[08:10] Cimi: seen email about left/center align? [08:11] tsdgeos, yup [08:12] oki [08:12] tsdgeos, still summer in barca? [08:12] not really [08:13] we got quite heavy rain yesterday [08:13] tsdgeos, is quite hot in london [08:13] tsdgeos, I was wondering how was in warmer places [08:13] 19 degrees at the moment [08:19] tsdgeos, pushed! [08:21] Cimi: cool [08:21] Cimi: scopes more things on memory got qmluitests passing at last [08:21] Cimi: can you re-review? [08:41] tsdgeos, sure === seb128_ is now known as seb128 [08:54] tsdgeos: hey, can you take over this one from Gerry? https://code.launchpad.net/~mzanetti/unity8/async-dashcommunicator/+merge/235272 [08:54] tsdgeos: should be mostly ok now.. I.e. we did already a few rounds of review back and forth [08:55] let me know if you want me to explain why this ended up so complicated [08:56] * tsdgeos read [08:56] s [08:56] How do I make an application short cut on the desktop? [09:00] mzanetti: why moved the visibility thing? [09:01] tsdgeos: hmm... dunno tbh... that was gerry :D. I assume because of the lib [09:01] tsdgeos: yeah. I guess we should only have this for qml plugins, but not for the lib [09:01] tsdgeos: but then I don't know exactly [09:02] want me to investigate? [09:02] tsdgeos: so at some point I merged a branch from gerry that added the libunity8-private with the AbstractDBusMonitor [09:02] as he was kind enough to prepare that for me [09:02] this was in there [09:03] mzanetti: yeah i mean we have more code that was previosuly covered by the visibility [09:04] want to know why we need the move [09:04] and if we really need to remove it that much [09:04] yeah [09:07] mzanetti: should DashCommunicator destructor delete m_dashConnection ? [09:09] i guess it's one of those long lived things that doesn't really matter in practice [09:09] but just to make it "correct" [09:09] and it's not hard [09:12] tsdgeos: yeah... will fix. also realized it needs merging with trunk [09:21] mzanetti: ok, now explain me what the branch is for : [09:21] :D [09:21] tsdgeos: basically the issue is that the dash was never suspended. so I fixed that in lp:qtmir. [09:21] i see dash never being suspended as a good thing :D [09:21] tsdgeos: that caused the issue though that unity8 hangs when it tries to send something through dashcommunicator and the dash is sleeping [09:21] tsdgeos: its not [09:22] tsdgeos: so.. dashcommunicator needs to be async [09:22] am i speaking to the guy that don't like suspending stuff?? :D [09:22] j/k [09:22] no but the dash is actually the least useful one in terms of background activity [09:23] tsdgeos: and it does cause about 1% CPU per favorited scope [09:23] tsdgeos: anyhow... dashcommunicator needed to be async [09:23] that's something we should fix in favorited scopes [09:23] tsdgeos: the issue now is that its not enough to use asynCall() because that would solve the issue [09:23] when its calling [09:23] right [09:24] but not when its connecting [09:24] so the whole createdbusconnection thing needs to be threaded [09:24] in case it tries to connect to a suspended dash we at least only hang the thread, but not unity [09:24] tsdgeos: yes, I agree we should improve in the dash itself too [09:25] tsdgeos: still I guess we can't always as there is 3rd party stuff in there [09:25] sure [09:25] so if a backend keeps on refreshing things we'd paint them [09:26] tsdgeos: also see the related branches [09:26] they allow waking up the dash on left edge swipe etc [09:29] i'm a bit "concerned" by eitehr the dashActive naming [09:29] or its valu [09:29] e [09:29] why? [09:29] i mean dashActive = launcher.shown || launcher.dashSwipe [09:30] yes..? still don't see the problem [09:30] looks like it should be active in