think we need tidy, though that is not a karmi problem [00:05] Is there a list of who is the foundations team anywhere? [00:06] maxb: I wrote the frankenscript that we use for linting. it breaks with most Ubuntu releases. I have replaced it in my editor. I plan to replace the entire lint checking with a script that does not need packages. [00:06] maxb: flacoste, gary poster, and stub are the three people who manage that [00:09] ok, I'll catch flacoste on irc some time, since I've been working with him for the rest of the .deb stuff [00:10] I see they did not organise themselves into a team [00:13] I see the team page has been hit by the ugly stick. [00:14] * sinzui postpones writing a bug report because it will ruin his evening [00:14] What's wrong with it? [00:16] The details portlet should be the first one in the left column; contact the team should be a link in action menu. owner, created should be first in the details [00:17] I also think the map could probably go in the right column, as it would often fit alongside the very long participation list. [00:17] polls should not appear if I cannot create one [00:18] I filed that one last night. [00:18] wgrant: !! I suggested the same thing and continue to suggest it for the profile page [00:19] sinzui: Well, I've no idea what the 3.0 Person index looks like, as it seems to be hiding on devpad. [00:20] can you see people.canonical.com...I set myself up on that today [00:20] Ah, it's up again. [00:20] Yes, that one is public. [00:22] http://people.canonical.com/~curtis/ [00:22] ^ The user pages are very old, They do show the content we want. When we started the team page, we had made big changes to the header, so I proposed that the details go to the left like pillars, and place the map on the right. [00:23] Ah, I see. [00:28] It is tempting to un-break the Packaging-related views now that it's used for something visible. [00:29] yes it is [00:29] sinzui: What do you think of the "This is a Restricted Team.(?)" thing? Doesn't HTML have a nice tag for doing that in a less confusing way? [00:30] rockstar: ping [00:30] thumper, pong [00:30] rockstar: call? [00:31] thumper, yea, lemme bring the dogs in the house really quick. [00:31] thumper, call when you're ready. [00:32] wgrant: Thanks for point that out. That is in the wrong place. I have no idea what the icon means. That belongs in the details...which I think we decide would be names Team information [00:32] sinzui: The icon has an explanation of "restricted team" as a title. [00:32] I see the related projects portlet did not get fixed [00:33] I believe it should be inline help [00:33] What's broken about it? === jkakar_ is now known as jkakar [00:34] rockstar, you there? [00:35] Ursinha, yes, but on the phone. [00:35] oops. [00:35] sinzui, are you available for a call? [00:35] rockstar, can we chat briefly after that about two code imports that are failing? [00:35] jml: I will be in a few hours. My children want me to feed them [00:35] sinzui, cool, thanks. [00:46] thumper, http://people.canonical.com/~beuno/branch-index.png [00:46] Ursinha, yes ma'am [00:47] oooh, mockups [00:48] rockstar, there are two imports from sf's svn failing similarly [00:48] rockstar, http://launchpadlibrarian.net/31252118/timer-applet-trunk-log.txt and http://launchpadlibrarian.net/31258356/whirlwind-sf-trunk-log.txt [00:48] rockstar, is it our problem or sf's? [00:48] or the branches.. [00:59] Ursinha, can you change the import path to not use https, opting for http? [01:02] jml: I see what happened. [01:03] kfogel, are you available for a call now [01:03] jml: two things in combination: 1) we didn't put it in the online calendar, and 2) Jono *Bacon* and I did put a call in the calender, for this Friday. He did the invite title as "Karl/Jono". Today I looked at my calender, thinking "Oh, don't Jono (Lange) and I have a call today?", and I saw "Karl/Jono" listed for Friday, and nothing today, so I thought "Guess I must have been wrong." [01:03] jml: I can't -- I accepted a friend's invitation for dinner. [01:04] kfogel, ok [01:04] jml: let me get my cal out [01:04] kfogel, UTC 2300 tomorrow then? [01:04] rockstar, sure [01:05] jml: tomorrow night completely booked (http://thechangeyouwanttosee.com/event/copyright-and-creative-practice-jamaica-and-beyond-dj-dance-party). How about 2300 Friday? That's 6pm EDT, which is right after my call with OtherJono. [01:05] * jml considers [01:06] kfogel, sure. [01:06] 'jml' for reduced ambiguity :) [01:06] jml: you should have an invite now [01:07] jml: I did it as "Karl/Jono" again, finding that I cannot resist tempting fate. [01:11] :) [01:15] jml: i forgot about you guys tonight didn't i :( [01:15] barry, you did, I think. [01:15] barry, but that's probably ok :) [01:16] phew :) [01:16] jml: we had a bit of late night excitement today. we got a cat [01:17] barry, I saw [01:17] barry, yay for cats :) [01:18] \o/ [01:18] rockstar, I've changed them both and asked to reimport, let's see [01:18] oh well, we'll *cat*ch up next week [01:19] rockstar, I have another question, unrelated :) you are the guy working on answers templates, right? [01:19] barry, hahahaha [01:19] Ursinha, yes. I saw the bug, will work on it as soon as thumper stops talking to me. :) [01:19] rockstar, actually was more a question than a bug.. where is the assignee of the question? am I blind or is it really missing [01:20] barry, ... [01:20] thumper, I accidentally submitted a merge proposal just before the cron job was scheduled to run. [01:21] thumper, man it would be nice if it always was that quick [01:21] :) [01:21] RabbitMQ ftw [01:21] I was just talking with rockstar about message queues in that position [01:23] it's totally what we should do first with message queues [01:24] is there anything actually stopping us from trying that today? [01:24] other than having other non-related things to do instead [01:26] well, that last point is what stops us doing all the fun things, isn't it? [01:26] we could do landing queues today, if we didn't have other things to do [01:26] but i guess there's some infrastructural stuff to make decisions on (or at least find out if decisions have already been made) [01:28] mwhudson, hmm. sounds very mysterious. [01:28] jml: i wasn't trying to be [01:29] jml: but before we can use message queues, we need a message queue set up and running [01:29] i don't know if that's been done yet [01:29] mwhudson, sorry, I didn't mean to say you were. Just that the whole process isn't particularly easy to inspect. [01:30] jml: true [01:30] jml: there may be a wiki page about this (ha ha!) [01:33] * jml looks [01:34] not on the public wiki [01:35] jml: I believe that the only thing stopping us using rabbit mq is resources to work on the problem [01:36] thumper, I was just thinking about how long it would take to get the merge proposal use-case working on a local laptop [01:36] thumper, pretending we had a magic wand to wave away all the production deployment issues. [01:36] :) [01:37] thumper, I'm serious! [01:42] jml, thumper, mwhudson: we are using message queues in landscape. we should probably have a pow-wow on that. maybe on softarch? [01:42] sidnei: what is softarch? [01:42] heh heh [01:42] thumper: the softarch CoP i mean [01:43] i'm not in softarch yet [01:43] i need to pummel gustavo into adding me i guess [01:43] sidnei: I'm still not clear what you're talking about [01:43] * thumper has vague recollections of an email from gustavo [01:44] yeah, that /me does jedi handwave [01:44] * thumper can't find the email [01:44] * thumper has to dash, back later === thumper is now known as thumper-afk === thumper-afk is now known as thumper [03:46] w00t [03:47] LCA2010 paper accepted: Using Launchpad for Code Reviews [03:48] \o/ [03:48] congrats thumper, great news! [03:48] school run, bbs === abentley1 is now known as abentley [03:49] thumper: nice, will be seeing you there [04:07] thumper: yay [05:09] thumper, woot [05:11] thumper, do you want to talk at some point about Code team hand off? [05:11] jml: sure [05:11] jml: please rephrase :P [05:11] 20m now? [05:11] lifeless: he said hand off, not hand job [05:11] thumper, 20m from now, or 20m starting now? [05:12] now for 20m [05:12] cool. [05:12] lets' === jtv is now known as jtv-eat [06:55] * jml off for a bit [06:57] for most of the older bugs tagged build-infrastructure, i have no idea what a fix might look like [07:07] * thumper EODs [07:27] netsplit :( [08:27] good morning [10:39] * wgrant plays whackamole with sprites. [10:42] * jml is finishing up package-permission-lovwe [10:42] love, rather. [10:42] jml is my hero [10:43] aww shucks. [10:45] jml: I think our review thread has finished, unless I take it way OT :) [10:46] bigjools, heh, yeah, I think so :) [10:46] ooo and see lifeless's reply [10:47] yeah, just did :) [10:47] there ya go then, exceptions suck :) [10:48] let's use return codes for everything :) [10:50] there's one more thing I need