=== _salem is now known as salem_ === salem_ is now known as _salem [04:56] Saviq: I'm not using quilt, but maybe it works only when sources are in the same repo as packaging? === duflu_ is now known as duflu [08:00] Mirv, of course [08:45] Mirv, both branches linked to https://bugs.launchpad.net/unity-lens-applications/+bug/1271158 - if you need me to do anything else with them, pray tell :) [08:45] Ubuntu bug 1271158 in qtbase-opensource-src (Ubuntu) "Application Lens Icons are blurry, compared with home screen" [Undecided,In progress] [09:00] Saviq: thanks [09:00] I'll look at those [09:01] Mirv, you're in London, right? [09:10] Mirv, were you *still* or *already* up at 5am? ;) [09:15] dednick, hey, could I ask you to do https://code.launchpad.net/~saviq/unity8/add-doxygen/+merge/204123 [09:16] Saviq: sure [09:16] dednick, probably no need to read through the whole Doxyfile.in [09:16] dednick, but you might want to grep for @ and see if the substitutions make sense [09:17] dednick, hmm actually now that I think of it... [09:17] dednick, we probably shouldn't build the docs all the time [09:18] since it takes a few seconds [09:18] Saviq: yeah, i just saw that [09:18] let me tweak it a bit [09:18] graphvis can probably take a bit of time [09:19] or dot rather... [09:20] dednick, doxygen itself is enough [09:20] dednick, especially since it needs to pipe the qml through doxyqml, which is python [09:21] Saviq: but it needs to be built for packaging though... [09:21] dednick, yup, tweaking debian/rules to do that [09:21] ok [09:21] let's see if I did good [09:27] dednick, pushed [09:27] Saviq: ta [09:28] dednick, I just saw an issue - "/build/unity8-KAx5z5/unity8-7.84+14.04.20140129" got included in the paths, /me fixes [09:32] sil2100, automerger still ignoring that one branch... what's up with that? (https://code.launchpad.net/~marcustomlinson/unity-scopes-api/smartscopes_proxy/+merge/203943) [09:32] dednick, fixed [09:32] Saviq: haha. no, I'm in London only next week, not this week [09:32] Mirv, oh ok :) [09:32] Saviq: I expect to wake up at weird hours anyhow also next week, even if I wouldn't want to ;) [09:33] Mirv, 3h difference? [09:33] Saviq: I think only 2h. the thing is I also wake up locally at 6am normally because of the daily schedule [09:33] Mirv, right ;) [09:33] so it furthers the observed delta [09:34] Mirv, just go for drinks on Sunday, it'll help you accommodate :D [09:34] but in London the work begins at 9am (ie. 11am here) so it's 4h difference in practice [09:37] morning [09:37] hey ricmm [09:37] Saviq: so yea reading back I think daniel got the point [09:38] dednick, https://code.launchpad.net/~saviq/unity8/packaging-and-cleanup/+merge/203829 another small one [09:38] some logic needs to happen in unity8 to setup the new ipc [09:38] and then plug that in p-api [09:38] seb128, could you do a quick checkup on packaging changes for: [09:38] https://code.launchpad.net/~saviq/unity8/add-doxygen/+merge/204123 [09:38] https://code.launchpad.net/~saviq/unity8/packaging-and-cleanup/+merge/203829 [09:39] Saviq: ok [09:39] ricmm, what changed between then and now? why can't we use android's copypaste service? [09:39] android has no copy paste service [09:40] ClipboardService is an implementation in the android ubuntuappmanager [09:40] ricmm, ah! [09:40] as in, code we wrote [09:40] ricmm, that was the detail I was missing [09:40] that uses binder for ipc and manages all through the app manager [09:40] it would be the same thing to be fair [09:41] just that living in unity-mir somewhere [09:41] Cimi, forgot to put prerequisite on textSummary [09:41] Saviq, yeah!! [09:42] Saviq, done [09:43] Cimi, also, while you're at it... you could implement the See more | See less buttons there, and a "title" for it [09:43] Cimi, https://docs.google.com/a/canonical.com/file/d/0B-a_7E3tDxOgY2daVlFzaVB3ZFk/edit [09:43] Saviq, cool [09:43] Saviq, was bored of simply adding a label :) [09:43] Cimi, make the see more / see less a component [09:43] Cimi, we'll be using it in the dash as well [09:44] Saviq, ok [09:44] Cimi, I'm adding "title" to the text widget JSON [09:44] Cimi, I'd say optional [09:44] Cimi, so hide if empty [09:44] Saviq, so textSummary needs what? [09:44] column of two labels [09:45] plus see more/less [09:45] ? [09:45] Cimi, yes, and it needs to accept a modelData object for its data, not text directly [09:46] Saviq, but no array [09:46] Cimi, no, a single value, yes [09:46] Saviq, var modelData then? [09:46] Cimi, yeah, or so [09:46] Cimi, make it an interface, btw [09:46] Cimi, so PreviewWidget { property var data } [09:47] Cimi, or so [09:47] Cimi, and base all of the preview widgets off of that [09:47] adding an interface to all widgets you mean [09:47] ok [09:47] Cimi, yup, the only common prop between widgets will be data, I'd say [09:47] Cimi, also, use Text.Wrap [09:47] Saviq, any chance to try to hookup the new previews today? [09:48] although we landed columns api, should change the models to support it [09:48] mhr3, I'd rather have at least one widget first ;) [09:48] yea, that would help i guess :) [09:48] Cimi, otherwise long words will overflow / get clipped [09:49] mhr3, so Monday? I reworked a few things with the Card (+CardTool), so carousel will work now [09:50] meaning when I complete the tests and merge it ;) [09:50] and will make sure today grid is centered [09:50] Saviq, k, re columns, i'll make scope.preview() return a master preview model that will have columns role which will have per-column widget models [09:51] widget model* [09:51] mhr3, yup, sounds good [09:51] mhr3, so we'll wrap it in a stack of previews later? [09:51] Saviq, activations and stuff will go to the master object though [09:51] mhr3, fine [09:51] Saviq, still not sure where to put the stack [09:52] mhr3, mhm [09:52] Saviq, i guess the master will emit a signal when it gets a nested child [09:52] s/nested/stacked/ [09:52] mhr3, mhm [09:53] so don't think we really need to expose the entire stack to the shell [09:53] always just the active one [09:54] but maybe you don't think so? :) [09:55] mhr3, well, I don't want to destroy the inactive one [09:55] mhr3, since there can be data the user entered [09:55] true [09:55] mhr3, and they expect it there when, went "back" from the stacked one [09:57] i just find the model hierarchy pretty complex already, but well [09:57] mhr3, well, it's modeling a complex thing, so... [10:00] Saviq, ok, so scope.preview() returns PreviewStack, that contains PreviewModel, which has columns and those have PreviewWidgetModel which finally has the individual widgets [10:00] mhr3: give me a moment and I'll check [10:05] Saviq: is unity8.shell.tests.test_system_integration.SystemIntegrationTests.test_networkmanager_integration a known failure at the mo? [10:08] Saviq: http://pastebin.ubuntu.com/6848599/ [10:08] dednick, https://code.launchpad.net/~mterry/unity8/disable-nm-test/+merge/203980 [10:09] Saviq: that was just using dpkg-buildpackage [10:09] dednick, uh, what that... [10:10] :/ it was second time was building. maybe not cleaned? [10:10] Saviq: if there's been any chages to card-tool brach, push 'em up please, reading now :) [10:11] karni, no, not yet [10:11] dednick, semop is "systemv semaphore operations"... [10:11] Saviq: hm. i had to ctrl-C out. build got stuck. [10:11] dednick, try a clean one, or better yet - https://wiki.ubuntu.com/SimpleSbuild [10:12] dednick, it builds everywhere else, so... [10:13] ack [10:13] yay. an updated [possibly] not impossible to understand sbuild wiki! [10:14] dednick, oh, didn't you see it before? [10:14] Saviq, btw ok to push http://paste.ubuntu.com/6848630/ ? it doesn't exactly work, but makes it easier