more cases no? [09:30] its also active when its focused [09:30] but other then that, no, I *think* that's it [09:30] right, so you're using dashActive as "activateDash" function [09:30] yeah [09:31] sort of [09:31] that's why i complain about the name [09:31] when its focused [09:31] but its a property [09:31] it'll be set to false [09:31] which is weird [09:31] tsdgeos: yeah... that's weirdness caused by the cat that we still handle suspension tied to focusing in qtmir [09:31] tsdgeos: number 1 prio to fix after RTM for us AppManager guys [09:32] s/cat/fact/ :D [09:32] mzanetti: well maybe it should not be a property but a function? [09:32] i mean it's always dash -> qtmir no? [09:32] it can be requestActivation() or something? [09:32] requestDashActive [09:32] tsdgeos: meh [09:32] then I need to call shit in many places [09:32] or just turn the Bindiing {} into a handler that calls stuff when a intermediate property changes [09:33] well, either that or you add a huge ass warning saying [09:33] "dashActive" is not really when the dash is active [09:33] tsdgeos: there is documentation in unity-api what this property is for :D [09:34] mzanetti: call it forceDashActive ? [09:34] tsdgeos: ok... can do [09:34] it's more really waht it does, no? [09:35] yeah... you're probably right :) [09:35] thanks :) [09:35] mzanetti: i guess you need someone to review https://code.launchpad.net/~unity-team/qtmir/fix-some-lifecycle-bugs/+merge/234087 too? [09:36] oh... didn't see he set it to needs fixing again [09:36] it was approved already [09:36] tsdgeos: will fix first and let you know [09:36] oki [10:00] noooooooooooooooooooooooooooooooooo [10:00] make testDefaultIndicatorPage [10:00] doesn't compile! [10:00] dednick: ↑↑↑ [10:01] tsdgeos: compile? [10:02] tsdgeos: what branch? [10:02] dednick: trunk [10:02] tsdgeos: erm [10:02] yes? [10:02] that sounds odd [10:02] it does [10:02] :D [10:02] and i dont get "compile". you mean the test doesnt succeed? [10:03] dednick: the test can't start because there's broken qml that can't be compiled [10:03] tsdgeos: something about messaging menus? [10:04] /home/tsdgeos_work/phablet/unity8/unity8/tests/qmltests/Panel/Indicators/tst_DefaultIndicatorPage.qml:32,9: Cannot assign to non-existent property "contentActive" [10:04] dednick: ↑ [10:05] tsdgeos: ok. give me a sec [10:06] tsdgeos: does CI not run qmltests? [10:06] dednick: it does [10:06] that's why i found its failing [10:06] s/why/how [10:06] tsdgeos: dont know how it landed then [10:07] CI has nothing to do with landing [10:07] we land broken stuff all the time [10:07] tsdgeos: silo dont run tests? [10:07] ah [10:07] dednick: afaik no, no tests other than make check/test is run in silos [10:08] tsdgeos: ok, well i found the problem. give me a sec and i'll sort a branch [10:09] dednick: cool [10:28] tsdgeos: https://code.launchpad.net/~nick-dedekind/unity8/broken-indicator-tests/+merge/236292 [10:36] tsdgeos: updated all 3 MPs related to suspending the dash [10:36] tsdgeos: kicking the ppa (phone-right-edge) for a test run [10:37] oki [10:40] please vote for this :) https://bugs.launchpad.net/ubuntu/+source/lxc-android-config/+bug/1367683 [10:41] tsdgeos, a couple of lines to edit [10:41] tsdgeos, one indent and an updated copyright year [10:41] tsdgeos, I would like to have this branch in ASAP [10:53] dandrader: https://code.launchpad.net/~dandrader/unity8/touchOwnership/+merge/236152 fixes the "press on stuff that is too close to the edge" thing? [10:53] tsdgeos, yes [10:53] cool [11:01] Holas [11:05] pstolowski: any idea why https://launchpadlibrarian.net/186037191/buildlog_ubuntu-utopic-i386.unity-scopes-shell_0.5.4%2B14.10.20140925-0ubuntu2_FAILEDTOBUILD.txt.gz failed? [11:05] looking [11:05] qmlplugindump: could not find a Qt installation of '' [11:05] ah [11:06] it may be me [11:11] tsdgeos, ah, good you spotted it, i was scratching my head [11:11] tsdgeos, how about this for the image gallery slideshow? http://paste.ubuntu.com/8454572/ [11:11] boooooooooooooooo [11:11] that ppa doesn't have arm [11:12] tsdgeos, right.. i remember mhr3 asked somebody to enable arm for his ppa [11:12] Cimi: screenshots seems like the wrong text === mpt_ is now known as mpt [11:21] tsdgeos, yeah, we need something better [11:22] tsdgeos, you know how we can pass the information down the image gallery? [11:22] Cimi: why you need the text? we didn't have it, why the need now? [11:22] tsdgeos, which text then? [11:22] Cimi: none? [11:23] tsdgeos, we need to fill the header with something no? [11:23] tsdgeos, application name? [11:23] Cimi: why do we need a header? [11:23] is it in the spec? [11:23] tsdgeos, they said app name [11:23] iirc [11:24] tsdgeos, none is actually quite good too [11:24] but that's not for apps only [11:24] tsdgeos, I'd go for none [11:24] ebay has gallery of images [11:24] go for none [11:24] easier === MacSlow is now known as MacSlow|lunch [11:25] Cimi: which indent do you want in that "indent" comment you made? [11:26] i'd have it in a single line [11:26] but there's people that complain [11:26] so i just wrapped it arbitrarily [11:26] there's no "proper" indent imho [11:31] right, no line should be longer than 120 chars, says our coding style [11:32] dednick: Did you make sure that your branch does not contain spurious tags? [11:32] dandrader: can you point me to it? i always forget where it is [11:33] tsdgeos, ouch, I don't care then [11:33] tsdgeos, I would have the "- " under the element on top [11:33] tsdgeos, but nevermind [11:33] Cimi: but see it's part of the same thing [11:33] it's not after the , or anything [11:33] or even the - on top l;ine [11:34] it's part of the same "expression" [11:34] Cimi: i'll put it wherever you want [11:34] keep Cimi happy :) [11:34] tsdgeos, lp:canonical-client-development-guidelines [11:35] tsdgeos, you have to build it in order to view it :D [11:35] awesome documentaiton \o/ [11:35] not [11:36] heheheheh, yeah, it's crazy. I think there's some URL around that shows the built result but I don't know it [11:36] it's essentially a lightly modified version of google's C++ guide [11:37] and there's nothing on QML [11:38] tsdgeos, you so nice on monday morning, thanks :)) [11:39] tsdgeos, I'd put the - on the top line [11:40] tsdgeos, and my personal taste could have cagoryview under baseItem, since they are part or Math.max [11:40] but that's me :) [11:40] dednick: you here? [11:41] Cimi: ok [11:41] Cimi: there's a broken qmluitest now but that's on trunk too [11:41] meh... I'd like to go for a run... but can't start building the silo [11:42] wrong channel :) [11:42] mzanetti: why can't start building? [11:42] because building the silo makes a copy of the code [11:42] and when merging that copy will be merged, [11:42] which would import tags [11:43] ok [11:43] mzanetti: fork it! [11:43] and put it in unity-team [11:43] :) [11:44] tsdgeos: https://bugs.launchpad.net/ubuntu/+source/oxide-qt/+bug/1375215 [11:44] Ubuntu bug 1375215 in oxide-qt (Ubuntu) "White screen in webbrowser and webapps / system runs out of memory and kills oxide-renderer" [Critical,Confirmed] [11:44] tsdgeos: seems its unity8-dash [11:45] meh [11:46] tsdgeos: mind reapproving? https://code.launchpad.net/~unity-team/unity8/broken-indicator-tests/+merge/236304 [11:47] mzanetti: done [11:50] Hey all. we're an ISV and getting customer complains, that our tray icon is missing from unity since 14:04 [11:51] it seems that com.canonical.Unity.Panel/systray-whitelist