to do for the review... move makeGPGKey [10:50] (and fix it) [10:57] Morning. [10:58] bigjools: it actually sucked a lot [10:58] bigjools: there was a place we really wanted a catch to transform an underlying failure into a clean message, without catching too much [10:58] lifeless: I used to work somewhere where they were banned in our near-realtime env [10:58] bigjools: poolie put it in... and we had to take it out again [10:58] bigjools: c++ or python? [10:58] c++ [10:59] yeah [10:59] c++ ones aren't that cheap either [10:59] python and realtime don't go together to well [10:59] bigjools: they do so; the end of the universe requires entropy! [10:59] arf :) [11:01] ok, added a getUniqueHexString function to the factory [11:02] and submitting to ec2test. === bigjools is now known as bigjools-afk [12:33] Hi deryck, I've got two questions about the inlineedit widget. It has a margin-top: -10px; - but this causes some issues for me when integrating it - do you remember what the reason for it was? [12:33] Secondly, the width seems to be fixed, so when the browser is made narrow, the edit-button/tab overlaps with other content... is that known? [12:33] * noodles775 checks to see if there's a bug for it. [12:34] noodles775, I think the top margin is to get the tab and edit icon at the top of the widget to line up correctly against the heading, i.e. "Bug description" in my case [12:34] noodles775, and yeah, the fixed width is known and problematic [12:34] noodles775, but not simple to fix [12:35] deryck: great, as long as it's known. Thanks! === matsubara-afk is now known as matsubara [13:59] Hi, can I run rocketfuel-setup twice or will this break things? [14:02] simon-o, it shouldn't break anything. if it does, it's a bug that we need to fix [14:02] salgado: ok, I thought that. I'll report back if it worked :) === henninge_ is now known as henninge [14:07] salgado: it doesn't work if the folder $LP_TRUNK_NAME exists, but is incomplete [14:07] the bzr branch command failed and it was incomplete [14:08] maybe you could check first if it's a valid bzr branch and nothing is missing [14:11] simon-o, indeed. care to file a bug about that? [14:11] salgado: sure, I'll try to attach a merge proposal too, if that's ok ;) [14:12] simon-o, that'd be perfect! [14:19] morning launchpad [14:34] I have a clean Ubuntu install, I've followed the guide exactly and in the "make run" stage I get "psycopg2.OperationalError: FATAL: Ident authentication failed for user "launchpad_main"" after that it quits, I've tried doing a "make clean.... make run" but same deal [14:35] jamone1313: You forgot to run launchpad-database-setup? [14:36] Were you following https://dev.launchpad.net/Running? [14:36] wgrand: No, I ran it [14:36] //help [14:37] erg [14:37] * wgrant doesn't know, then. [14:38] wgrant: Just to be sure I just reran the db-setup and that fixed it. Maby it failed last time [14:38] https://code.edge.launchpad.net/launchpad/+activereviews is 403ing for me ATM. [14:38] That shouldn't happen. === barry` is now known as barry [14:43] salgado: do you know of any existing script to automate importing a bunch of old series/releases and their respective download files into launchpad? [14:47] sidnei, I don't think we have one, but let's see what sinzui says about it. [14:48] sidnei, salgado : no, script, but the product-release-finder will create the milestones and releases for each series it is setup to find [14:48] i guess it's time to give it a try again :) [14:49] sidnei: there is a bug that a user claims he will fix that will ensure all the release files are downloaded. [14:55] allenap: will your formatter work for compound file endings, like .tar.gz? [15:00] bigjools-afk: Yeah, should do. mimetypes.suffix_map contains some .tar.gz-like ones. [15:00] allenap: awesome! === bigjools-afk is now known as bigjools === kiko-afk is now known as kiko [15:38] Chex, gary_poster, rockstar, bigjools, henninge, sinzui, intellectronica, Ursinha: LP production meeting in 22 minutes at #launchpad-meeting [15:38] matsubara, roger [15:39] roger, matsubara === cprov is now known as cprov-lunch [16:11] Hi, I fixed a small bug in rocketfuel-setup. I linked a branch in bug 402187. What to do next? Propose for merge or find a reviewer? [16:11] Bug #402187: rocketfuel-setup can't resume if fails to complete [16:26] simon-o, that was quick! you can propose a merge and then poke the on call reviewer on #launchpad-reviews [16:28] salgado: It was only a small