/srv/irclogs.ubuntu.com/2014/01/17/#ubuntu-unity.txt

RAOFkarni: That looks like library version skew.00:26
RAOFWhat's the demo-stuff PPA? It's probably the thing that's broken.00:27
karniRAOF: Yes, I'm sure that's the problem ;)00:28
karniThat's why I raised it, so maybe someone from Unity team will pick it up tomorrow/today morning.00:28
RAOFYeah, looks like something needs a rebuild.00:29
=== duflu_ is now known as duflu
Saviqmzanetti, cool, thanks06:05
Saviqkarni, she's off until Thursday I think06:05
Saviqkarni, so yeah, it's fine06:05
Saviqkarni, you can access old comments on gdrive - the comments button will take you to all comments06:06
Saviqmhall119, they were, but I had to redo a few things, so they're coming06:06
Saviqmhall119, I'm waiting for a new release of unity-scopes-api basically06:07
* mhall119 hopes Saviq is just starting work, not just ending it06:24
Saviqmhall119, I'm in Cape Town, so yeah, just got in06:25
mhall119ok, I'll look for the new stuff in ~8 hours then06:25
Saviqricmm, https://lists.launchpad.net/ubuntu-phone/msg04549.html06:44
didrockswaow, but people are staring on emails!06:44
didrocksah no, I was thinking it was mine ;)06:45
didrocks(see #ubuntu-eng-ci)06:45
didrocksSaviq: btw, should we increase the timeout?06:45
Saviqdidrocks, why?06:45
Saviq(you mean the kill timeout?)06:46
didrocksSaviq: as you are still seeing corrupted stacktrace06:46
didrocksyep06:46
Saviqdidrocks, it was only corrupted in the jobs before we installed u8-autopilot06:46
didrocksSaviq: are you sure? I remember we saw some after this package being installed06:46
Saviqdidrocks, which were not smashed, but just didn't retrace06:47
Saviqdidrocks, so what I'm actually thinking is moving it to unity8 itself06:47
didrocksSaviq: yeah, I think it's making sense06:47
Saviqdidrocks, as it only ever changes things when you go "initctl unity8 stop"06:47
didrocksI was about to suggest the same :)06:47
Saviqwhich normal users won't do, and if it crashes runtime, it will wait for apport to complete anyway06:48
didrocksexactly06:48
Saviqdidrocks, https://code.launchpad.net/~saviq/unity8/move-upstart-timeout/+merge/20204106:52
ricmmgreyback: do you know enugh about the OSK controller thing?06:55
ricmmneed to fix that thing where keyb is the wrong orientation for sidestage apps06:55
greybackricmm: I do. Point me to the bug please06:56
ricmmgreyback: https://bugs.launchpad.net/ubuntu-keyboard/+bug/125738306:57
ubot5Ubuntu bug 1257383 in ubuntu-keyboard "On tablet; keyboard sideways" [Undecided,New]06:57
ricmmold report, but its basically the bug06:58
ricmmin sidestage, app thinks its portrait, rotation is set to -9006:58
greybackricmm: yeah. It's actually ubuntu-keyboard issue, shell/unity-mir is not responsible for how OSK draws itself06:59
greybackricmm: so ubuntu-keyboard needs to be made aware of the stage the app is running on. Eeek07:00
ricmmgreyback: why? cant the keyboard just follow gravity07:01
greybackricmm: it could07:01
ricmmit already does for main stage I believe07:01
ricmmits just getting confused with what portrait means for sidestage07:02
greybackricmm: now i am too. Did we add a hack to rotate the orientation reported to sidestage apps by 90 degrees at one stage?07:08
* tsdgeos complains about getting up too early to get an earlier EOD07:09
ricmmgreyback: none that I remember, but I do know that rotation is reported through the OSK controller07:09
ricmmand no longer directly by maliit07:09
* tsdgeos starts to work07:09
tsdgeosSaviq: i did try to test it and did not really succeed but i'll give it another go07:10
Saviqtsdgeos, thanks07:10
tsdgeosshouldn't be *that* hard since karni was hitting it :D07:10
