[07:06] morning [07:06] starting earlier today so I can watch italy later on :D [07:26] Saviq, I am doing tests [07:26] Saviq, found more bugs fixing them [07:26] (some conditions where the layout was breaking) [07:31] Saviq: taaaaaaaaags [07:32] hah [07:34] tsdgeos, damn :| [07:34] no idea how they got to my unity8 [07:35] tsdgeos, I do [07:35] seems they are not upstream [07:35] tsdgeos, they were, yesterday [07:35] tsdgeos, or well, a few days before [07:35] but somehow i ended up infecting swipe_down_close_no_remove_dash [07:35] tsdgeos, Kaleo's branch [07:35] cleaning up now [07:35] tsdgeos, I will do some checking again [07:36] before the next landing [07:36] oki [07:37] ooh 5.3 is a landeded [07:38] crap [07:54] facundobatista: hey, was on holiday yesterday [07:55] facundobatista: if you have an utopic machine now, you can branch lp:unity8 and run ./build.sh in there. Then you'll have a "cardtool" somewhere in the builddir [07:55] facundobatista: ping when you're around I can give you better instructions then [09:57] tsdgeos, hmm we seem to be missing some "don't animate height" on the VJ [09:58] tsdgeos, I did http://paste.ubuntu.com/7673731/ [09:58] AAARGh [10:00] Saviq: i think i missed half of what you said [10:00] i only got [10:00] tsdgeos, hmm we seem to be missing some "don't animate height" on the VJ [10:00] tsdgeos, https://code.launchpad.net/~aacid/unity8/dashviews_integration/+merge/222192/comments/537179 [10:00] tsdgeos, I did http://paste.ubuntu.com/7673731/ [10:01] tsdgeos, and there's a small inline comment on the doxy comment [10:01] ok [10:01] * tsdgeos reads [10:01] oh come on, got disconnected again :/ [10:01] tsdgeos, no, you're still here [10:01] wow [10:02] Saviq: known troubles with QtFeedback? [10:02] oh come on, got disconnected again :/ [10:02] tsdgeos, no, you're still here [10:02] <-- tsdgeos has quit (Read error: Connection reset by peer) [10:02] tsdgeos, you knew it before it happened ;) [10:02] mzanetti, context? [10:02] plugin cannot be loaded for module "QtFeedback": The plugin '/usr/lib/x86_64-linux-gnu/qt5/qml/QtFeedback/libdeclarative_feedback.so' uses incompatible Qt library. (5.3.0) [10:03] mzanetti, looks like you didn't upgrade completely [10:03] yeah... but apt doesn't agree on that [10:03] yeah it's kind of weird [10:03] i was disconnected when i wrote that [10:03] then reconnected [10:03] and client quite [10:03] no idea why [10:03] mzanetti, you using a mirror? [10:03] mzanetti, you probably upgraded in-between migrations from proposed [10:03] looks like... [10:03] qtsensors coming in noe [10:05] Saviq: 4-indented ? you mean with tabs? or 4 spaces? [10:05] tsdgeos, 4 space [10:05] tsdgeos, like my patch did ;) [10:05] tsdgeos, I'll prep one for the other places where that happens [10:06] but separate MP [10:08] tsdgeos, you know what I meant for resize-on-creation? [10:10] Saviq: hmmm, not really [10:10] can't see anything wrong after applying your patch [10:10] tsdgeos, scroll fast [10:10] tsdgeos, expand first category, scroll past it, scroll back up [10:11] tsdgeos, you should see it not being expanded (which it should) and animating height (which it shouldn't) [10:12] erhmmmm [10:12] something's very wrong [10:12] i'm just not getting the VJ :D [10:12] tsdgeos, did you build? ;) [10:12] i did [10:13] tsdgeos, did you build your VJ branch? [10:13] yeah [10:13] http://paste.ubuntu.com/7673799/ [10:14] tsdgeos, well, it might be grid in fact, it's not like it's easy to discern ;) [10:14] well i replaced the carousel too [10:14] and still get it :D [10:14] tsdgeos, so maybe my bug is in trunk ;) [10:14] it may [10:14] i can't repro though [10:15] tsdgeos, anyway there's a few template defs above as well [10:15] ah but that's for special [10:15] ah i changed the wrong ones [10:15] ... [10:16] silly me [10:17] yeah, why is it doing that? [10:17] * tsdgeos checks [10:32] Saviq: ok, should be better now, can you check? [10:37] tsdgeos, yeah, checking [10:39] tsdgeos, it needs 4 spaces *more* than normal text ;) [10:39] ¿?