to fix :) [10:14] Saviq: only this on. https://wiki.ubuntu.com/SecurityTeam/BuildEnvironment [10:14] mhr3, fine [10:15] dednick, yeah, I extracted the info useful for the normal person in the SimpleSbuild one [10:16] :D [10:17] dednick, https://wiki.ubuntu.com/CrossBuilding some info about cross-building here, too [10:18] dednick, unity8 is cross-buildable through that [10:18] didrocks, maybe you could you do a quick checkup on packaging changes for: [10:18] https://code.launchpad.net/~saviq/unity8/add-doxygen/+merge/204123 [10:18] https://code.launchpad.net/~saviq/unity8/packaging-and-cleanup/+merge/203829 [10:18] Saviq: ta [10:19] Saviq, just gave a +1 to the first one [10:20] Saviq, I'm unsure about the second one, maybe ping Mirv? (did that package got renamed in 5.2?) [10:20] Saviq, well unsure in the sense of "I don't know why you have the alternative name, I didn't follow the qt packaging enough to say if that's correct" [10:21] seb128, see http://packages.ubuntu.com/saucy/qtdeclarative5-private-dev vs. http://packages.ubuntu.com/trusty/qtdeclarative5-private-dev [10:21] seb128, qtdeclarative5-private-dev in trusty depends on libqt5v8-5-private-dev [10:22] seb128, and libqt5v8-5-private-dev doesn't exist for qt >= 5.2, it's "built into" qtdeclarative5-private-dev [10:22] Saviq, ok, thanks for the pointer, makes sense now [10:22] Saviq, +1 that one as well [10:22] seb128, thanks [10:23] Saviq: dpkg not generating unity8-doc package [10:24] dednick, :/ [10:24] need dh_install rule as well for doc? [10:24] dednick, https://jenkins.qa.ubuntu.com/job/generic-mediumtests-builder-trusty-amd64/2792/artifact/work/output/ [10:24] dednick, jenkins disagrees [10:24] dednick, building from the vcs linked previously? [10:24] dednick, you have the new .install? [10:24] Saviq: it's building, but not generating the package [10:25] Saviq: no, just using dpkg still. Working on sbuilder env now [10:26] seb128: yes Qt's V8 is no more, it's replaced by V4 insside qtdeclarative. [10:26] Saviq: hm, actually, it's not giving me autopilot either.... [10:26] Saviq: but, qtdeclarative5-private-dev 5.0.2 depends on libqt5v8-5-private-dev [10:26] nowadays, to avoid that need [10:26] Mirv, not in saucy [10:26] Mirv, that change is for saucy [10:26] Mirv, https://code.launchpad.net/~saviq/unity8/packaging-and-cleanup/+merge/203829 [10:26] dednick, seems like you don't build arch all packages? [10:27] Saviq: aha.. a stable release backport, no PPA usage? the SDK Release PPA has qtdeclarative for saucy too that has that. [10:27] seb128: er. dpkg-buildpackage -us -uc -B ? [10:27] dednick, drop -B [10:27] (and precise/quantal/raring) [10:27] -B not [10:27] hm [10:27] dednick, -B is the issue indeed [10:28] Saviq: https://launchpadlibrarian.net/159906274/qtdeclarative-opensource-src_5.0.2-6ubuntu4_5.0.2-6ubuntu5~saucy1~test1.diff.gz [10:28] ah, no arch-independant [10:28] Mirv, hmm we require the sdk ppa anyway, maybe that change isn't needed then [10:28] doh [10:28] 3rd times a charm! [10:28] Saviq: yes we require SDK PPA if it's about developers [10:29] Mirv, ok, /me undo's [10:29] thanks [10:29] Mirv, thanks for reviewing it ;-) [10:30] seb128, Mirv, done [10:32] Saviq: sorry, too many fuzz here [10:32] thanks seb128 for the review :) [10:32] didrocks, yw! [10:32] didrocks, no worries, got many eyes on it already :) [10:33] approved [10:33] Mirv, seb128, thanks! [10:33] dednick, pushed OPTIONAL for the install of doc, so that cmake doesn't complain if you didn't build it [10:40] Saviq, interface https://code.launchpad.net/~unity-team/unity8/unity8.previews_audioPlayer/+merge/203387 [10:41] Cimi, 55 + model: data [10:41] Cimi, that's wrong [10:41] Cimi, you will get the whole object as described in the docs [10:41] Cimi, so you need data["tracks"] [10:41] damm [10:42] Cimi, also, be explicit - name the top level