greybackricmm: it's not. OSKController just waits for ubuntu-keyboard to create a surface, it makes it visible and sets up an InputArea for it. That's all07:10
ricmmok then not osk controller07:11
ricmmbut that socket based comms that we have07:11
ricmmthat reports geom07:11
tsdgeoskarni: there?08:22
Saviqtsdgeos, https://code.launchpad.net/~aacid/unity8/bad_loop_dash_test/+merge/201802 works doesn't it?08:31
tsdgeosSaviq: sure08:36
tsdgeosSaviq: but we don't seem to be hitting the thing it fixes atm08:36
Saviqtsdgeos, ok, let's see08:37
tsdgeosdamn now i can't reproduce the crash in the journals in karni's branch08:39
tsdgeoshe seems to have changed something else that makes it not happen :/08:39
tsdgeosah good bzr08:40
tsdgeoslet me go back to the rev it fails08:40
mzanettiSaviq: hey, can you check this one? https://code.launchpad.net/~elopio/unity8/app_preview/+merge/20171809:50
mzanettiSaviq: it contains code from me and I guided elopio through this. So I'd prefer having another pair of eyes on it09:51
=== dandrader is now known as dandrader|bbl
mzanettianyone up for confirming this? https://bugs.launchpad.net/hud/+bug/126978810:09
ubot5Ubuntu bug 1269788 in Unity HUD "hud-service goes to 100% cpu if unity8 is not running" [Undecided,New]10:09
tsdgeosSaviq: added the tests, took me more than expected10:09
tsdgeosmzanetti: destkop or device?10:09
mzanettitsdgeos: device10:09
tsdgeosmzanetti: so you do /sbin/stop unity8 ?10:09
mzanettitsdgeos: yeah10:10
mzanettitsdgeos: well... I do ./run_on_device10:10
tsdgeos:D10:10
mzanettitsdgeos: which stops the service and runs another one. if that one crashes10:10
mzanettithen hud-service is stuck at 100%10:10
tsdgeos/sbin/stop unity810:10
tsdgeosworked here10:10
tsdgeoshud is fine10:10
tsdgeosand then starting a new one10:11
tsdgeoshud went to 100% briefly10:11
tsdgeosbut is back at doing nothing10:11
mzanettitsdgeos: hmm... my device is constantly burning here10:11
tsdgeostoday's image?10:12
tsdgeosi'm on yesterdays10:12
mzanettiit's happening since weeks already10:12
mzanettiat least since a week before xmas10:12
tsdgeos13110:12
tsdgeosoh10:12
mzanettitsdgeos: so to be more precise, what I do is this:10:12
mzanettiI install my patched libunity-mir10:12
mzanettithen stop the installed unity810:12
mzanettiand start my patched unity810:13
mzanettiwhen I stop my patched unity8, the installed one doesn't start any more because it's not compatible with the installed unity-mir10:13
tsdgeosmaybe unity-mir patch breaks hud?10:13
tsdgeosi mean10:13
tsdgeoshud wants some stuff exported by unity-mir10:13
tsdgeosand it may get unhappy if we're not providing that10:13
mzanettioh... interesting10:13
mzanettitsdgeos: well, I only added stuff10:13
tsdgeosthe dbus application stack model thing10:13
mzanettibut sure... might cause woes10:14
tsdgeosis consumed by hud afair10:14
mzanettimhm... interesting10:14
mzanettitsdgeos: but whenever I start my patched unity8, hud-service goes down to 010:14
mzanettiwhen I stop unity8 it freaks out again10:14
Saviqmzanetti, I probably won't be able to look at it before Monday in over a week10:36
mzanettiSaviq: ok, I'll ask someone else10:37
Saviqthostr_, mhr3 lunch10:45
tsdgeosCimi: https://code.launchpad.net/~cimi/unity8/fix-1214423/+merge/192868 ?11:03
tsdgeosCimi: dandrader|bbl: https://code.launchpad.net/~cimi/unity8/searchIndicator-swipe/+merge/198258 ?11:04
tsdgeosmzanetti: you doing https://code.launchpad.net/~elopio/unity8/app_preview/+merge/201718 right?11:05
mzanettitsdgeos: I'm trying to find someone else because it contains code from me and basically I told elopio to do this. So I'd like to have another pair of eyes on it11:05
