[07:21] <tsdgeos> Mirv: 5.3 is out, can you update the ppa with it?
[07:23] <Mirv> tsdgeos: yeah, just started working on it. I try not to remove symbols this time so it'll take some time, but I guess at least I can at least update the components as soon as they're ready (no such thing as component mismatches)
[07:24] <Mirv> I mean version mismatches. so once qtbase is done, I can copy it to the usual PPA:s and s on.
[07:27] <tsdgeos> oki
[07:41] <Saviq> mzanetti, FYI, the nonmirplugins dir is prepended to the import path when not running on Mir (so under X11 for example)
[07:41] <Saviq> mzanetti, we needed it to be on some other path than the usual mocks, it's not a fortunate name or solution, I agree, we might want to rethink this setup
[07:51] <Mirv> tsdgeos: awesome! qt5-daily is just now uptodate again with qtdecl + unity8 and I have scopes
[07:51] <Mirv> tsdgeos: launcher is broken though among else
[07:53] <tsdgeos> Mirv: but doesn't crash anymore on start?
[07:58] <tsdgeos> yes it does
[07:58] <tsdgeos> good stuff
[07:58] <tsdgeos> and yeah launcher is all black :/
[07:58] <tsdgeos> and carousel stuff is borked too
[07:58]  * tsdgeos shakes fist
[08:01] <Mirv> tsdgeos: doesn't crash. carousel seems ok though or am I missing something? (is carousel the music scope thing?:)
[08:02] <tsdgeos> Mirv: the carousel of music
[08:02] <tsdgeos> he bottom corners
[08:02] <tsdgeos> of the "grayish" square
[08:02] <tsdgeos> should be cut
[08:02] <tsdgeos> by the shape
[08:02] <tsdgeos> they are not
[08:02] <Mirv> tsdgeos: so I'll now rebuild some more stuff, qtubuntu-* and qtvideonode
[08:02] <Mirv> tsdgeos: ok I don't immediately see what's wrong but anyhow
[08:03] <tsdgeos> interestingly the launcher works on the desktop :/
[08:03] <Mirv> tsdgeos: did you have any features you know are in 5.3 that we want? to add to the "why we want 5.3" list of things, in addition to newer/better/morestable?
[08:05] <tsdgeos> Mirv: nothing obvious comes to mind tbh
[08:05] <Mirv> tsdgeos: ok. it was asked about.
[08:05] <Mirv> tsdgeos: so, hmm, we should probably go ahead with the landing-003 to get the 5.2 backport in?
[08:05] <tsdgeos> Mirv: +1
[08:06] <Mirv> I rebuilt the unity8 there yesterday after another unity8 had landed on Friday
[08:06] <Mirv> ok, I'll go back to 5.2 later today and look at it again
[08:06] <mzanetti> Saviq: yeah... just like I was expecting
[08:14] <Cimi> tsdgeos, is this C++ in fake_categories correct? http://paste.ubuntu.com/7489256/
[08:15] <tsdgeos> Cimi: what's wrong with it?
[08:16] <Cimi> tsdgeos, I don't know
[08:16] <Cimi> tsdgeos, I wrote it but my c++ knowledge is null
[08:16] <Cimi> tsdgeos, I was wondering if there is a better way to write 1.1, 1.3...
[08:17] <tsdgeos> Cimi: any reason you don't store the string but do all that weird stuff of turning it into a bytearray and then a char* ?
[08:23] <Cimi> tsdgeos, I think artMap["aspect-ratio"] is a char
[08:23] <Cimi> expects a char
[08:24] <tsdgeos> ?
[08:24] <tsdgeos> how would it expect a char? are we from the past?
[08:53] <greyback> Saviq: you knew of work on improving the orientation sensor output, any idea who was doing it?
[08:54] <Saviq> greyback, check with tvoss/ricmm
[08:54] <greyback> ok
[08:54] <Saviq> greyback, I don't think there's actual work happening
[08:54] <Saviq> Cimi, it's a QVariantMap, it can virtually take anything
[08:54] <Cimi> Saviq, indeed works with QString
[08:55] <Cimi> Saviq, but yesterday compiler said it was expecting char*
[08:55] <Cimi> Saviq, probably as a result of my bad way of concatenating variables
[08:55] <Cimi> I just used QString and works now
[08:57] <Cimi> mzanetti, init is called before every qmltest, right?
[08:57] <Saviq> Cimi, it probably just said it couldn't deal with what you gave it, and that char* was one of the supported types (probably just the first one)
[08:57] <Cimi> Saviq, yeah
[08:58] <mzanetti> Cimi: yes, I think
[08:58] <Cimi> mzanetti, ok so we have a lot of useless code in test DashContent
[08:58] <mzanetti> Cimi: there's 2, init() and initTestCase(). one is called before every test, the other once before all the tests
[08:59] <Cimi> mzanetti, init is before all?
[08:59] <Cimi> me adds a console.log
[08:59] <mzanetti> I think the other way round. but Imo they are named wrong
[08:59] <Cimi> init is before every test
[09:01] <Saviq> greyback, hey, I'm doing a proper review of unity-mir now - first instalment is over here https://code.launchpad.net/~gerboland/unity-mir/fix-upstart-closed-apps2/+merge/218721/comments/526045
[09:05] <Saviq> greyback, nothing major, mostly nitpicks and "your call"s
[09:06] <greyback> Saviq: as long as you found nothing broken functionally, I'm over the moon :)
[09:06] <Saviq> greyback, I didn't indeed
[09:06]  * Saviq is PITA
[09:06] <Saviq> but that's my job ;)
[09:14] <Cimi> Saviq, https://code.launchpad.net/~cimi/unity8/carousel-shadow-fixes/+merge/220213
[09:18] <Saviq> Cimi, k
[09:18] <Cimi> I noticed there's a console.log in this test
[09:18] <Cimi> any reason for that?
[09:19] <Cimi>             categoryListView.contentY = units.gu(11);
[09:19] <Cimi>             console.log("contentY", categoryListView.contentY);
[09:24] <Cimi> tsdgeos, https://code.launchpad.net/~cimi/unity8/carousel-shadow-fixes/+merge/220213
[09:25] <apw> larsu, i see that PS jenkins Bot had a hissy fit about a lack of changelog on that append bubble merge
[09:28] <tsdgeos> Cimi: why the qml/Dash/CardCarousel.qml change?