fix. I have to merge into devel not db-devel, right? [16:29] simon-o, right [16:34] salgado: Thanks for your help. Do I also need to sign the Contributor Agreement Form, or is this only for larger contributions? [16:36] simon-o, I think you have to sign it for all contributions. let's confirm with kfogel [16:37] salgado, simon-o: the contributor agreement is something you sign once and once only. [16:37] salgado, simon-o: should happen with the first contribution to any Canonical project. [16:37] kfogel: ok, than I have to sign it now. [16:38] Even if it's a small fix, I'd recommend just doing it now, to get it out of the way for the future, and because (due to international copyright law being inconsistent) what counts as a "small fix" in one place maybe be considered copyrightable somewhere else. [16:38] simon-o: thanks [16:38] salgado: you know the process? (http://canonical.com/contributors) [16:38] kfogel: Thanks for your explanation. I'll send it right now [16:39] great [16:39] kfogel, yes, I do. I was just wondering whether or not this applied to small contributions as well, just failed to make that clear. ;) [16:40] salgado: yup, it does -- see http://canonical.com/contributors/faq (which is a relatively new document) [16:42] kfogel, ah, nice to know about that. thanks [16:42] simon-o: (above URL might be of interest to you too) [16:43] bac: i've made some changes to my branch which will affect the headers. i'm going to get some lunch, but i'd love to see how your pages look with my branch. maybe we can try some things afterward? [16:44] barry: ok. mired in test failures ATM [16:44] kfogel, salgado: just saw that too. Do I need to send this to Francis Lacoste or to whom? [16:44] bac: y'know, that's a good idea. i think i'll run my branch through ec2 while i eat [16:44] * barry fears the output [16:44] barry: wow, long lunch! [16:45] bac: ;) [16:45] simon-o: fcontributor-agreement@canonical.com, francis lacoste, and CC'ing salgado too [16:45] simon-o: oops, paste error, ignore the "f" on the front there :-) [16:45] bac: if i can't run the full suite on my new machine in under 15m i'm taking it back :) [16:47] kfogel, salgado: Thanks to both of you for your help :) [16:48] barry: if so i think elmo would be beating a door to costco === matsubara is now known as matsubara-lunch === salgado is now known as salgado-lunch === deryck is now known as deryck[lunch] [17:19] Anyone know of any proposals or just requests to support Gantt charts? === gary_poster is now known as gary-lunch === cprov-lunch is now known as cprov [17:36] awilkins: https://dev.launchpad.net/Wishes [17:38] Guess that fits in https://dev.launchpad.net/Wishes/ProjectManagementTools then [17:38] Thanks :-)( [17:39] yep! [17:47] maxb: are you a moin syntax guru? [17:47] moderately so, yes [17:48] maxb: I'm trying to get external hyperlinks to work inside a {{{...}}} block (see https://dev.launchpad.net/Contributions/Draft -- I'm trying to make those bug numbers into links in the commit messages). [17:48] maxb: I've got the code to linkify the bug numbers all done, but figuring out how to make something inside {{{...}}} into a live link is harder. [17:49] hmm [17:49] I have this nasty feeling that this is not going to work without a custom plugin [17:50] afaik plain {{{ }}} is "ignore all other markup inside me" [17:51] maxb: that's what I was afraid of. If I knew some other way to preserve the formatting of the log messages without using {{{...}}}, I'd do that (of course, I'm fickle and don't want any *other* wiki syntax besides my hyperlinks to be live in there). [17:52] hm. Sorry, short of writing a custom commit message formatter, I don't see any easy way of making this wkr [17:52] *work [17:53] maxb: np, thanks for checking [17:54] s/checking/reading/ [17:54] s/reading/commiserating/ === matsubara-lunch is now known as matsubara === salgado-lunch is now known as salgado === gary-lunch is now known as gary_poster === deryck[lunch] is now known as deryck [18:54] flacoste, ping. currently, it's Navigation._publishTraverse() that appends objects to request.traversed_objects, but that means only instances of a class that has a Navigation will be appended to traversed_objects (e.g. DistroMirror and Announcements are not). I was wondering if there's any reason why we do that in Navigation or if it would be OK to try and find another place for doing that [18:55] salgado: no particular reason, might be a good idea to put it into the