mzanettitsdgeos: can you?11:06
tsdgeosmzanetti: sure, i'll add me to it11:06
mzanettithanks11:06
dednickmzanetti: want to see that solution i came up with for the MouseArea which doesn't consume events? :)11:14
* mzanetti senses a trap11:14
dednickmzanetti: hehe.... it's a bit "interesting"11:14
mzanettisure I want to know :)11:14
mzanettiI had this problem quite often already11:15
dednickI'll just pastebin the cpp. And it only works for pressed, released, clicked, dblclick at the moment.11:15
dednickhttp://pastebin.ubuntu.com/6767344/11:16
mzanettifor example it would allow me to get rid of a workaround in the right edge thing11:16
mzanettiah ok...11:16
mzanettiwell sure... if there is a proper way to do it, I guess this is it11:16
mzanettidednick: so this first evaluates the qml code inside the PassthroughMouseArea. if there is accepted == true it behaves like a regular mousearea, right?11:18
mzanettidednick: if accepted == false, it still keeps on getting the following events, but they are additionally re-injected into QCoreApplication11:18
mzanettiam I reading this correctly?11:18
dednickmzanetti: if accepted==true (ie the PassthoughtMouseArea was pressed), then it re-sends the mouseevent through core application11:19
dednickmzanetti: but it bypasses this mouseArea, because m_enabledEvents==false now.11:20
mzanettiah... 'cause if accepted == false, the MouseArea itself forwards it already... got it11:20
mzanettidednick: hmm... will this land in unity8 anytime soon?11:20
mzanettithat would really solve the one dirty workaround I have in the rightedge stuff11:21
dednickmzanetti: er. maybe. there's a few other branches it depends on.11:21
dednickmzanetti: and it's not actually in MP yet.11:21
mzanettiok. no problem. the rightedge stuff is still a bit away too. just let me know when I can start using this please11:22
dednickmzanetti: ok.11:22
=== _salem is now known as salem_
=== MacSlow is now known as MacSlow|lunch
dednickmzanetti: um, what's currently the best way to get a click event through qmltests?11:55
mzanettidednick: huh? mouseClick() ?11:56
dednickmzanetti: what about tap?11:56
mzanettidednick: I think everything is converted to a tap internally11:56
dednickdont really understand why we have different ones for touch...11:56
mzanettiso we shouldn't have different ones in unity11:56
mzanettidednick: dandrader knows all the details though11:57
mzanettidednick: but afaik, if there is a mouse click, the uqmlschene we use for testing should convert it to a tap11:57
dednicki c11:58
dednickdoesnt seem to work.11:59
dednickdandrader|bbl: ping when you're back11:59
mzanettidednick: pun intended? :P12:17
dednickmzanetti: :)12:18
dednickno!12:19
mzanetti:D12:19
karnitsdgeos: hi :)12:24
tsdgeoskarni: hi, was nothing, Saviq answered all your questions, right?12:24
* karni reads e-mail12:24
karnitsdgeos: If I understand that correctly, for testing purposes, line 292 [1] is fine, because we're not on the CardVerticalJournal.qml level yet? [1] https://code.launchpad.net/~unity-team/unity8/new-scopes-vj-integration/+merge/20193212:29
karnitsdgeos: It seemed strange to me I had to set ResponsiveVerticalJournal's columnWidth and delegate width separately12:30
karniIf it where in CardVerticalJournal.qml, setting CardVerticalJournal columnWidth would suffice.12:30
karniBy the way guys, I wrote this here last night, maybe someone can have a look at it:12:34
karni Unity8 won't start with packages installed from demo-stuff PPA: http://paste.ubuntu.com/6764850/12:34
karnitsdgeos / Saviq: Would you guys have a moment today to review the merge [1] above? :)12:37