[09:29] <Cimi> tsdgeos, that was the issue causing the wrong shadow in video carousel
[09:29] <Cimi> tsdgeos, it was a typo of someone
[09:30] <Cimi> cardCarousel.components doesn't exist
[09:30] <Cimi> it's inside cardTool
[09:30] <larsu> apw: I'm on holiday
[09:30] <larsu> seb128: can you help apw out on that notify-osd thing please?
[09:30] <tsdgeos> Cimi: oh right
[09:31] <apw> larsu, heh no worries.  i shall just call you sad for even noticing while on holiday :)
[09:31] <seb128> larsu, apw: I put a commit message on the mp and I pinged MacSlow for review
[09:31] <apw> seb128, ahh that explains why the MP says commit message needed, and there seemed to be one ... strange system
[09:31] <larsu> seb128: thanks!
[09:32] <seb128> yw!
[09:36] <tsdgeos> Cimi: clean your branch
[09:36] <tsdgeos> Cimi: tags
[09:37] <Cimi> which tags to clean?
[09:37] <tsdgeos> 0.1.16               ?
[09:37] <Cimi> I don't have this locally
[09:37] <Cimi> http://paste.ubuntu.com/7492304/
[09:38] <tsdgeos> hmmmm
[09:38] <tsdgeos> weird
[09:38] <tsdgeos> something infected by "clean" unity8 branch
[09:38] <tsdgeos> sorry bout that
[09:44] <Saviq> greyback, ok, two more small comments, otherwise good!
[09:44] <greyback> Saviq: thanks, am addressing right now
[09:44]  * Saviq tries to revive unity8 session
[09:45] <Saviq> biab
[09:53] <Cimi> I have to go for a blood test, catch you later!
[09:55] <Saviq> Mirv, hey, https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2 says to install silo 006 on desktop, seems wrong?
[09:56] <Mirv> Saviq: oops, copy-paste
[09:56] <Mirv> Saviq: fixed
[09:56] <Saviq> Mirv, the "ppa-purge" part also mentions landing 006?
[09:57] <Mirv> Saviq: tsdgeos: btw qt5-beta2 now has a) qtdeclarative with delegate ranged patch + unity8 branch, and b) rebuilds of more of qtubuntu-* so it might be also camera, media playback etc works but I'm currently not able to test since I'm testing the 5.2 delegate range patch
[09:57] <Mirv> Saviq: oops2
[09:58] <Saviq> Mirv, k, I'm checking whether the black launcher also happens on desktop (so all-mir), or only on phone
[09:58] <Mirv> desktop not tested btw. the only problem there'd be is that not all modules are recompiled so you might be missing something. but at least Qt Creator doesn't anymore require the Qt Quick 1 so maybe it's all ok
[09:58] <Saviq> (on desktop+Mir that is)
[09:58] <Mirv> Saviq: ok, great!
[10:03] <tsdgeos> Saviq: http://paste.ubuntu.com/7492381/
[10:04] <tsdgeos> if you use that as Shell.qml
[10:04] <tsdgeos> you'll have a nice black recangle in the middle of the scren
[10:04] <tsdgeos> for no particular reason
[10:05] <tsdgeos> Saviq: maybe http://paste.ubuntu.com/7492386/ is less "artificial"
[10:05] <tsdgeos> so you get a black vs red rectangle
[10:05] <Saviq> tsdgeos, yeah, easier to explain, thanks
[10:16] <tsdgeos> Saviq: ok, i'll leave that in your hands for now and see if we can add some Mir people or something
[10:16] <Saviq> tsdgeos, Mirv, everything seems fine on Mir+desktop, installing on flo and manta now
[10:16] <tsdgeos> oh damn, that was fast :D
[10:17] <Saviq> tsdgeos, but yeah, I'll take it from here
[10:17] <tsdgeos> Saviq: oki
[10:20] <Wellark> mzanetti: did you notice my comments in the pinlockscreen MR?
[10:20] <mzanetti> Wellark: yes. already replied to them
[10:20] <Wellark> oh, ok.
[10:20] <Wellark> ah, I see
[10:20] <Wellark> sorry for the interrupt
[10:23] <mzanetti> no problem :)
[10:24] <tsdgeos> mhr3: Saviq: so what we do with the unity-api changes? land those first and then scopes-shell and unity8 in one go? or land the three in one go?
[10:24] <Saviq> tsdgeos, let's just land them all together, no need to split
[10:26] <tsdgeos> Saviq: well, splitting makes reviewing of the other stuff a bit easier maybe
[10:26] <tsdgeos> Saviq: so can we get a silo with the 3 things?
[10:27] <mhr3> tsdgeos, let's land it standalone, as you pointed out, it's not necessary to sync it, and the less sync required the better
[10:27] <tsdgeos> Saviq: ↑↑↑
[10:27] <tsdgeos> mhr3: will you review https://code.launchpad.net/~aacid/unity-scopes-shell/use-unity-api/+merge/219221 ?
[10:27] <Saviq> tsdgeos, mhr3, ok, I'll put in a silo asap
[10:28] <mhr3> tsdgeos, sure
[10:28] <tsdgeos> mhr3: cool tx
[10:31] <tsdgeos> mzanetti: and will you do https://code.launchpad.net/~aacid/unity8/use-unity-api/+merge/219222 ?
[10:32] <tsdgeos> Cimi: can you do https://code.launchpad.net/~aacid/unity8/noiconsfakeunity/+merge/219478 ?
[10:32] <tsdgeos> Cimi: and https://code.launchpad.net/~aacid/unity8/noiconpreviewheader/+merge/219502 that is already approved by mhr3 but he wants a second view
[10:37] <mzanetti> tsdgeos: yes, can do
[10:37] <tsdgeos> mzanetti: g8, tx
[10:56] <Saviq> tsdgeos, Mirv, confirmed on both flo and manta, so it seems to be android-specific
[10:56] <tsdgeos> :/
[10:58] <mhr3> Saviq, do we have the sync meeting today with design?
[10:58] <Saviq> mhr3, good question, I doubt it
[10:59] <Saviq> mhr3, although Mike is still in London
[10:59] <Saviq> mhr3, mikenagle, are we doing the sync meeting today?
[10:59] <mhr3> ah, is he?
[11:00] <mhr3> interesting answer :)
[11:00] <mikenagle> saviq mhr3 - we were just wondering the same thing :)
[11:01] <Saviq> ;)
[11:01] <mikenagle> is thostr_ in Malta?