item [10:42] Cimi, and reference data by that name [10:42] root.data [10:42] ok [10:42] Cimi, yup, something like that [10:42] Saviq, data["tracks"] will contain the model then? [10:43] so I'll have to redo the mocks? [10:43] Cimi, yes [10:43] Cimi, well, not "redo", just wrap them in { "tracks": [] } [10:43] Saviq, got a link handy [10:43] of the json? [10:44] Cimi, https://docs.google.com/a/canonical.com/document/d/1NmiM4UCnJgf6IEawmfyTOHRNAA5ZGrqpyrPqPOibwc8/edit#heading=h.6fng39a8onr [10:44] Saviq: does doxygen not generate for the qml files? [10:44] dednick, it does, when they're documented [10:44] dednick, Carousel should be there [10:44] Saviq: ahh [10:44] dednick, I've hidden non-documented files, thought it didn't make sense otherwise [10:45] Cimi, 405 + audioPlayer.stop() I don't think that's needed - and if it would be, then it needs to be part of the interface [10:45] Cimi, or maybe not.. [10:45] Saviq, currently [10:45] Saviq, on current previews [10:45] Saviq, there's a stop when the preview isCurrent turns falser [10:46] Saviq: they're not listed under 'Files', only 'Classes' [10:46] Saviq, since this is no longer top level preview but a widget, thought of adding a function [10:47] dednick, yeah, that seems a limitation of doxyqml [10:48] dednick, or well, it doesn't add .cpp files either [10:48] dednick, I'll have a chat with agateau (maintainer of doxyqml) [10:48] Cimi, I'm just thinking what would be calling it [10:49] Saviq, the preview page [10:49] Saviq: approved. [10:49] Cimi, yeah, but that would basically mean that the preview should go away, so the player would get destroyed anyway [10:49] Saviq, do they go away? [10:49] Saviq, or they are listviews? [10:50] Cimi, there's only 3 in a ListVIew [10:50] Cimi, so if you move away by more than one [10:50] Cimi, it gets destroyed [10:50] Cimi, only use case I can think of is when you get a preview on top of your current one - but whether it should pause I'm not sure [10:50] Saviq, but you want to move one, not two [10:50] Saviq, you switch album [10:50] and it should stop [10:50] Cimi, yes, I know [10:51] Cimi, just wondering whether it should really get destroyed or not [10:51] Cimi, anyway, so let's say PreviewWidget gets a new property [10:51] Cimi, "focused" [10:51] ok [10:51] Cimi, and preview widgets react to that [10:54] sil2100, so what's the news on the branch? [10:54] sil2100, you gave us a silo and that mp would like to be landed [10:57] mhr3: ok, so I poked fginther about it, since it seems it's something stranger [10:57] mhr3: oh, and we just recently published some silo, right? [10:58] Saviq, data seems to cause troubles as a name [10:58] var name [10:58] can I change it to something else? [10:58] previewData ? [11:02] sil2100, mhall119, the MP is failing to merge because jenkins is complaining that it has an unapproved pre-requisite, but that's not the case (it's actually merged). Will need to dig more. [11:02] mhr3: ^ [11:02] ;) [11:02] fginther: thanks! [11:03] fginther, so i'll just remove the pre-req since it's merged [11:04] * mhr3 really wants it in [11:05] Cimi, right, of course [11:05] Cimi, widgetData [11:05] Cimi, "data" is an internal QML name [11:06] Cimi, sorry about that [11:07] mhr3, can I re-approve that MP? the last push to merge to trunk does not match the approved version. [11:08] fginther, i resubmitted it a minute ago [11:08] and approved [11:09] so should be good afaict [11:10] mhr3, ack. I'll watch it in case it doesn't start [11:15] Cimi, you could add some docs - we'll be generating them soon enough [11:15] Cimi, see http://bazaar.launchpad.net/~saviq/unity8/card-tool/revision/677 for an example === _salem is now known as salem_ === alan_g is now known as alan_g|dentist [11:27] /food [11:29] i think you client does not suppor that [11:29] *your [11:34] greyback, hey. it unity8 already