¿? [10:39] * tsdgeos is confused [10:39] tsdgeos, normal text is * foo [10:39] the ascii art [10:40] needs to be * foo [10:40] so i'm missing 1 space still? [10:40] tsdgeos, yes [10:40] tsdgeos, make -C builddir doc [10:40] and builddir/doc/unity8/class_responsive_vertical_journal.html [10:40] uh oh [10:41] do you guys get http://paste.ubuntu.com/7673886/ on make testShell? [10:41] tsdgeos, looks like it, yes :| [10:41] http://nooooooooooooooo.com/ [10:42] tsdgeos, yeah, same trace [10:44] this crash in valgrind seems veeeeeery similar to the other one we were having [10:44] * tsdgeos builds with debug again [10:45] tsdgeos, ACK on VJ [10:45] \o/ [10:49] MacSlow, there's still inline comments in https://code.launchpad.net/~macslow/unity8/combo-button-support/+merge/221499 that need action (reply at least) [10:49] Saviq, I'm on those... [10:49] MacSlow, ah ok, thought you were done with that one, too [10:49] Saviq, was surprised I didn't see them (all) before. [10:49] MacSlow, yeah, they're not easily visible indeed [10:50] Saviq, the other three branches seem fine [10:50] MacSlow, and then if you change, the diff changes and so you need to switch to the previous diff to be able to see them [10:50] Saviq, yeah... by now I'm aware of that new LP-diff-comment feature [10:51] Saviq, after all that new way is nicer for comments... especially for the commenter [10:51] MacSlow, indeed, well, as submitter you can reply to those, so that's good, too [10:52] there's stuff to improve for sure [10:52] but it's a start [10:53] Saviq: one for the lazyimage test https://code.launchpad.net/~aacid/unity8/fix_lazyimagetest_53/+merge/223896 [10:53] then there's the multigreeter one [10:53] in which html behaviour has changed [10:54] someone needs to check if we're fine with that change or not [10:54] mterry probably [10:54] and then the crash i'm investigating [10:54] i'll create a bug for mterry and the multigreeter test failures [10:54] tsdgeos, great, thanks [11:05] k folks, taking second half of my holiday o/ [11:08] Saviq, you off now? [11:18] Cimi: i'm pretty sure this was fixed in our latest release, can you double check it ? [11:18] https://bugs.launchpad.net/mir/+bug/1319907 [11:18] Ubuntu bug 1319907 in Mir "Last shell frame isn't sent to the display" [High,New] [11:19] kgunn, phone dead at the moment and without carousel [11:19] kgunn, let me ask mzanetti [11:19] mzanetti, do you remember the bug with the carousel? [11:19] Cimi: i can do it actually [11:19] ok [11:19] kgunn, thanks! [11:19] Cimi: hmm... which one? [11:19] ah... that one [11:19] haven't seen it any more lately [11:19] mzanetti, flick on carousel, no overlay [11:19] kgunn, ^ [11:19] thanks [11:20] I'll reopen it in case [11:20] Cimi: for future...how to reproduce ? add like 10 music files or something ? [11:20] kgunn, yes, just flick [11:20] kgunn: yep, just drop some music or videos on the phone [11:20] and carousels start appearing [11:20] kgunn, quite almost every time the overlay wasn't shown [11:20] no title/song [11:21] that said, I think I saw the one in the recent apps category again lately, where whe thought it would be related [11:21] but I didn't really investigate, so... take it with caution [11:22] can't reproduce it right now... [11:31] mzanetti: Cimi yeah...seems pretty solid, i can't repro [11:35] jhodapp, got a u8 crash with 5.3 coming from media-hub http://paste.ubuntu.com/7674074/ sorry for crap debug, don't have all the symbols, but fwiw it should be uploaded to daisy === MacSlow is now known as MacSlow|lunch [11:37] tsdgeos, I refactored my card emblem as saviq asked https://code.launchpad.net/~cimi/unity8/card_emblems_single_container/+merge/223899 [11:39] tsdgeos, I tested a bit, seems fine [11:39] tsdgeos, I am happy to get it merged and wait for bugreports if there are [11:39] there are so many nested cases that there is a chance something is broken [11:40] but I tested most [11:40] Cimi: i'll be away next week, and i'm having a look at a qt crash right now, maybe you should look for someone