[11:01] <mhr3> not yet
[11:01] <mikenagle> right - decision time - should we do the meeting?
[11:02] <mhr3> let's do it, and make it quick if there aren't things to discuss
[11:02] <mikenagle> k
[11:24] <Mirv> Saviq: ok
[11:27] <Mirv> tsdgeos: I've finished running all image AP:s + unity8 manual test plan, so I'd publish the unity8 (+ qtdeclarative) now
[11:29] <tsdgeos> Mirv: sounds good :) Saviq ↑↑↑
[11:29] <Saviq> Mirv, hit it!
[11:29]  * Mirv hits it
[11:30] <Mirv> now I need to practice that -gles package update thingy
[11:30] <Mirv> but I can also upgrade back to 5.3 and see if camera + videos work now with the rebuilds of qtubuntu-*  qtvideo-node in qt5-beta2
[11:34] <tsdgeos> Saviq: why isn't a unity-ci job running for https://code.launchpad.net/~cimi/unity8/carousel-shadow-fixes/+merge/220213 ?
[11:35] <Saviq> tsdgeos, because it failed?
[11:35] <tsdgeos> Saviq: what do you mean "failed"?
[11:35] <Saviq> tsdgeos, https://code.launchpad.net/~cimi/unity8/carousel-shadow-fixes/+merge/220213/comments/526139
[11:35] <tsdgeos> man
[11:36] <tsdgeos> that's sad timing
[11:36] <tsdgeos> wasn't there 2 mins ago
[11:36] <tsdgeos> :D
[11:36] <Saviq> ;)
[11:36] <Cimi> fixed
[11:38] <Saviq> AAARAAAAAGGEGRDFGH
[11:38] <tsdgeos> what?
[11:39] <Saviq> nothing
[11:40] <Saviq> couldn't move the launcher in a multimonitor setup
[11:40] <Saviq> well, it did move, but moved right back
[11:40] <Saviq> coupled with the fact that the barrier got b0rked and barely would let me through, I got a bit frustrated
[11:40] <tsdgeos> :D
[11:42] <Saviq> I'll be back, letting some steam off
[12:03] <Cimi> tsdgeos, I read about this qt quick compiler of qt 5.3
[12:03] <Cimi> how is it?
[12:15] <mzanetti> Wellark: ok. fixed most of the things.
[12:21] <Wellark> mzanetti: thanks!
[12:21] <Wellark> mzanetti: ready for round 2? ;)
[12:26] <mzanetti> sure
[12:27] <mzanetti> Wellark: I even got jouni to have a look at it and agree :)
[12:28] <mzanetti> Wellark: but I promised him that he can come up with changes when he has more time to look at it
[12:30] <Wellark> well, as long as we can get the current one in while waiting :)
[12:31] <Saviq> greyback, so, are we landing this thing or what? (I got a silo I could use)
[12:32] <greyback> Saviq: ok let's do it
[12:33] <Wellark> mzanetti: we now have two magical values for the pin lenghts
[12:33] <Wellark> namely 0 and -1
[12:33] <Wellark> should we just use 0
[12:35] <Wellark> mzanetti: where is this autoconfirm behavior coming?
[12:36] <mzanetti> Wellark: min == max && min > 0
[12:36] <greyback> Saviq: https://code.launchpad.net/~gerboland/unity-mir/land-devel-222/+merge/220245
[12:38] <Wellark> mzanetti: I mean where is the requirement for it coming from?
[12:38] <mzanetti> Wellark: design
[12:38] <Wellark> it is so rare to have exact lengths on codes
[12:38] <mzanetti> Wellark: I had tons of discussion about it already (I don't like it either)
[12:38] <mzanetti> Wellark: design wants to restrict the phone lock code to 4 digits
[12:39] <mzanetti> Wellark: I protested, I lost, I'm not going to discuss it again
[12:39] <Saviq> greyback, hmm I wonder if it wouldn't be better to just MP devel into trunk? that's what others seem to be doing?
[12:39] <Wellark> mzanetti: has this been raised with the security team?
[12:39] <greyback> Saviq: is it? Huh, lots of criss-cross merging there then.
[12:39] <Wellark> I'm sure they will protest
[12:40] <greyback> Saviq: ok let me re-do
[12:40] <Saviq> greyback, yeah, it's not great either, but it's clear at least
[12:41] <mzanetti> Wellark: they even wanted to restrict the SIM one to 4 digits. I managed to convince them for that at least
[12:41] <cwayne1> hm, unity-scope-tool failing for me: http://paste.ubuntu.com/7492832/ any ideas?
[12:42] <Wellark> mzanetti: well, I don't think we would have passed certification if we would have done that
[12:42] <mzanetti> Wellark: to calm down the security team, you still can have longer passphrases for the phone lock code, but you have to use the alphanumeric one (which ignores the length settings)
[12:42] <mzanetti> Wellark: yeah, that was my argument that convinced them
[12:42] <Wellark> as the GSM spec clearly says that SIM PIN is between 4 and 8
[12:42] <Wellark> so what is this 4 digit code then?
[12:42] <Wellark> I'm confused
[12:42] <mzanetti> Wellark: anyways... if you want a phhone lock code with only numbers, you're restricted to 4 digits.
[12:43] <mzanetti> Wellark: if you want higher security, use an alphanumeric passphrase
[12:43] <greyback> Saviq: try that: https://code.launchpad.net/~mir-team/unity-mir/development-branch/+merge/220249
[12:43] <Wellark> mzanetti: so I have a checkbox when setting the thing "use alphabeticals" ?
[12:43] <mzanetti> Wellark: I have no idea how settings is gonna handle that
[12:43] <greyback> Saviq: the "unmerged revisions" thing looks wrong now in LP
[12:43] <Saviq> greyback, I can see that
[12:43] <Wellark> mzanetti: sweet. ok.
[12:43] <Wellark> let's see then :)
[12:44] <mzanetti> Wellark: yeah... that might be the point when we can drop the autoconfirm again when designer see it sucks :D but right now there's no point in discussing it again
[12:45] <Saviq> greyback, it merges fine, so...
[12:45] <mzanetti> Wellark: and with the SIM pin you're not affected anyways
[12:45] <Saviq> greyback, I think it'll sort itself out once we follow a certain workflow
[12:45] <greyback> Saviq: okey dokey
[12:46] <Wellark> mzanetti: well, I'm just concerned about the PUK :)
[12:46] <mzanetti> Wellark: why?