ceased to work [11:52] bregma: are you guys still the ones in charge or unity7? ↑↑↑↑ [11:53] and are you awake already? [11:53] :D [11:53] tsdgeos, yep, that's us [11:53] bregma: see sebastian_'s question [11:53] sebastian_, the whitelist was removed in 12.10 and has not been available for years [11:54] bregma: well, that doesn't help us. what can we do about that? [11:55] sebastian_, the "tray" was removed more than 4 years ago, applications that want to integrate in the menu bar need to write to the application indicator spec available on freedesktop.org and supported by most major desktops [11:56] sebastian_, here's an older wiki entry on the topic: https://wiki.ubuntu.com/DesktopExperienceTeam/ApplicationIndicators [11:58] is there some kind of wrapper, that wrapps the old tray icons to ApplicationIndicators? [12:01] tsdgeos, I am trying to implement a zoom in effect when I tap on images [12:01] well the main problem ist, that we're based on Qt4 and it's QSystemTrayIcon implementation doesn't support ApplicationIndicators [12:02] tsdgeos, there is something wrong in my code... [12:02] tsdgeos, http://paste.ubuntu.com/8454940/ [12:03] tsdgeos, I am using "red" to show a dot, it is in the top left of each image in the gallery, but when I am using it as origin.x and .y of a scale transform, something behaves wrong [12:03] tsdgeos, how did you do in the overview? [12:08] sebastian_, it's possible a patch for that got dropped with the switch to Qt5, there may be a bug..... [12:11] mzanetti: ya. i did [12:11] mzanetti: hm... i know i did! [12:12] I will use scale and transform... [12:13] oh. maybe i didnt scroll all the way to top :/ [12:17] bregma: ok. then, is there a sample implementation that also works wihtout an installed unity desktop? [12:17] sebastian_, I'm poking some of the upstream developers to find out what's going on -- some of them are not online yet [12:19] bregma: thanks! I'll stay here at least for the next three hours. === MacSlow|lunch is now known as MacSlow [12:50] bregma: sni-qt ? [12:50] bregma: it has no qt5 version afaik [12:52] yep, wonder how that could have gotten missed? [12:53] no idea [12:53] https://bugs.launchpad.net/sni-qt/+bug/1347128 [12:53] Ubuntu bug 1347128 in owncloud-client (Ubuntu) "Owncloud client indicator has wrong position - absolutely on the left" [Undecided,Confirmed] === karni is now known as karni-lunch [12:56] dednick: no worries. I've resubmitted it [12:56] dednick: to be able to kick the silo build [12:57] mzanetti: yup. i saw. thanks [13:07] robotfuel: hey there, do you know where the jenkins scripts are that auto land stuff ? [13:08] specifically if we wanted to alter unity8 landing to include a test [13:08] kgunn: that's cupstream2distro-config [13:09] kgunn: https://code.launchpad.net/cupstream2distro-config [13:09] mzanetti: ^ [13:10] robotfuel: who owns that if we we'd like to propose a change ? [13:10] we can't run qmluitests there [13:10] is that you? [13:10] i mean we can [13:10] but it needs X and stuff [13:10] or xvfb-run at least [13:10] tsdgeos: no... make qmltests runs in xvfb [13:10] may not be trivial [13:10] kgunn: ci [13:10] tsdgeos: but yeah, that's the story that triggered this request :) [13:11] robotfuel: ack [13:12] mzanetti: so we'd need to work with fginther === tedg is now known as ted === dandrader is now known as dandrader|afk === _salem is now known as salem_ === karni-lunch is now known as karni === dandrader|afk is now known as dandrader [13:56] sebastian_, it looks like Qt5 integration into the Unity indicators should be accomplished by the appmenu-qt5 package -- can you check to see if that is installed and if not, install it and see if it resolves the problem? [14:01] sebastian_, you