else to review [11:40] ok Saviq then [11:42] Cimi: did the new infographics already got merged? [11:42] tsdgeos, no [11:42] tsdgeos, unity side they work [11:43] tsdgeos, but camera app doesn't populate the service [11:43] tsdgeos, so we wait next week to talk with pete who is on holiday [11:44] tsdgeos, basically you can manually add infographics to the backend and you see them in unity... but there is something broken on the backend [11:45] oki === alan_g is now known as alan_g|lunch [12:04] mhr3: yeah, we weren't ready to switch to qt 5.3 === MacSlow|lunch is now known as MacSlow [12:41] mzanetti, saw your instructions from some hours ago; I'm currently installing stuff in this Utopic machine for build unity, thanks! === alan_g|lunch is now known as alan_g === _salem is now known as salem_ [13:40] dpm_: kgunn: hey, one of you should move the meeting to a different time slot. I don't mind which one :) [13:40] talking about reminders weekly and kevin/michael 1:1 [13:41] mzanetti, I can move it, sorry, didn't want to step on your 1:1 [13:41] dpm_: no... kevin stepped on the reminders meeting [13:41] aha! :) [13:41] dpm_: but I figured we're moving the reminders one often enough to probably move it for real [13:42] mzanetti, yeah, the last weeks have been quite hectic with other things and that's meant some movement there. Let me change the time for the one this week and then I'll find a better weekly time for the one next week [13:43] dpm_: appreciate it....i got a lot of guys...so shuffling is a challenge of all the 1on1's [13:44] kgunn, no worries [13:45] mzanetti, ok, moved this one to Monday, as Riccardo pinged me earlier he might not be able to make it today [13:54] dpm_: works for me [13:58] ok, thanks [14:25] greyback: i had a moment to tinker, export renderer_timings & when i tried to relaunch using [14:25] QT_QPA_PLATFORM=mirserver [14:25] it aborted [14:25] am i doing it wrong ? [14:25] kgunn: need to also set MIR_SOCKET=/run/mir_socket [14:26] kgunn: if you pulled today, the PPA is broken again, since Qt5.3 landed [14:26] working on fixing it now [14:27] greyback: ack... sorry man [14:27] kgunn: np [14:27] * kgunn wonders about qt5.3 so close to rtm [14:27] kgunn: I'll let you know when PPA back up and running [14:27] the gcc transition also didn't help [14:28] yeah :-/ [14:28] kgunn: i just discovered a few crashers too :/ [14:28] in qt5.3 [14:29] olli_: ^ to keep in mind for discussions about how upstream migrations occur [14:33] greyback: mirserver doesn't seem to exist... [14:33] Available platform plugins are: eglfs, kms, linuxfb, minimal, minimalegl, offscreen, ubuntu, ubuntumirclient, ubuntumirserver, xcb. [14:33] i tried the obvious ubuntumirserver [14:34] Failed to connect to "/run/user/32011/ubuntu-keyboard-info" after 10 failed attempts [14:35] kgunn, if you have time today, you could also be my welcome-wizard guinea pig. Not urgent, still waiting on some test script changes to be reviewed. But might be a nice Friday activity [14:35] mterry: is there something available now ? [14:36] i'm taking off for the day at lunch [14:36] kgunn, just install ubuntu-system-settings-wizard and run 'adb shell rm /home/phablet/.config/ubuntu-system-settings/wizard-has-run' to reset state [14:36] kgunn: you'd better just hold on until I have the PPA working again [14:37] kgunn, again, not urgent though [14:38] greyback: my apologies...i'm a distraction, just ignore me...note: i'm not on the ppa [14:39] kgunn: I'm confused, what are you testing? Trunk? [14:39] greyback: yep...just straight devel-proposed image [14:40] again..just ignore me [14:40] kgunn: ok, sorry, "mirserver" threw me off - that's PPA only. [14:40] kgunn: QT_QPA_PLATFORM=ubuntumirserver and MIR_SOCKET=/run/mir_socket should be enough [14:41] greyback: yeah that worked...but failed on keyboard [14:41] kgunn: those "Failed to connect" messages are normal, as maliit-server hasn't been started (it's usually started by unity8's upstart job) [14:42] maybe i didn't wait long enough [14:43] Cimi: hi. Can