[12:46] <mzanetti> Wellark: just set min: 8, max: 16
[12:46] <Wellark> because it's fixed 8 digits
[12:46] <mzanetti> Wellark: oh, is it?
[12:46] <Wellark> no. it has to be 8 digits and allowing entering 9 would be an error
[12:47] <Wellark> but anyway
[12:47] <mzanetti> well, in that case yes. I'm sure design *wants* the autoconfirm
[12:47] <Wellark> well, how many times have you typod a digit when using your phone?
[12:47] <Wellark> I'm doing it a lot
[12:47] <Wellark> but anyway
[12:47] <mzanetti> Wellark: their argument is: "you don't want to make the user press the OK button when its clear that's the only allowed action anyways"
[12:47] <Wellark> it's not fatal
[12:47] <Wellark> the user just has to cancel the dialog and start again if he notices he typoed the last one
[12:48] <mzanetti> Wellark: there's still a backspace
[12:48] <Wellark> not for the last digit
[12:48] <mzanetti> Wellark: you just can't make the typo on th every last character :D
[12:48] <Wellark> as it's autoconfirm
[12:48] <Wellark> ed
[12:48] <Wellark> yep
[12:48] <Wellark> don't do it :)
[12:48] <Wellark> I just don't see any major value on the autoconfirm
[12:49] <greyback> Saviq: I need to change the commit message as there's a shutdown fix in there too
[12:49] <Wellark> or the value is so minor that it's almost like design micro-optimization ;)
[12:49] <Wellark> mzanetti: anyway. see my last comment on the MR
[12:49] <Wellark> after that I'm fine with it
[12:50] <Saviq> greyback, didn't the shutdown change get released already?
[12:50] <greyback> Saviq: apparently not
[12:50] <Saviq> greyback, I'll do
[12:50] <greyback> Saviq: LP is telling the truth
[12:55] <Saviq> greyback, erm, sorry for jumping channels...
[12:56] <greyback> keepin' me on ma toes
[13:01] <tsdgeos> Cimi: it's a non free thing so no idea
[13:02] <mzanetti> Wellark: huh? Dialog is always modal. At least I can't interact with anything behind it while the dialog is shown
[13:08] <Wellark> mzanetti: well, there is a separate "modal" property and AFAIK it defaults to false
[13:08] <mzanetti> Wellark: just looked it up. it defaults to true
[13:09] <Wellark> mzanetti: is it documented to be the default?
[13:09] <Wellark> could you just humor me and make it explicit?
[13:09] <mzanetti> Wellark: yes: http://developer.ubuntu.com/api/qml/sdk-1.0/Ubuntu.Components.Popups.Dialog/#modal-prop
[13:09] <mzanetti> ok
[13:09] <Wellark> even though it's the default. defaults can change quite rapidly :)
[13:09] <Wellark> mzanetti: thanks!
[13:10] <mzanetti> Wellark: pushed
[13:14] <Wellark> mzanetti: thanks! btw, modal dialogs are generally evil. we should not default to them ;)
[13:15] <mzanetti> Wellark: hmm... this type of dialog really is intended to be a modal one... as it puts a black overlay above the whole application
[13:16] <Wellark> mzanetti: only if modal property is true
[13:16] <Wellark> the black overlay is controlled by the modal property
[13:16] <mterry> Saviq, hello!  Do you know if it's possible for me to fake SIM unlocking?
[13:16] <Wellark> mzanetti: approved.
[13:16] <mzanetti> Wellark: thanks
[13:17] <Wellark> mzanetti: what's the convention with unity8 ?
[13:17] <mzanetti> Wellark: regarding?
[13:17] <Wellark> will I top-approve now or only after the testing is done?
[13:17] <Wellark> I can't type that fast!
[13:17] <tsdgeos> mhr3: pstolowski: how's the departments api going? Is there anything i could start basing off? I remeber we had that API sketch but the you guys said you wanted to change it
[13:17] <Wellark> :)
[13:17] <mzanetti> Wellark: no, you copy paste the checklist from here: https://wiki.ubuntu.com/Process/Merges/Checklists/Unity8
[13:17] <tsdgeos> Saviq: do you know if the api is done? ↑↑
[13:18] <mzanetti> Wellark: and fill it in. Then you top approve
[13:18] <Saviq> tsdgeos, http://pad.ubuntu.com/dash-departments-shell-api should be up-to-date now
[13:18] <mzanetti> Wellark: next time we get a wagon on the CI train, all the top-approved ones will get a ticket for the train
[13:18] <Saviq> tsdgeos, the "usage" part needs a bit fixing
[13:18]  * tsdgeos clicks
[13:21] <Saviq> tsdgeos, should we do a quick HO on how this should be used?
[13:22] <Saviq> tsdgeos, I'm afraid the QML part of that pad should probably be written from scratch
[13:22] <Saviq> since when I wrote it I assumed wrong
[13:22]  * Saviq deletes
[13:24] <Saviq> tsdgeos, mhr3 was writing the actual thing in -scopes-shell to see if everything fits, but we should be ok to start the UI
[13:27] <Saviq> mterry, let me know what I can do to help with the SIM lock thing
[13:27] <mterry> Saviq, pitti told me in #ubuntu-devel I can fake it with the ofono phonesim tools
[13:27] <Saviq> mterry, I was hoping that would be the case
[13:31] <tsdgeos> Saviq: sure, let's do it
[13:32] <Saviq> tsdgeos, https://plus.google.com/hangouts/_/grqpoxzwvq3rsrsdwfpnhbq2qia
[13:50] <dandrader> I should have only a single unity-system-compositor process running, right?
[13:50]  * dandrader sees 5 after restarting unity8 a nmber of times during develpment
[13:51] <dandrader> after "sudo stop lightdm" I now have 6...
[13:53] <mterry> Saviq, so using phonesim, unlocking appears to work in both greeter and session.
[13:53] <mterry> Saviq, so what was your sequence of actions?
[13:57] <mterry> Unless maybe the phonesim is masking some issue that real SIMs have
[13:58] <dandrader> mzanetti, so the qt comp. ppa rebuilds automatically whenever it sees a new unity8 (or qtmir) revision?
[13:59] <sil2100> pstolowski, mhr3, pete-woods: hi guys!