request/publiation traversal hook [18:57] flacoste, cool, thanks. I'll see if I can do that later today [19:01] barry, bac, salgado: release planning meeting in 2 minutes [19:02] sinzui: did you send an email about it? [19:02] yes, the standup minutes as I said on the call this moring [19:02] * salgado tests skype [19:02] barry: salgado: bac: we are looking at http://people.canonical.com/~beuno/conversions.html#registry [19:24] bac, sinzui, salgado: http://pastebin.ubuntu.com/264578/ [19:33] bac: also, the person.py conflicts were not too hard to resolve, so don't worry about it. i'll do that in my branch once yours lands [19:34] yeah, they did turn out to be easy [19:34] cool. smerge makes it /really/ easy :) [20:12] Entity-body was not a well-formed JSON document. [20:15] just got that when changing the description of a bug [20:15] deryck, have you seen that before? (^) [20:16] salgado, no, I haven't. [20:16] * deryck looks back through description bugs [20:17] looks similar to bug 331990 [20:17] Bug #331990: The inline editor widget reports a JSON error when saving non-ASCII characters [20:17] but I tried again with the same description and didn't get the error [20:17] salgado, that's weird. [20:19] it does sound like the same thing Tom already fixed, though. [20:19] the multi-line editor is based of the inline editor. [20:20] indeed, but I used the same description, so I'd expect it to keep failing [20:20] deryck, I've filed bug 331990 just in case [20:20] Bug #331990: The inline editor widget reports a JSON error when saving non-ASCII characters [20:21] and attached a screenshot [20:21] I've filed bug 423924, that is [20:21] Bug #423924: Entity-body was not a well-formed JSON document when updating bug description [20:21] salgado, excellent, thanks. [20:25] gmb: https://dev.launchpad.net/GettingStartedIRCSession [20:26] gmb: (I reformatted it a bit to be more readable; I'm trying to think of where best to link to it from now.) [20:26] sinzui: do you have a preference for which bug i grab next? [20:26] * sinzui looks [20:27] sinzui: bug 421976 [20:27] Bug #421976: Update keys/wikiname pages to UI 3.0 [20:27] or bug 422974 ? [20:27] Bug #422974: Update productrelease form pages to UI 3.0 [20:27] bac: 422974 should be sane lpfv [20:28] yes, the last one may be the easiest [20:28] sinzui: ok, i'll take it and up my velocity [20:28] bac: there are other bugs about the view...I will fix them over the weekend. [20:28] sheez, that bug has 5 templates. [20:29] bac: page_title, label main_only pagetitle deletes. It will be easy. I think all the views are alreayd tested [20:29] cool [20:30] edit and add are minor test changes [20:30] * bac sets time box countdown timer to 75 minutes [20:32] bac: The delete release rules are tested in milestone and series (it's a mixin) so you need to add a short test to create_view() and verify the properties [20:33] ok [20:34] bac: I guess that means you will not see that the delete rules are broken (the series and milestones do no unsubscribe structural subscriptions) [20:35] sinzui: i'm confused. is that brokenness filed as another bug or part of my task? [20:35] bac files in 3 other bugs. I just didn't want you to see one of the problems and try to fix only one. [20:36] I will fix all three [20:36] ok [21:14] sinzui, oh noes ping. [21:16] rockstar: oh no? [21:17] sinzui, yeah, I can't figure out how to link a branch to a series (setting a development focus) [21:17] sinzui, nevermind, the user found it. [21:18] sinzui, I thought we had some royal breakage there for a second. [21:19] rockstar: yes there are really two problems. One setting a branch to the series, and the other is "This is the development focus". I started working on this by making it more prominent: http://people.canonical.com/~curtis/productseries.png [21:19] Code for this series will get both actions! [21:20] sinzui, awesome. I have three users that had no idea what the development focus was. [21:21] rockstar: Series are an enigma, and the can be used for forward planning and backward ass covering. I think we need a tutorial about them [21:27] salgado: ping [21:27] hi barry [21:27] salgado: hi. i'm so close i can taste it, but i'm having a small problem with breadcrumbs that i'm stuck on [21:27] salgado: have a few minutes? [21:28] barry, sure, what's up? [21:28] salgado: so, this page: https://code.launchpad.dev/~mark/firefox/release-0.8/+merge/1/+review?claim=name16&review_type= [21:28] salgado: the breadcrumbs