may need to set QT_QPA_PLATFORMTHEME=appmenu-qt5 [14:03] pstolowski: build fails on arm :S https://launchpadlibrarian.net/186048336/buildlog_ubuntu-utopic-armhf.unity-scopes-shell_0.5.4%2B14.10.20140925-0ubuntu5_FAILEDTOBUILD.txt.gz [14:04] cd /build/buildd/unity-scopes-shell-0.5.4+14.10.20140925/obj-arm-linux-gnueabihf/src/Unity && UNITY_SCOPES_NO_LOCATION=1 UNITY_SCOPES_LIST_DELAY=10000 /usr/bin/qmlplugindump -notrelocatable Unity 0.2 /build/buildd/unity-scopes-shell-0.5.4+14.10.20140925/obj-arm-linux-gnueabihf/src/Unity/../ > /build/buildd/unity-scopes-shell-0.5.4+14.10.20140925/obj-arm-linux-gnueabihf/src/Unity/plugin.qmltypes [14:04] Error: SEGV [14:04] Error: SEGV [14:04] qemu: uncaught target signal 11 (Segmentation fault) - core dumped [14:04] mzanetti: kgunn: any idea? just retry ad infinitum? ↑↑ [14:05] tsdgeos: not following [14:05] mzanetti: have you ever seen a ppa fail on qemu for arm? [14:06] tsdgeos, uhm, any more details?.. [14:06] tsdgeos: ah... qemu segfaulting sounds like its trying to instantiate a QQuickView [14:06] tsdgeos: won't work [14:06] so can't build unity-scopes-shell in a ppa? [14:06] tsdgeos: I reported a bug for that when I was still in QA :D [14:06] tsdgeos: you can, but the ppa needs to use real builders, not qemu [14:06] * tsdgeos does the unhappy dance [14:06] tsdgeos: your personall ppa won't work, but you can use the phone-right-edge one [14:07] or mir staging [14:08] kgunn: may I request a second unity-testing ppa? :) seems we're using this quite a bit [14:08] kgunn: somewhat low priority though [14:08] mzanetti: but you're using phone-right-edge, no? [14:09] tsdgeos: yeah... but there is no scopes in there... [14:09] I guess we can mix it in this case [14:10] mzanetti: well i also need unity8 [14:10] tsdgeos: then yeah... need to wait a little [14:10] tsdgeos, before I jump into tests... do we need to add any smoothing or cache/loaders here? http://paste.ubuntu.com/8458732/ [14:10] kgunn: ok, so no bottom list ppa then [14:11] tsdgeos: my stuff should be built soonish, then we test it asap and switch to your stuff [14:11] Cimi: can you push it to a branch? it's easier to get :D [14:11] ahah ok [14:12] tsdgeos: mzanetti alternative, stuff it in a silo mark it test....easier to keep up to date imho [14:13] kgunn: yeah... if we're always getting so many silos... seems a bit out of scope use though [14:13] tsdgeos, lp:~cimi/unity8/preview-image-gallery_slideshow [14:15] tsdgeos, i've updated my branch with tests + one minor fix [14:15] oki cool [14:16] Cimi: can you merge it with trunk? [14:16] it doesn't compile! [14:16] tsdgeos, sure [14:16] @_@ [14:17] tsdgeos, done === dandrader_ is now known as dandrader [14:23] mzanetti, kgunn, is it a known issue that the orientation sensor remains on while the display is off? do we have a bug for it? like is not being turned off in qtmir [14:24] tsdgeos: mzanetti keeping a silo busy is not a big deal for big ticket stuff...but up to you, seems like phone-right-edge ppa gets us what we want...save maybe building on rtm [14:24] dandrader: is that on n4 ? [14:24] dandrader: not known to me... please report a bug [14:25] kgunn: yeah... we use it mostly to test stuff with inter-project-dependencies before approving [14:25] kgunn: sometimes also to give other's a way to test some not-landed feature [14:25] kgunn: I think we should get more of them. at least one more [14:25] Cimi: looks good ui wise (well the back thing looks a bit weird but can't suggest anything better) [14:26] maybe call it unity-testing instead of some-feature-from-last-year this time :) [14:26] tsdgeos, which black thing? [14:26] tsdgeos, header? [14:26] Cimi: "back" thing [14:26] sorr [14:26] tsdgeos, or the rectangle