=== alan_g is now known as alan_g|lunch
tsdgeoskarni: i can try having a look later12:38
karnitsdgeos: thank you12:38
karnitsdgeos: silly question - should this part of code ('category-layout': 'vertical-journal') already work in unity-scope-tool? It shows nothing when I switch to vertical-journal. If it should show stuff, I may have gotten something wrong.12:40
karnithe try and testResponsiveVerticalJournal work fine12:41
tsdgeoskarni: if you're not hooking category-layout vertical to the vJournal stuff12:42
tsdgeosi'd say no12:42
tsdgeosand i don't see you doing it, no?12:42
* mzanetti needs to leave for an hour and a half. bbl12:42
karnitsdgeos: line 160 in qml/Dash/GenericScopeView.qml: case "vertical-journal": return "CardVerticalJournal.qml";12:43
tsdgeosi hate it when ctrl+f fails on me :D12:43
* karni same here ^ ^12:43
tsdgeosthen i'd say it should work™12:44
tsdgeosbut i'm not very uptodate in the new scopes architecture12:45
karnitsdgeos: ok, I think that might be because CardVerticalJournal is missing: width: template['card-size'] (the GU value of it)12:45
karnitsdgeos: I guess not many of us are :D12:45
karnisuper fresh12:45
Saviqmzanetti, you're famous now ;)12:45
Saviqdidrocks, I'm off next week, mzanetti would you like to be our first trained landing engineer in the team?12:51
didrocksSaviq: ok, switching place sounds fine, mind forwarding that to mzanetti?12:51
didrocksSaviq: he will have to do the paperwork, you win! :)12:52
mzanettiSaviq: didrocks: ok...12:56
=== dandrader|bbl is now known as dandrader
mzanettiSaviq: yeah... didn't expect that much fame :D12:56
didrocksmzanetti: did you get the email or should I forward it to you? ;)12:57
dandraderdednick, ping12:57
mzanettididrocks: only got the invitation12:58
dednickdandrader: howdy. was wondering what the diff is between mouseClick and sending press/release events in qml tests. Seems that click doesnt work with draghandle hinting.13:02
=== dednick is now known as dednick|lunch
=== MacSlow|lunch is now known as MacSlow
dandraderdednick|lunch, dragHandle deals with touch events, not mouse events13:05
dandraderdednick|lunch, so you should send a touch tap instead of a mouse click13:06
=== dednick|lunch is now known as dednick
=== alan_g|lunch is now known as alan_g
Saviqkarni, you haven't assigned the model to ResponsiveVerticalJournal in CardVerticalJournal, or bound CVJ's height to RVJ13:47
* karni looks13:47
Saviqkarni, basically because you're not doing DashVerticalJournal, you need to bind much more than you have - see DashFilterGrid13:49
karniack13:49
Saviqkarni, also, template and components are undefined on ResponsiveVerticalJournal13:49
Saviqkarni, so            template: cardVerticalJournal.template13:50
Saviq            components: cardVerticalJournal.components13:50
Saviqis wrong13:50
cwayneheya guys, is there a timetable for scopes-as-click-packages13:50
Saviqmhr3, do you know anything ↑?13:54
mhr3thostr_, ^^13:55
karniSaviq: thostr_: I know it's quite late notice, but would you guys like to hangout later and update us on progress made during the sprint?13:56
thostr_cwayne: click packages are planned for 14.04 but not done by mid of Feb meaning won't be available for mwc13:56
thostr_cwayne: do you need click package by mid Feb?13:56
karniSaviq: I'm processing your comments13:56
thostr_karni: I need to leave in couple of minutes for the airport13:57
karnithostr_: maybe monday then?13:57
thostr_karni: yes, sure13:57
cwayneneed might be a strong word, but it would certainly simplify things quite a bit for us thostr_13:57
karnithostr_: anyway, we'll schedule the meeting13:57