you help me review this branch? https://code.launchpad.net/~paulliu/unity8/lp1330957/+merge/223779 [14:43] paulliu, sure [14:43] Cimi: not a large one but fixes some FAILED in testCard [14:43] Saviq: any chance of the vertical journal fixes landing today? [14:44] cwayne: i think Saviq is off... tsdgeos might know [14:44] paulliu, is this a typo? headerLeftAnchorHasMagin [14:44] cwayne: it's approved [14:44] cwayne: https://code.launchpad.net/~aacid/unity8/dashviews_integration/+merge/222192 [14:44] tsdgeos: do you just need a silo to land it ? [14:44] cwayne: but it needs a lander [14:44] not sure if Saviq has something prepared or not [14:44] kgunn: but you're not starting the unity8 job, you're starting unity8 manually. So no dependent processes (like maliit, indicators) won't autostart. [14:45] tsdgeos: i'll check and get a silo if not... [14:45] oki [14:45] Cimi: No. The original code are using that variable. Or I should change all of them? [14:46] paulliu, you can skip the fix [14:46] paulliu, I will add it in my emblems branch [14:46] kgunn: you could run "initctl set-env QSG_RENDER_TIMING=1" and then "start unity8" as usual, and the ~.cache/upstart/unity8.log file will have the timings printed there [14:46] paulliu, the entire fix I mean [14:46] Cimi: ok. because I'll rebase my another branch about the attribute UI on your emblems branch. [14:46] paulliu, or you can do that [14:46] but that means *every* Qt process will do the same, unless you unset that right after [14:46] kgunn: ^ [14:46] paulliu, but I think is easier for you if you leave it to me [14:47] kgunn, no, no silo - we could land stuff that's there in u8 queue, unity-notifications could use a landing, too [14:47] paulliu, that code is quite awful to see (that js) [14:47] Cimi: OK, maybe I should do that. Because this bug fixing branch is smaller. [14:47] Cimi: yours are large. [14:47] paulliu, I mean I can add the fix to my branch [14:48] Cimi: ah. got it. Please do. [14:48] paulliu, but you can keep fixes for the card tests [14:49] Cimi: actually I need to fix the margin. [14:49] paulliu, ok fix it then [14:49] Cimi: ok. [14:49] paulliu, but please rename the variable [14:50] Cimi: ok. [14:52] Saviq: you know time off means off :)...but since you're on, do you mean you want me to create a silo and land everything that's approved+ready for unity8+unity-notifications ? [14:52] (sorry parse error on your last post) [14:52] kgunn, yeah [14:53] ack [15:06] cwayne: just an fyi...might not land until monday...looks like ci train is unmanned for the moment [15:06] so suspect i won't get a silo until late afternoon [15:07] kgunn: ack, thanks for the update, monday should be fine === dandrader is now known as dandrader|afk [15:18] mhall119, mind running your nice script to update the scopes docs? [15:18] mhall119, or is it running as a cron job somewhere? [15:18] mhr3: no cron yet, the changes are in trunk? [15:18] mhall119, in U even, yes [15:19] I'm still on trusty, so doing it from the branch is easier than from the package [15:27] paulliu, hey, charles fixed the failure in indicator-session, but unity8 still didn't log out for him [15:28] paulliu, I kicked a rebuild of everything in silo 008 [15:28] saviq, I sent paulliu a mail about it on (wednesday?) [15:29] charles, yes, seb128 fwd'ed it back to me [15:29] charles, so I don't have a reply, if there was any :) [15:29] mhr3: please check http://91.189.92.89/api/scopes/sdk-14.10/ [15:29] paulliu, are you able to trigger a logout in unity8 via dbus-send? [15:29] I'd like to decouple the testing of the indicator-session parts and the unity parts [15:30] paulliu, you could test on the phone btw, if your desktop doesn't run unity8 [15:31] paulliu, you just need to run the dbus command, or you could even export the UNITY_INDICATORS_PROFILE="desktop" into upstart to get the session indicator on the phone [15:33] mhall119, +1, we moved some stuff to a new unity::scopes::experimental namespace, but it's actually good that it's not in the docs, discouraging use of