box? [14:26] y [14:26] but let's not argue about that now [14:26] do the tests [14:26] and we review+merge [14:26] and let others complain if they don't like it later [14:26] rsalveti, how would you change? [14:27] rsalveti, sorry [14:27] Today I tried to update my nexus 4. And adbd doesn't want to run. It doesn't export the adb interface. Any way to recover it or I need a totally wipe from fastboot? [14:27] tsdgeos, ? [14:27] kgunn, seem it in krillin, but should happen with any device [14:27] i don't know how i would change it, just saying it feels a bit strange [14:27] my immediate feeling was "tap to close" [14:27] but that is probably done nowhere [14:27] in the rest of the ui [14:27] tsdgeos, maybe I can put an X to close [14:28] so not good [14:28] Cimi: i think i'd like an X more [14:28] tsdgeos, are you aware of an X in the icons for the toolbar? [14:29] "back" [14:29] err [14:29] "close" [14:29] sorry :D [14:29] tsdgeos, can you take a look at this trivial fix (for the problem you spotted last week)? https://code.launchpad.net/~stolowski/unity-scopes-shell/fix-delete-later/+merge/236349 ; btw, the test i added fails without that fix [14:29] back is < [14:29] close yeah [14:29] close is X [14:29] tried [14:29] works [14:30] pstolowski: sure [14:31] tsdgeos, on my question for smooth and loaders? [14:31] tsdgeos, which settings we want for the image? [14:31] tsdgeos, I don't remember how you did your tests on sourcesize and that option [14:32] Cimi: not sure i get what you mean, let's do the standup and come back [14:32] ok [14:35] bregma: that's a solution for Qt5. there is no such thing for our current stable release based on Qt 4? [14:36] tsdgeos, thanks! [14:37] bregma: ahh, I'll have a look at https://launchpad.net/appmenu-qt [14:37] sebastian_, the Qt4 solution was sni-qt, which should still be working [14:42] tsdgeos, up ^ :) [14:42] Cimi: so [14:43] what do you mean for smooth and loaders? [14:43] tsdgeos, if we need smooth true in the slideshow delegates [14:43] bregma: ok, thanks [14:44] or antialiasing [14:44] I am not 100% sure on those [14:44] smooth true maybe yes [14:44] Cimi: smooth is on by default [14:44] tsdgeos, ok, delete me! [14:44] :) [14:45] Cimi: so that's all? [14:47] tsdgeos, well, not sure I should use a loader for this slideshow or not [14:47] tsdgeos, I would avoid delayes on the zoom animation when you open it [14:48] Cimi: it's a listview [14:48] listviews are async already [14:52] Cimi: it feels good here [14:52] Cimi: and Image also has another async in case you want to set it to true [14:52] tsdgeos, nah, I don't want to slow down [14:53] tsdgeos, I mean, seing black boxes [15:00] dednick, the plan is that we dismiss a prompt session once its "host" session is backgrounded or suspended, right? [15:02] dandrader: yep [15:02] dednick: hey. I like the new indicators look :) [15:02] dandrader: for now at least [15:02] mzanetti: yeah, it's quite nice what they've done. where did you see them? [15:03] dednick: on devel-proposed [15:03] dednick: I think there's more to come which I haven't seen yet. [15:03] ah. the menus. you havent seen panel yet ;) [15:03] dednick: but I'm referring to the things like the green highlight on active wifi [15:07] meh. my krillin wont let me ssh on :( [15:11] mzanetti, are you planning to take https://code.launchpad.net/~gerboland/qtmir/desktopFileReader/+merge/235223 in you next landing? [15:11] tsdgeos: phone-right-edge built [15:12] dandrader: hmm... its not in the current silo. will put it in the next one, ok? [15:12] mzanetti: url is ppa:waht [15:12] ? [15:12] mzanetti, ok [15:12] tsdgeos: ppa:unity-team/phone-right-edge [15:15] @unity does anyone have mako flashed with devel-proposed version 261? [15:15]