[13:59] <sil2100> pstolowski, mhr3, pete-woods: so, I did the landing of unity-scopes-api and unity-scopes-shell and it seems to be breaking autopkgtests for unity-scope-click
[13:59] <Mirv> Saviq: tsdgeos: ok scrap it, I'll work with the symbols hell in parallel so that we'd actually have some packages ready soon
[13:59] <sil2100> pstolowski, mhr3, pete-woods: and to be more clear - it seems to not build correctly
[14:00] <sil2100> pstolowski, mhr3, pete-woods: it's failing on unit-tests
[14:00] <pete-woods> sil2100: that's the unity-scope-click tests that are failing?
[14:00] <sil2100> pete-woods: yes, you can see the failure here:
[14:00] <sil2100> https://jenkins.qa.ubuntu.com/view/Utopic/view/AutoPkgTest/job/utopic-adt-unity-scope-click/lastBuild/ARCH=amd64,label=adt/console
[14:01] <Mirv> meaning that I'll disable symbols for one build and continue trying to get them all updated in the background
[14:01] <pete-woods> sil2100: yes, I think I can guess why, sorry about that, we'll get right on it
[14:01] <sil2100> pete-woods: where will the fix be needed? :)
[14:02] <pete-woods> sil2100: it's (probably) the test fixtures, we renamed some of the config file options
[14:02] <pete-woods> sil2100: so in this case their test cases will need tweaking
[14:03] <pete-woods> sil2100: fortunately that won't affect the already released packages
[14:03] <sil2100> pete-woods: awesome, if it's unity-scope-click which will need a release, just fill a new landing with that and we'll try releasing
[14:05] <mzanetti> dandrader|afk: it builds automatically once a day if there's a change
[14:05] <mzanetti> dandrader|afk: if you want more builds, go to the build recipe and kick it manually
[14:06] <pete-woods> sil2100: will obviously have to confirm it's actually that first
[14:06] <pete-woods> before we get too happy
[14:06] <mzanetti> dandrader|afk: e.g.: https://code.launchpad.net/~unity-team/+recipe/unity8-qtcompositor
[14:08] <sil2100> pete-woods: thanks :)
[14:12] <tsdgeos> Mirv: i'm confused, we released the new 5.2.1 with the displayMargin but didn't release the new unity8? I thought we were doing it together
[14:13] <pstolowski> pete-woods, I wonder if recently aded strict checks for empty values are not culprit as well, I see an exception for set_uri
[14:13] <pete-woods> pstolowski: I think you're right
[14:13] <Saviq> mterry, for me it seemed to be unlocked straight away (without me doing anything) when I booted the phone...
[14:13] <pete-woods> they don't have any ini file test fixtures
[14:13] <Saviq> mterry, but it was not, really
[14:14] <Saviq> mterry, I'll try again with the new packages soon
[14:15] <mterry> Saviq, huh, yeah I get the "unlocked" message at top appropriately, and it all seems to do the right thing (but I'm not familiar with unlocking SIMs and maybe the phonesim isn't matching your realsim behavior)
[14:16] <Cimi> mhr3, office tomo?
[14:16] <mhr3> pete-woods, see #ferrets
[14:16] <Saviq> mterry, the "real" thing should be: 1. boot phone, panel says "SIM Locked"; 2. in network menu, tap "Unlock SIM"; 3. PIN UI shows up and you unlock your SIM; 4. the "SIM Locked" message is replaced with signal strength icon
[14:17] <mhr3> sil2100, the tests were misusing a mock feature, it's being fixed
[14:17] <mhr3> sil2100, tests in -click
[14:17] <mhr3> sil2100, do we just add it to that silo?
[14:17] <mterry> Saviq, yeah, that's what I see then
[14:17] <sil2100> mhr3: yeah, let's do that this way, I'll reconfigure
[14:17] <Saviq> mterry, ok, I'll test it again in a few
[14:18] <sil2100> mhr3: please build it then and test ;) Sicne we're low on silos, I wouldn't be able to assign a new one for you ;p
[14:18] <mhr3> sil2100, well i guess the test is that it builds
[14:25] <Mirv> tsdgeos: so there's a problem with unity-scopes-api mhr3 and sil2100 are trying to deal with. the scopes-api broke unity-scope-click autopkgtests so it also prevents unity8 migration :(
[14:25] <tsdgeos> oh :/
[14:25] <Mirv> tsdgeos: and that's a problem of course since there wasn't a declared dependency, but let's see if it's sorted out quickly now
[14:25] <sil2100> Mirv, tsdgeos: yeah, so mhr3 and pete-woods have a fix in the works
[14:28] <Saviq> mterry, btw, the gatekeeper seems to disagree with you on the ap tests passing: http://q-jenkins.ubuntu-ci:8080/job/autopilot-release-gatekeeper/140/label=daily-mako/testReport/junit/unity8.shell.tests.test_notifications/InteractiveNotificationBase/test_modal_sd_with_greeter_Native_Device_/
[14:28] <Saviq> mterry, I mean http://q-jenkins.ubuntu-ci:8080/job/autopilot-release-gatekeeper/140/#showFailuresLink
[14:28] <Saviq> mterry, this is a job that basically runs everything smoke testing runs
[14:29] <Saviq> mterry, on a silo
[14:29] <mterry> Saviq, ps-jenkins was passing the ap tests.  I'll retest
[14:29] <Saviq> mterry, looks like all the unity8 failures say the same - ligthdm mock seems to be misplaced or something
[14:29] <Saviq> mterry, yeah, please get the packages and use phablet-test-run -n unity8 to run them on deviec
[14:30] <Saviq> mterry, not sure -ci does the same (but probably should!)
[14:30] <mterry> Saviq, also I'm looking at that "unity8-greeter spins" issue you saw on manta -- looks like unity8-greeter is taking about 10% cpu constantly.  Unacceptable, and I'm trying to figure out why
[14:30] <Saviq> mterry, yah it was a whole core for me, so slightly different, but agree very relevant
[14:31] <Mirv> Saviq: tsdgeos: sadly the qtubuntu-* rebuilds suggested by jhodapp did not bring functional video playback or a camera (still black) on Qt 5.3
[14:31] <Saviq> mterry, maybe there's just an issue with how the gatekeeper runs tests, I'll actually run the suite here in a bit, too
[14:31] <tsdgeos> Mirv: may it has to do with the same blackness that makes the launcher black
[14:31] <tsdgeos> s/may/maybe
[14:31] <jhodapp> Mirv, do you get the mediaplayer-app's UI or is it completely black?