under the title say "Mozilla Firefox >> +review" [21:29] salgado: but i've added code to class Hierarchy to put the reversed breadcrumbs in the [21:29] <barry> @cachedproperty [21:29] <barry> def page_title(self): [21:29] <barry> """The page title, constructed from the reversed breadcrumbs.""" [21:29] <barry> return COLON.join( [21:29] <barry> breadcrumb.text for breadcrumb in reversed(self.items)) [21:29] <barry> [21:30] <barry> salgado: but the title says "+review : Mozilla Firefox : Mark Shuttleworth" [21:30] <barry> salgado: i can't figure out why the breadcrumbs doesn't have mark as the first component but the page_title does [21:30] <barry> salgado: does this ring any bells? [21:31] <barry> salgado: i'm looking at launchpad-hierarchy.pt and can't see where that might be conditional, and i'm using the same Hierarchy.items property (i think anyway) [21:31] <barry> salgado: first question i guess is which would you consider right? [21:31] <barry> salgado: i.e. maybe the in-page breadcrumbs really should be mark >> firefox >> +review [21:32] <barry> salgado: oh, and this page: https://launchpad.dev/~name16/+karma [21:32] <salgado> barry, that page uses a custom hierarchy view (see BranchHierarchy in lib/lp/code/browser/branch.py), and I guess you're using the standard Hierarchy view to generate the title [21:32] <barry> salgado: has Foo Bar >> +karma as the breadcrumbs and +karma : Foo Bar as the title [21:32] <barry> salgado: gah! that must be it [21:33] <salgado> the breadcrumbs on that page are broken, though. someone has to fix that custom Hierarchy view [21:33] <barry> salgado: thanks. i'd seen that but didn't but 2 and 2 together [21:34] <barry> salgado: i'm not going to worry about that :) when they fix that, both will be fixed. i just need to add another adapter for @@+page-title [21:34] <barry> salgado: let me try that... thanks! [21:34] <salgado> barry, you're welcome. :) [21:36] <salgado> barry, I don't know how your adapter works, but you might want to use getMultiAdapter((obj, request), name='+hierarchy') to get the correct Hierarchy view instead of hardcoding the standard one there and redefining it for the objects that use a custom Hierarchy view [21:39] <barry> salgado: yeah, dang. just adding the browser:page to configure.zcml did not work [21:39] <mwhudson> morning [21:40] <barry> salgado: i didn't add an adapter, i just added a page_title property to Hierarchy, and then added the browser:page zcml [21:40] <thumper> morning [21:40] <thumper> can someone explain to my why the build has been broken for over 10 hours? [21:40] <mwhudson> gary_poster: did you see my buildbot merge request? [21:40] <barry> salgado: i would have thought BranchHierarchy.objects would be enough to make it work [21:41] <gary_poster> mwhudson: probably. on call, one sec [21:41] <mwhudson> ok [21:42] <mwhudson> thumper: "psycopg2.OperationalError: could not write to file "base/195483/2667": No space left on device" [21:42] <mwhudson> slaves seem to be running out of space!? [21:42] <barry> salgado: ah, there are more browser:pages to add... [21:42] <salgado> barry, it does, but I think the problem is that the Hierarchy view is hardcoded somewhere, so it's being used to generate the title of that page instead of BranchHierarchy [21:44] <gary_poster> mwhudson, yes, I got it all nice and queued up for me to read this morning and have not looked at it since. :-P looking now [21:45] <mwhudson> gary_poster: :) [21:45] <barry> salgado: joy! [21:45] <gary_poster> I also looked at the draft last night [21:47] <thumper> mwhudson: oh arse [21:48] <thumper> barry: on the new active reviews page we have approved merges shown too [21:48] <thumper> barry: you have some old ones showing [21:48] <gary_poster> mwhudson: fwiw, I contributed some bzr bits to the "contrib" directory (or equivalent, I forget the exact name) of twisted. They included a bzr push plugin, and a poller that at one time bore some resemblance to the one we are using. IIRC, neither of those had tests either. [21:48] <thumper> barry: which I think are just artifacts due to looms [21:48] <gary_poster> It might be nice in a theoretical sort of way to sometime get to the point that we are using the contributed versions for some of these things--so that our changes continue to enrich upstream, for instance. Probably obviously, I didn't do that because of ease and expedience, and that may continue to be a driving factor. [21:48] <thumper> barry: can you mark them merged