classes from there was why we did it [15:33] Saviq, there was no reply [15:33] mhr3: do any old classes need to be removed? remember that the import script never deletes things, only adds and updates [15:34] mhall119, hm, even if they were you wouldn't be linked from anywhere, no? [15:34] they wouldn't be* [15:35] mhr3: if you moved things from a supported namespace into experimental, they would still be in the old location [15:35] mhall119, right, but there wouldn't be any pages linking to that now [15:36] mhall119, or wait the index page is hardcoded, right? [15:36] no [15:37] http://91.189.92.89/api/scopes/sdk-14.10/ is generated from the API website's data [15:37] where does it get the categorization from? [15:37] mhr3: for scopes, I hand-crafted them [15:37] because the source docs don't have that data [15:38] so it is kindof hardcoded?! [15:38] for QML docs, I can get that from the sources [15:38] mhr3: only the categories [15:38] and those are hardcoded in the importer, not in the api-website [15:38] mhall119, ok, nonetheless, it looks good to me [15:39] ok, I'll push to production === dandrader|afk is now known as dandrader [15:41] Saviq: yes, I can do the logout by dbus-send [15:41] charles, ↑ [15:41] mhr3: please verify http://developer.ubuntu.com/api/scopes/sdk-14.10/ [15:41] charles: Can't you logout with dbus-send? [15:42] paulliu, can you paste to me the recipe you're using for that? [15:42] charles: I think tsdgeos also tested the dbus-send already. [15:42] charles: https://bugs.launchpad.net/unity8/+bug/1302213 [15:42] Ubuntu bug 1302213 in Unity 8 "API to bring down the session" [Critical,In progress] [15:43] paulliu, no it didn't work for me when I tried dbus-send from the command line. [15:43] charles: the unity8 didn't kill itself? [15:44] paulliu, I'm happy to re-test, could you paste the invocation you used, so I can repeat it hee? [15:45] mhall119, oh wait one leftover - "Deprecated List" [15:45] mhall119, should be gone now [15:46] mhr3: the full page? [15:47] mhall119, yep [15:47] ok, gone [15:48] mhr3: removed from staging as well, so it won't cause confusion on a future update [15:48] mhall119, awesome, thx [15:53] charles: ok. wait. [15:56] charles: first, bzr branch lp:~paulliu/unity8/logout and then ./run_on_device.sh (you might want to adb shell install some build-deps first). [15:56] paulliu, just upgrade from silo 008 [15:56] https://launchpad.net/~ci-train-ppa-service/+archive/landing-008 [15:56] paulliu, thanks. I'll try that here [15:57] still building latest unity8, but shouldn't be a problem (and should finish soon) [15:57] charles: And source this shell script. http://paste.ubuntu.com/7675317 [15:57] paulliu, if you ssh onto the device, that should be there in your env already [15:57] charles: And use qdbus com.canonical.Unity /com/canonical/Unity/Session com.canonical.Unity.Session.RequestLogout [15:57] Saviq: ah. ok [15:57] or use sudo -u phablet -i [16:00] well, I was testing on desktop since indicator-session doesn't have a profile on the phone [16:00] unless I'm misunderstanding what I'm needing to test here... [16:01] charles: Another way to test on desktop is use KDE, and then run ./run.sh, qdbus will also trigger a kill of unity8. [16:10] charles: I'm thinking if it is because I didn't modify debian/*.install. The plugin might not be installed. [16:10] charles: I'll check the silos [16:11] charles: hmm, no, there is the plugin. [16:14] charles: I'm testing silos right now. But I'll need some time for that. The network to ports and ppa isn't stable. [16:15] charles: Some Taiwan sites are under attack now. === salem_ is now known as _salem [16:22] paulliu: okay, on my laptop I'm logged into a unity8 session [16:23] paulliu: I ssh in from my desktop and run that script to get DBUS_SESSION_BUS_ADDRESS set [16:23] (nice script, by the way. I'd been sourcing the upstart file, but this is more convenient ;) [16:24] charles: try "qdbus com.canonical.Unity". It should prints all the objects [16:25] paulliu: doesn't exist. I'm wondering