using the mir socket and protocol as a general purpose mechanism to exchange information with apps (and vice versa)? [11:34] s/it/is [11:41] Saviq: FYI I'm on app/sdk sprint next week in Orlando, my presence may vary [11:57] karni, ok [11:57] karni, will you get us an updated look at your guys' priorities? [11:58] Saviq: just noticed something interesting on my phone. flashed trusty-proposed, added demo-stuff ppa, rebooted, still see old scopes [11:58] Saviq: yes, will do [11:58] karni, apt-cache policy unity8? [11:59] anpok, ;) [11:59] Saviq: looks correct http://paste.ubuntu.com/6849035/ [12:00] karni, no it doesn't, there's a newer unity8 from distro than there is from ppa [12:00] oh, indeed [12:01] didrocks, deps between landed packages not really handled? https://launchpad.net/~ci-train-ppa-service/+archive/landing-009/+packages [12:01] hrm [12:01] karni, probably because I didn't merge trunk into new-scopes [12:01] karni, pushed, kicking build now [12:02] Saviq: thank you, sir! [12:02] didrocks, what will i have to do once the dep pkg publishes? [12:04] mhr3, click "merge & clean" [12:05] mhr3, and for the deps, just click "retry this build" [12:05] mhr3, oh no you can't [12:05] didrocks, ↑ valid issue [12:05] yea, i can't [12:06] the good news is that the build today was about 20x faster than last time [12:07] mhr3: it's in dep wait [12:07] didrocks, yes, and the dep built now [12:07] mhr3: ah, you have to wait for a cronjob IIRC in launchpad [12:07] it's not triggered by signals) [12:07] but launchpad is handling that [12:07] didrocks, does that really work in ppas? [12:08] AFAIK, it does [12:08] didrocks, yeah, I've never seen "dependency wait" to switch to "needs build" by itself [12:09] let's see, then... [12:09] Saviq: just trying to understand unity-mir at the mo. how does mir know where to draw the surface? ie how does it know its position given a stage? Can't find unity telling it anywhere... [12:09] dednick, currently unity-mir has it hardcoded [12:09] Saviq: mhr3: you can ping wgrant about it (I'm ddosing on pings right now) [12:09] Saviq: ahh. i c. [12:09] ;D === MacSlow is now known as MacSlow|lunch [12:10] thought i was going a bit mad [12:10] didrocks, k [12:10] @unity: any branches you guys would like landed? There's five in queue currently [12:11] elopio, can you please fill checklist on https://code.launchpad.net/~elopio/unity8/generic_preview/+merge/202793 [12:11] mzanetti: did you take a look at https://code.launchpad.net/~nick-dedekind/unity8/plugin-qml/+merge/203810 ? [12:12] dednick, mzanetti's refusing to approve branches before CI goes green again ; [12:12] ;) [12:12] ahha. dang [12:12] then no, i guess [12:12] dednick, which it will with this lannding [12:12] Saviq: which row of landing spreadsheet where you referring to? [12:12] Saviq, no [12:13] * karni doesn't see unity8-fix-mascot in the sheet [12:13] karni, none, yet, just https://code.launchpad.net/~unity-team/unity8/trunk/+activereviews [12:13] ah [12:13] dednick: I looked at it, but CI seems still red [12:13] cool [12:13] mzanetti: no worries, no rush on that one [12:16] sil2100, I can has silo for unity8? [12:17] Saviq: let me look at that [12:17] sil2100, thanks [12:18] karni, I'll have patched (blurriness) qtbase packages for you in a few minutes [12:18] Saviq: w00t! [12:23] karni, new unity8-newscopes built in demo-stuff for amd64/i386, soon there for armhf [12:23] Saviq: so, we have a limited number of silos right now, so it might take a while at most - we're trying to bump the number of silos anyway, but it's something not for today [12:23] Saviq: very neat, very neat :) [12:23] sil2100, so no spot for us yet? ;( [12:24] sil2100, 002, 007 and 010 look empty? are they queued for someone or waiting for cleanup? [12:24] sil2100, not complaining, just trying to get more data [12:27] thostr_, "When