[14:33] <Mirv> jhodapp: UI is there, and you can click play/pause which switches its mode
[14:34] <jhodapp> Mirv, ok
[14:37] <kgunn> mterry: how critical is the life-cycle exposure for usc ?...like is that gonna hold up landing split greeter ?
[14:37] <mterry> kgunn, no
[14:38] <kgunn> mterry: thanks...camako has been "distracted" :)
[14:38] <mterry> kgunn, I think it was related to boot animations, which I dropped from the split landing
[14:40] <kgunn> dandrader: are you testing on n4 and n7 ? or just n4?
[14:42] <Mirv> I now filed a bug against media-hub (the first place I can think of about that) bug #1321319 since media playback is definitely something is needed for testing 5.3 properly
[14:42] <dandrader> kgunn, just n4
[14:42] <Saviq> mterry, hmm there's still something wrong for sure... don't you get a GSM icon on manta? (I do on both manta and flo)
[14:42] <Saviq> mterry, and managed to get black session on manta again
[14:42] <mterry> Saviq, I was testing sim unlock on mako...  Let me try manta
[14:42] <mterry> Saviq, black session?  you mean with an app up?
[14:43] <kgunn> dandrader: ok, i'm seeing a bit more weirdness on n7
[14:43] <Saviq> mterry, even with no app
[14:43] <Saviq> mterry, dash is black for me at the moment until full unlock is done
[14:43] <mterry> Curious...
[14:43] <dandrader> kgunn, what?
[14:43] <mterry> Saviq, I'll resume testing on manta, maybe there is a device difference
[14:44] <dandrader> kgunn, i mean, like what
[14:44] <Saviq> mterry, k, and I'll look again on mako after I land silo 18
[14:44] <kgunn> dandrader: in some case (not sure what makes it happen), the app area turns a full blank yellow
[14:44] <kgunn> i thot maybe i launched notepad...
[14:45] <kgunn> but then rotated it back, and view returned to normal
[14:45] <dandrader> kgunn, it's not yellow, it's khaki :)
[14:45] <kgunn> :)
[14:45] <dandrader> kgunn, it's the background color of the stage
[14:45] <dandrader> kgunn, I guess it should only appear when something goes really wrong, right mzanetti?
[14:47]  * mzanetti ready scrollback
[14:47] <mzanetti> reads
[14:47] <kgunn> dandrader: also, i dunno how, but it got into spread mode (at least i didn't mean to enter it) and i couldn't get out of it
[14:48] <mzanetti> dandrader: can't figure what the question is from the scrollback
[14:50] <Saviq> mterry, ah, and again on manta I get two voicemail messages that are only in session... and why are they there at all in the first place?
[14:50] <mterry> Saviq, oh you're saying that you never got calls at all -- I understand why the voicemails are odd to you
[14:50] <Saviq> mterry, well, it's manta and flo ;)
[14:50] <Saviq> mterry, no calls to be had there
[14:51] <dandrader> mzanetti, if you see the khaki color of the stages Rectangle it means something is really broken
[14:51] <mterry> Saviq, right...  odd
[14:51] <Saviq> mterry, it looks like it's some test message injected
[14:51] <mterry> Saviq, is this with phonesim?
[14:51] <Saviq> mterry, no, fresh flash + silo 002
[14:51] <mterry> hmm
[14:51] <Saviq> mterry, at least not that I know of
[14:51] <Saviq> mterry, but yeah, it seems to think it has a phone
[14:52] <Saviq> mterry, and a voicemail
[14:52] <Saviq> mterry, on all my devices, for that matter
[14:52] <Saviq> mterry, which could explain my SIM issues as well
[14:52] <mterry> but only with the split greeter version...  odd
[14:52] <Saviq> mterry, if there's a fake SIM inserted
[14:52] <Saviq> mterry, yeah
[14:52] <Saviq> mterry, carrier says "oFono (T-MeeGo)"
[14:53] <Saviq> mterry, so sounds like it's a fake SIM indeed
[14:53] <greyback> dandrader: yes something is really broken :)
[14:53] <mterry> Saviq, I assume you don't have ofono-phonesim installed (i.e. silo doesn't pull it in)
[14:53] <Saviq> mterry, you assume wrong
[14:53] <Saviq> mterry, I do have it installed, so silo must be pulling it in
[14:53] <mterry> Saviq, what?  We install phonesim?
[14:53]  * mterry installs fresh to confirm
[14:54] <mterry> Saviq, well at least that's an explanation of something...
[14:54] <Saviq> mterry, indicator-network-autopilot pulls it in
[14:54] <mterry> Why is the autopilot installed though?
[14:55] <Saviq> mterry, ah wait
[14:55] <Saviq> mterry, I think I see what's the problem
[14:55] <Saviq> mterry, robru's citrain-push installs *all* binary packages...
[14:55] <Saviq> mterry, from a silo
[14:55] <Saviq> instead of just dist-upgrading from it :|
[14:55] <mterry> Ah..  I use add-apt-repository
[14:56]  * Saviq shakes fist at robru
[14:56] <Saviq> ;)
[14:56] <Saviq> mterry, ok, ignore me for now, let me clear this thing up
[15:00] <Saviq> mterry, ah, I don't think I understood your "hack" for locale in greeter, though
[15:00] <mterry> Saviq, understood or understand?  Meaning you want clarification or you figured it out?
[15:01] <Saviq> mterry, if the current design is that user's locale should not affect the greeter, we should make it so (although we have no way to set system locale on phone do we)
[15:01] <Saviq> mterry, understand, sorry :)
[15:02] <mterry> Saviq, current design for multi-user and phone are a little different.  My design mandate on phone is "don't introduce any changes" and for tablet mode "by default do what the desktop does".  Hence my "hack".  But I need to see clarification from design on whether the inherited behavior of no-switching in multi-user is what they want or not
[15:03] <Saviq> mterry, well, I don't think we can keep it separate for much longer
[15:03] <Saviq> mterry, when we split the greeter, people will go into multi-user mode pretty soon
[15:03] <Saviq> on phone, too
[15:04] <Saviq> mterry, so we should converge on something (and if system language change is required to change the greeter language, we need to make this possible somehow)
[15:04] <mterry> Saviq, I get that.  But both behaviors create design problems (having two language settings or having greeter language flip-flopping)
[15:04] <mterry> Saviq, so we just need to lock a designer in a room
[15:04] <Saviq> mterry, let's!