if an update I've run since Wednesday overwrote the silo [16:25] paulliu: 5 minutes :) [16:26] charles: what? Session might not be there, but there should be some other objects there already. [16:27] charles: I now only have 5000 B/s to ports and ppa. [16:27] charles: Cannot get anything. [16:28] charles, paulliu: I'm installing that ppa to try here [16:29] paulliu: to answer your question, this is what I see: [16:29] $ qdbus | grep Unity [16:29] com.canonical.Unity.WindowStack [16:29] com.canonical.Unity.Launcher [16:29] com.canonical.UnityGreeter [16:29] charles: hmm. So Session isn't there. I'll check. [16:30] paulliu: but I suspect I'm not installed right from silo 8, so as I said, give me a minut :) [16:30] charles: ok [16:30] seb128, thanks :) [16:31] charles, paulliu: the ppa doesn't work on utopic unity8-mir-desktop session :/ [16:32] seb128, do you mean when trying to log out through indicator-session, or when manually sending the RequestLogout call over the bus via dbus-send / qdbus? [16:34] the indicator [16:34] tryin qtdbus now [16:34] seb128, thanks [16:35] if that works, maybe there's still a different bug remaining in indicator-session [16:35] mterry: only thing i noticed is that the delay between the wizard and the spinner is kinda big... [16:36] long enough that as a user i wanted to start punching buttons [16:37] charles, trying to get qdbus to work first... [16:37] seb128, you have qdbus5 installed? [16:38] * charles is waiting on apt-get upgrade :) [16:38] charles, installing it [16:38] thanks [16:40] "qdbus com.canonical.Unity /com/canonical/Unity/Session com.canonical.Unity.Session.RequestLogout" doesn't do anything to unity8 [16:40] charles, paulliu: ^ [16:41] seb128: ok. [16:41] seb128: I'll check. I'm still on apt-get update. [16:43] seb128: what does "qdbus com.canonical.Unity" say? [16:44] charles, paulliu: unity8-mir.log has "Signal QQmlEngine::quit() emitted, but no receivers connected to handle it." [16:44] displayed every time the dbus request is emitted [16:45] paulliu, http://paste.ubuntu.com/7675560/ [16:51] paulliu, is the "no receivers connected" error helping? what code is supposed to be the receiver? [16:56] seb128: let me check. [16:56] that usually means in the main.cpp where the QQmlEngine is created, that no slot was connected to the QQmlEngine::quit signal to actually call QGuiApplication::quit() [16:57] greyback: yes. I think I did that already when you told me. [16:58] greyback: But might be cleaned after some trunk merge. I'll fix it. [16:58] paulliu: you've a better memory than I do. I don't remember saying that :) [16:58] :) [16:58] paulliu, https://launchpadlibrarian.net/178057888/unity8_7.89%2B14.10.20140618-0ubuntu1_7.89%2B14.10.20140620-0ubuntu1.diff.gz is what we got in the ppa upload [16:59] ups, sorry [16:59] launchpad seems confused === alan_g is now known as alan_g|EOW [17:04] Saviq: I push a new commit to logout branch based on seb128 and greyback's comments. [17:05] Saviq: add QObject::connect(view->engine(), SIGNAL(quit()), application, SLOT(quit())); in main.cpp [17:07] Saviq, charles, paulliu: I'm kicking a rebuild for unity8 so it picks up that change [17:08] paulliu, do you know how that code got dropped/why you had it in your local version and no in the branch? [17:09] seb128: My local branch also dropped sometime after merge trunk. But I don't know why it still works on all my devices. Even on phone. [17:09] ok, let's see if that rebuild makes it work for us === _salem is now known as salem_ [17:39] Saviq, charles, paulliu: ok, works now with the unity8 signal connect change [17:40] seb128: \o/ [17:40] well "works", it logout without asking for confirmation [17:40] finally [17:40] seb128, yeah, that comes later [17:40] but I guess that's what that change is supposed to do? [17:40] k [17:40] Saviq, so feel free to land the silo [17:40] seb128: right, i-session tries to pull up the dialog, but when it gets a failure response, that's treated as confirmation now... [17:40] seb128, Monday ;) [17:40] :) [17:41]