you allocate a silo just go ahead and hit build at the same time. Thanks!" delegating already, nice one! [12:28] Saviq: ? [12:29] Saviq: the comment. yes, I think that is a pattern we should apply [12:29] thostr_, indeed, no point in assigning a silo without building, agreed [12:29] Saviq: at least I started this and e.g. initiated bfillers requests, so when he starts his day he should have all his packages built [12:30] thostr_, yup [12:47] Saviq: so, I had a talk with Didier, and I would be grateful if you could wait with silo assignment for unity8 for now [12:48] Saviq: just a little bit ;) [12:49] sil2100, ok, can you let me know before you'd assign them in case we'd like another branch to get in? [12:49] Saviq: sure! [12:49] dednick, you have some time - merge the test fix and resubmit your branch with prerequisite if you want it in! ;D === dandrader is now known as dandrader|afk [12:56] aargh stupid optipng ;/ [12:57] Mirv, you really need to disable OPTIPNG for qt packages... it's a half hour already here that it's built and just mangling already-optimized pngs... [12:57] Mirv, or something... [12:58] * Saviq finds a way to disable optipng system-wide [12:59] Saviq: card-tool -- looking good :) let me know when you want further feedback/when it goes out of WIP [12:59] karni, yeah, I just need to add some tests, will happen soon (I hope) [13:00] very cool === alan_g|dentist is now known as alan_g|lunch [13:01] Saviq: it could be similarly temporarily disable to how I'm temporarily enabling debug options [13:02] Saviq: I know I've bookmarked your disabling trick _somewhere_ :) [13:02] Mirv, I just feel like it's wasting so much builders' time [13:02] Mirv, export NO_PKG_PNG_MANGLE=1 [13:02] export NO_PNG_PKG_MANGLE=1 [13:02] PKG_PNG vs. PNG_PKG :/ [13:05] sil2100, more train questions [13:05] Saviq: I already did: https://code.launchpad.net/~elopio/unity8/generic_preview/+merge/202793/comments/475007 [13:05] elopio, yeah, saw that [13:05] elopio, please put submitter checklists in description [13:05] elopio, reviewer in comments, as there may be more [13:05] elopio, I copied it already [13:05] Saviq: ok. Thanks. [13:06] sil2100, the build failed, do i need to check any of the options now when i try to rebuild? === dandrader|afk is now known as dandrader [13:10] Saviq, it's fine now https://code.launchpad.net/~unity-team/unity8/unity8.previews_audioPlayer/+merge/203387 [13:11] Cimi, \brief not needed if only one line [13:11] ok [13:11] mhr3: no, you can specify what specific packages you want to rebuild but you don't have to [13:12] done [13:12] Cimi, no tests for subtitle, but otherwise looks good, I'll leave to Albert to review properly on Monday [13:12] Saviq, how can I test subtitle? [13:12] oh sorry [13:12] sil2100, well, i need to rebuild the one that succeeded cause that broke the one that failed :) [13:12] subtitle I think I tested it [13:13] Cimi, not there in mocks? [13:13] Cimi, wrong, it's there [13:13] Cimi, sorry, it's good [13:13] Saviq, time is not tested [13:13] Saviq, not sure how I can test it [13:13] Saviq, because it's processed [13:13] from seconds I have a string [13:13] Cimi, well, yeah, you should supply one value and check that the string is what you expect? [13:13] sil2100, so i just click the rebuild without checking any of the options? [13:14] Cimi, that's the easiest thing to test :) - input, output [13:14] Cimi, have a separate test function for it [13:14] Saviq, yes but that will mean adding an extra field to the mock [13:14] mhr3: yep :) [13:15] I wanted to keep the mock like the one it will receive [13:15] but I can do it [13:15] mhr3: it will then take all the components, fetch latest revisions of the branches and build [13:15] Cimi, wait [13:15] Cimi, no, no more fields in mocks, why would you need it there? [13:15] Saviq, I don't know wehre [13:15]