[15:05] <Saviq> mterry, in the mean time I don't think we should introduce the hack
[15:05] <Saviq> mterry, as it's not going to affect the indicators anyway
[15:05] <mterry> Saviq, it will
[15:05] <Saviq> mterry, oh will it? so you pull in lang from user 32011 to the greeter session?
[15:05] <mterry> Saviq, yup
[15:06] <Saviq> mterry, oh ok then...
[15:06] <mterry> Saviq, silo as it is now, should have no language regressions
[15:06] <Saviq> mterry, big FIXME provided?
[15:06] <mterry> Saviq, yeah, there is a large comment
[15:06] <Saviq> mterry, ok, that's fine then
[15:06] <Saviq> mterry, that clears things up
[15:06] <Saviq> mterry, do we have a greeter session in Malta?
[15:06] <Saviq> re: locking a designer in a room
[15:06] <mterry> well, I suppose it "regresses" in multi-user mode, but only because previous behavior was dumb and fake
[15:07] <mterry> kgunn, ^ do we have a designer meetup for greeter issues in malta?  That would be nice
[15:08] <Saviq> mterry, "split greeter design sign off"
[15:08] <Saviq> mterry, but it's only 15 mins
[15:08] <kgunn> mterry: yep, what he said
[15:08] <mterry> Saviq, eh, there's nothing to sign off for the current silo really
[15:08] <Saviq> mterry, that's good ;)
[15:08] <mterry> Saviq, oh that's a currently scheduled meeting?
[15:08] <Saviq> mterry, yes
[15:08] <mterry> that means we can't land before that.  Bummer
[15:08] <kgunn> mterry: i spoke to john  he said land it
[15:08] <Saviq> mterry, we can expedite it
[15:09] <mterry> kgunn, then what is that meeting?  :)
[15:09] <kgunn> we'll use malta as a followup, treat anything as bugs
[15:09] <kgunn> that design finds to be an issue
[15:10] <Saviq> robru, one for you: bug #1321333
[15:11] <Saviq> kgunn, btw, we talked with greyback this morning, you said unity-mir/devel should only be used to adapt to mir/devel, why do you think that?
[15:12] <kgunn> Saviq: b/c it would slow down landing other things (app closing bug is a good example)
[15:12] <kgunn> and i couldn't get ricardo to adopt the same
[15:12] <kgunn> with papi devel
[15:12] <kgunn> so if i can't get all 3 (usc, papi, um)
[15:13] <Saviq> if only we used a proper VCS...
[15:13] <Saviq> we could just cherry-pick into trunk
[15:13] <kgunn> i know :-/
[15:13] <Saviq> at release time
[15:13] <kgunn> Saviq: its not that bad...just means discipline for those wanting to pre-build with staging to frequently rebase
[15:13] <kgunn> ...or sorry, merge trunk
[15:14] <Saviq> touché
[15:14] <Saviq> kgunn, we felt like this would cause confusion on what to put into trunk, what into devel...
[15:14] <Saviq> kgunn, greyback, but maybe it's fine OTOH
[15:15] <Saviq> but we should then rename lp:unity-mir/devel to "mirdevel" or something
[15:15] <kgunn> Saviq: i thot it was simple (which is unusual...that i think something is simple :)
[15:15] <kgunn> +1
[15:15] <kgunn> that's what it becomes
[15:15] <Saviq> and not actually have it as a series
[15:15] <kgunn> yes
[15:15] <Saviq> that branch is going to be neglected so much :|
[15:16] <kgunn> it should end up being used by mir devels...and a few on unity
[15:16] <kgunn> right?
[15:16] <Saviq> well, sure
[15:16] <Saviq> but I fear it's not gonna follow trunk
[15:16] <kgunn> ...maybe ci airline magically happens and then this doesn't matter :)
[15:16] <Saviq> unless something breaks
[15:16] <Saviq> and then someone will go "oh, we didn't merge trunk for a month, let's"
[15:17] <kgunn> yep, that is true...but iff no one is using staging
[15:17] <kgunn> which if that's the case...then
[15:17] <kgunn> its  not really useful/needed
[15:18] <kgunn> and actually....
[15:18] <Saviq> right, but you wanted devel to be built into Mir staging PPA
[15:18] <tsdgeos> mzanetti: i pushed a few small commits to https://code.launchpad.net/~aacid/unity8/use-unity-api/+merge/219222 to address your comments :)
[15:18] <mzanetti> tsdgeos: ack
[15:18] <kgunn> yeah...but look at the recipe...its devel + merge trunk
[15:18] <kgunn> so you do get the whole thing
[15:18] <Saviq> yeah ok that helps
[15:19] <greyback> Saviq: kgunn: I don't have a strong opinion, I can go either way
[15:19] <Saviq> assuming greyback doesn't want lp:unity-mir/devel for non-mir reasons :)
[15:19] <Saviq> I'd actually be happier without /devel
[15:19] <greyback> it's not quite the traditional - work in devel, release to trunk when you want - method, but it's fine, I can live with it
[15:20] <Saviq> I don't see it solving much, while it does bring issues of its own
[15:20] <Saviq> like having to prep a release branch
[15:20] <greyback> Saviq: sure, me too, if it wasn't for Mir changes. unity-mir/devel useful to keep up with mir/devel
[15:20] <kgunn> it mainly the mir team
[15:21] <kgunn> i was just chatting with AlbertA, that maybe we should just work on good cross-compile scripts
[15:22] <dandrader> greyback, going work now on fixing the indicators bar. full screen vs. non fullscreen apps and have it behaving correctly on app launch and restore. is that gonna conflict with anything you're doing?
[15:22] <dandrader> or plan to do
[15:22] <greyback> dandrader: nope, go right ahead
[15:22] <dandrader> greyback, ok
[15:23] <greyback> dandrader: let's use this to keep track: http://pad.ubuntu.com/PfyWZQxyTs - do add other tasks you think need doing
[15:24] <dandrader> greyback, i've been using the blueptrint :) https://blueprints.launchpad.net/ubuntu/+spec/client-1410-unity-ui-rotation
[15:24] <Saviq> greyback, ok, I vote for dropping the devel series of unity-mir in that case, and keeping a lp:~unity-team/unity-mir/devel-mir or so, just for Mir adaptations, and synced landing with Mir releases
[15:24] <greyback> dandrader: the reason I hesitate to use a blueprint is that this code isn't being released just yet
[15:24] <mterry> Saviq, mzanetti: do you folks know good ways to debug what qml is doing?  I think some animation or something is sucking up CPU
[15:25] <dandrader> Saviq, kgunn, thoughts ^^
[15:25] <dandrader> ?