karnithostr_: safe flight!13:57
thostr_karni: thanks13:57
Saviqdidrocks, greyback is on the spreadsheet for wave 1, too, and he's off next week as well13:58
Saviqdidrocks, I think it should be merged into team Unity TBH13:59
didrocksSaviq: yeah, I agree, kgunn: are you happy to merge that in the Unity team and only have one lander?13:59
Saviqdidrocks, kgunn, or we get tsdgeos to be the second one (still, for a single Team Unity)14:00
elopiotsdgeos: I updated my MP. Thanks for taking a look.14:01
tsdgeosno worries14:01
karniSaviq: re: no model in ResponsiveVJ - isn't it set by the user?14:02
Saviqkarni, the property is set on the top component - DashRenderer14:03
Saviqkarni, you need to bind it to RVJ14:04
karniack14:04
Saviqkarni, and probably a few others14:04
kgunndidrocks: i'm always happy if Saviq is happy14:07
didrocksok, modifying then, thanks guys14:07
karniSaviq: pushed bound properties of CardVerticalJournal top level component. Sadly, I don't know where the template and component values would come from.14:18
karniSaviq: If it seems I'm slowing you down, you should tell me :)14:18
Saviqkarni, they're there on DashRenderer14:22
karnioh. I should also create property aliases for 'em?14:22
karniSaviq: Conceptually, when someone instantiates CardVerticalJournal -- will they introduce their own delegate? if so, the current delegate in CardVerticalJournal makes little sense14:24
karniOr rather, that's just a good default value.14:25
Saviqkarni, that's the only value, yes14:25
Saviqkarni, so there should be no "delegate" property on CVJ14:26
karnioh14:27
* karni removed14:27
karniSaviq: Card (delegate) expects template and components on ResponsiveVerticalJournal (its parent), would this work then? http://paste.ubuntu.com/6768200/14:30
karniSaviq: I don't want to flood you with MP updates..14:31
Saviqkarni, why do you say it expects it on its parent?14:31
Saviqkarni, it has those as its own properties14:31
karniSaviq: you said it's undefined. should I point the card to the DashRenderer instead?14:31
karniSaviq: I'm not sure if I have enough brevity for those final touches :(14:32
karnibrevity - wrong word heh14:33
karnifluency14:33
Saviqkarni, that's where the values are14:33
Saviqkarni, so yes, you need: template: genericVerticalJournal.template instead14:33
Saviqkarni, cardVerticalJournal does not have a components or template properties14:33
tsdgeosdednick: standtup14:34
tsdgeosnic-doffay_: ↑14:34
nic-doffay_tsdgeos, on my way14:35
tsdgeosSaviq: wanna join just to tell us how you're on holiday next week?14:35
Saviqtsdgeos, can't14:35
tsdgeosSaviq: ok14:35
Saviqtsdgeos, not sure I understand your question though ;)14:36
karniSaviq: http://paste.ubuntu.com/6768235/ ? :)14:37
karniSaviq: seems that's still not sufficient to make unity-scope-tool accept the vertical-journal integration. :S14:40
Saviqkarni, yes, model and height were not propagated (model down, height up) between DashRenderer and RVj in CVJ14:44
Saviqkarni, although that wasn't enough either14:45
Saviqkarni, I can't get much further though14:45
karniSaviq: appreciated14:45
tsdgeosdandrader: ping14:53
dandradertsdgeos, pong14:53
tsdgeosdandrader: i had a look at the gallery14:53
tsdgeosand does the same thing i do14:53
tsdgeosi.e. the 4 element is put on the bottom row14:53
dandradertsdgeos, well I had a different impression14:53
tsdgeoslet me do a photo of the phone14:54
dandradertsdgeos, but we can merge it as it is14:54
dandradertsdgeos, and tweak this stuff later14:54
tsdgeosnow gallery doesn't want to start14:54
* tsdgeos reboots phone14:54
tsdgeosdandrader: got the email?15:00
dandradertsdgeos, yes15:01