if they are actually merged? [21:49] <gary_poster> mwhudson: s/twisted/buildbot/ [21:49] <gary_poster> that is, I contributed to buildbot, as you'd expect [21:49] <mwhudson> gary_poster: yes, that's a good goal [21:50] <mwhudson> gary_poster: our testfix mode voodoo seems a little special cased to us, perhaps, but maybe not... [21:50] <gary_poster> ah yes [21:50] <mwhudson> gary_poster: in any case, it needs testing in our environment before we do that :) [21:50] <gary_poster> agree [21:51] <barry> thumper: will do [21:52] <mwhudson> gary_poster: also, does this out of disk space error make any sense to you> [21:52] <mwhudson> ? [21:52] <gary_poster> mwhudson: "Carry on" link: +1 :-) [21:52] <thumper> ta [21:52] <mwhudson> gary_poster: yeah, that's hideous :) [21:55] <gary_poster> mwhudson: don't see out of disk space error--maybe started with a slash and was swallowed by IRC gods? Looked on lpbuildbot and didn't see anything pertinent at a quick glance. We have had things like that before. They usually seemed to be spurious random EC2 things. Sometimes we did identify some way that we were not cleaning up our logs, but most of the time it seemed unlikely that we were at fault [21:57] <barry> thumper: done [21:57] <thumper> barry: awesome, thanks [21:57] <thumper> barry: in which case you can reply to my email to encourage the others :) [21:58] <barry> :) [22:02] <gary_poster> mwhudson: r=gary. ship it. === salgado is now known as salgado-afk [22:11] <thumper> barry: easy _and_ fun? :-) === henrique is now known as henrique-out [22:13] <gary_poster> mwhudson: thumper may have already spoken to you. the next buildbot I think we will want is a JS buildbot, running once daily like the bzr buildbot. flacoste is the keeper of the keys of knowledge here, while mars is out. Please ping me about it when you have a moment, or go to thumper or flacoste. Note that the js tests are currently failing; having a reminder of this is the point. [22:14] <mwhudson> gary_poster: sounds like an excuse to tidy up master.cfg [22:14] <gary_poster> mwhudson: +1 [22:16] <thumper> mwhudson: skype? [22:16] <thumper> rockstar: skype? [22:17] <mwhudson> thumper: yeah, skype works better when it's running [22:20] <kfogel> jml: my other call tomorrow got moved; I can (& would like to) do our call earlier if possible. How much earlier is good for you? === matsubara is now known as matsubara-afk [22:30] <barry> thumper: okay. easy :) === henrique-out is now known as henrique [22:35] <thumper> sinzui: IRevisionCache(self.context).getRevisions.order_by(revision_date).config(limit=5) === cprov is now known as cprov-afk [22:39] <thumper> sinzui: getUtility(IAllBranches).getMergeProposalsForReviewer(person) [22:41] <mwhudson> gary_poster: in general, shouldn't it be possible for buildbot-poll to be replaced with something that listens to a statustarget we add to buildbot? [22:41] * mwhudson dislikes polling [22:58] <mwhudson> flacoste: you here? [22:58] <flacoste> mwhudson: still [22:58] <mwhudson> flacoste: do you want to talk about a windmill buildbot? [22:58] <flacoste> mwhudson: sure, i'll forward you the email from mars on this [22:59] <mwhudson> flacoste: how much longer are you going to be around for? [22:59] <mwhudson> thanks === thumper is now known as thumper-afk [22:59] <flacoste> mwhudson: i need to leave for an external dinner very soon [22:59] <flacoste> so not much actually [22:59] <mwhudson> flacoste: ok [23:00] <flacoste> mwhudson: i'll pop-in when i'm back later though [23:00] <mwhudson> flacoste: so if the mail doesn't make enough sense, i'll talk to you next week about it i guess [23:00] <mwhudson> flacoste: oh, ok [23:01] <flacoste> mwhudson: it's actually three emails [23:01] <flacoste> all sent [23:02] <mwhudson> flacoste: thanks [23:03] <flacoste> ok, i need to run [23:03] <mwhudson> bye for now [23:30] * mwhudson decamping to a cafe in town [23:33] <bac> hi sinzui [23:33] <bac> sinzui: i'm blocked on this test: http://pastebin.ubuntu.com/264677/ [23:34] <bac> sinzui: in the 'Adding a download file' i expected to get some view.errors since some required input is missing. i don't care to demonstrate uploading a file, just checking label and page_title, but i'm confused why the error processing isn't working [23:36] <bac> sinzui: the branch with the broken test is at lp:~bac/launchpad/bug-422974-productrelease [23:36] <mwhudson> biab [23:36] * bac -> dinner [23:42] <wgrant> I still can't access launchpad's +activereviews.