dandradertsdgeos, have you tried the situation I explained in my e-mail15:03
dandradertsdgeos, a module with only 2 items15:03
dandradertsdgeos, (e.g. a fully filled module and on its left a module with only 2 items15:03
dandrader)15:03
tsdgeoson the right?15:03
dandraders/left/right15:05
tsdgeosdandrader: karni: can any of you top approve https://code.launchpad.net/~aacid/unity8/journal_misc_fixes ?15:10
tsdgeosdandrader: so delete 2 items from the photo i sent you?15:10
dandradertsdgeos, yes15:11
* karni looks15:11
tsdgeosdandrader: ahh15:11
tsdgeosright that is different15:11
dandradertsdgeos, aha!15:11
tsdgeosi thought you were always speaking of the 4 item configuration15:11
tsdgeosnot of the 2 item one15:11
karnitsdgeos: looks good to me, shall I top approve? In previous teams the propose'er would do that.15:15
tsdgeoskarni: please15:15
karnitsdgeos: done15:15
tsdgeoswe only top approve our own stuff if autolanding fails because one of the known "unstable" tests15:16
karniI see. So the reviewer would top-approve? Also, do you usually seek more than one review?15:16
tsdgeosdepends15:16
tsdgeosi'd say no15:17
karniok15:17
tsdgeosit's hard enough to get one to review your stuff :D15:17
karnihaha15:18
karniWhat's this aboug guys? Unity8 won't start with packages installed from demo-stuff PPA: http://paste.ubuntu.com/6764850/15:23
tsdgeosi'd say rebuild needed somewhere15:25
karniIs there anyone who could look into that? (/me doesn't know people on the team well, yet)15:26
tsdgeosmhr3 or Saviq are the ones that rebuild that ppa afaik15:26
tsdgeossaviq is out next week15:27
tsdgeoslet's hope mhr3 is at work and not touring south africa too15:27
karnihaha15:27
tsdgeospstolowski: do you know if mhr3 is back next week?15:27
pstolowskitsdgeos, he should be, i think so15:27
tsdgeosoki15:27
karniphew15:28
* tsdgeos enjoys his early day start with an early eow15:31
* tsdgeos waves15:31
tsdgeosnic-doffay_: see you man!15:31
karnio/15:31
=== dandrader is now known as dandrader|lunch
didrocksmzanetti: if we move the bootcamp to 16 UTC instead of 15 UTC, would that work for you?16:53
didrocks(to work a little bit better with the NZ guys)16:53
mzanettididrocks: ok, sure16:55
didrocksthanks!16:56
=== dandrader|lunch is now known as dandrader
=== alan_g is now known as alan_g|EOW
karnimzanetti: I learned from omgubuntu about the new app switcher. woot \o/ :) (though, I was bit surprized about the horizontal scroll control, seemed way to responsive/delicate)18:19
mzanettikarni: yeah, it's a bit too fast still18:20
mzanettikarni: need to tweak the values still18:20
mzanettikarni: so, now you know what I'm talking about for days in the standup already :D18:20
karnimzanetti: hahahah yeah, from omgubuntu ;D18:20
karnimzanetti: looks great18:21
mzanettithanks18:21
karnimzanetti: do you know which trusty-proposed revision can I flash to work with demo-stuff ppa? The most recent trusty-proposed just doesn't work with our new stuff.18:22
karnimzanetti: you using trusty-proposed, right?18:22
* karni tries revno 13218:26
fgintherSaviq, the unity8 otto jobs are failing again with what looks like the same signature, but I've been able to identify a changed dependency that may be the culprit, libunity-scopes018:56
karnifginther: Hey Francis. He may have EOD'ed (haven't seen a while in here, he finished sprint today), he's on holiday next week.19:04
fgintherkarni, thanks. I'll find someone else to ping Monday19:04
=== dandrader is now known as dandrader|afk
=== dandrader|afk is now known as dandrader
=== salem_ is now known as _salem
=== _salem is now known as salem_

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!