[15:25] <greyback> Saviq: I think that's an excellent topic for next week :)
[15:25] <Saviq> greyback, ;)
[15:25] <Saviq> mterry, profile it
[15:25] <dandrader> kgunn, Saviq (on the blueprint vs pad discussion)
[15:25] <mterry> hm
[15:26] <mterry> Saviq, oh!  we have console.profile() built in support. that's nice
[15:26] <Saviq> mterry, you pass -qmljsdebugger=foo
[15:26] <Saviq> mterry, and then you can connect to the port you selected with QtC
[15:26] <Saviq> mterry, it will show you what's happening
[15:27] <Saviq> mterry, our run_on_device does that for example
[15:27] <mterry> Saviq, oh dang nice
[15:28] <josharenson> kgunn, btw there was a noticeable performance drop running glmark2 with shell rotation enabled. There are many factors however, so I'm investigating.
[15:29] <greyback> josharenson: I'd expect a performance drop, since rotated shell completely replaces the renderer
[15:29] <greyback> and we've not profiled/optimized the replacement renderer
[15:29] <josharenson> greyback, ack
[15:29] <greyback> but I'd still expect a perf hit nevertheless
[15:30] <kgunn> dandrader: greyback ...blueprints please if you don't mind
[15:30] <greyback> kgunn: ack
[15:30] <kgunn> remember white board section perfectly open for abuse
[15:31] <kgunn> josharenson: greyback ..ack on the perf drop....the good thing is, now we're aware ;)
[15:31] <greyback> kgunn: indeed
[15:32] <greyback> josharenson: can you give me a rough idea of the perf hit? Percentage?
[15:32] <josharenson> :-) kgunn, greyback it was't _huge_ and I think without framed dropping enabled, we still hit 60 FPS... I wanna say ~5%
[15:33] <greyback> josharenson: that I can live with :)
[15:33] <kgunn> yep, not bad
[15:35] <Saviq> greyback, btw, I feel like we should SIGCONT apps before upstart-stopping them
[15:35] <Saviq> greyback, otherwise we wait 5s before upstart SIGKILLs them, and cause unnecessary dirty shutdowns
[15:36] <Saviq> _or_ we should make upstart SIGKILL them straight away, if we don't want to resume them on stop
[15:36] <greyback> Saviq: true, please log bug for that so I don't forget
[15:36] <Saviq> greyback, will do
[15:42] <kgunn> Saviq: i'm otp, but yeah agree with your vote on unity-mir/mir-devel
[15:42] <greyback> I'm ok with it too
[15:48]  * tsdgeos does an early-ish EOD, need to go to the ophtalmologist
[15:48] <tsdgeos> see you tomorrows
[15:54] <dandrader> Saviq, btw rebased unity8/mirCompositor against latest trunk the other day and it still seems to take over a second to relayout due to a resize (landscape<->portrait)
[15:57] <Saviq> dandrader, ok, wanna file a bug or track it in a workitem in a silo?
[15:57] <Saviq> dandrader, ah, didn't answer on bp vs. pad
[15:57] <Saviq> dandrader, for things that progress really fast
[15:57] <Saviq> dandrader, I think it's fine to have a "big" item for unity-ui-team or so in a blueprint
[15:58] <Saviq> dandrader, and link to a pad with breakdown
[16:00] <dandrader> Saviq, I thought blueprint work items were meant for those "fast moving" items
[16:00] <Saviq> dandrader, I know, I know, you hate bps ;)
[16:00] <dandrader> linking to a pad kinda defeats the purpose, imho
[16:00] <Saviq> dandrader, as we know the overhead of updating workitems is a bit big
[16:01] <dandrader> Saviq, I think that updating bp work items is fine in terms of overhead
[16:02] <dandrader> although it might spam people with e-mails :)
[16:02] <greyback> dandrader: helps you look busy ;)
[16:02] <Saviq> yeah, that, and then you can only put so much info in a workitem without it becoming unreadable
[16:03] <dandrader> Saviq, that's true. I've noticed that the "work items" field is rather inflexible when it comes to formatting
[16:03] <Saviq> indeed, and then it takes a few seconds to parse/confirm
[16:04] <Saviq> mterry, hmm "phablet, radio, system" showed up as available users on my flo :)
[16:04] <Saviq> and on manta, that expected?
[16:04] <mterry> Saviq, not expected, and I don't see that on manta
[16:04] <Saviq> heh
[16:05] <Saviq> ok, I got more upgrades than wanted (distro moved to new qt, for example), maybe that's related again..
[16:10] <mhr3> Cimi, oh, and yea, will probably come
[16:10] <mhr3> Cimi, maybe just for the afternoon
[16:14] <Saviq> mterry, these are the additional packages installed when I upgrade from the silo, is that expected http://pastebin.ubuntu.com/7493587/ ?
[16:14] <mterry> Saviq, yeah
[16:14] <Saviq> mterry, ok
[16:15] <dednick> Saviq: hey. Shoud CI for merges into unity-mir/devel be building against mir/devel?
[16:15] <Saviq> dednick, not likely, no
[16:16] <Saviq> dednick, we'd need to put that in a stack with mir... not sure if stacks are at all supported still
[16:17] <dednick> Saviq: i c. so the devel in this case is a bit pointless?
[16:17] <Cimi> dandrader, on https://code.launchpad.net/~dandrader/unity8/unusedBackgroundImage/+merge/220288
[16:17] <Saviq> dednick, feels like it, yeah
[16:17] <Saviq> dednick, maybe it should build against the staging PPA
[16:17] <Cimi> shouldn't we have some wallpaper between the header and the rest of the mainview?
[16:17] <Cimi> Saviq, too on ^
[16:17] <Cimi> from designs I saw
[16:17] <Saviq> Cimi, that's going awayt
[16:17] <Saviq> -t
[16:18] <Cimi> ok then
[16:18] <Cimi> looked cool though
[16:19] <Saviq> mterry, so, unlock SIM indeed works when I don't have phonesim ;)
[16:20] <Saviq> mterry, only the UI (when unlocking in greeter) looks bad (on dark grey background for some reason)
[16:20] <Saviq> mterry, and edges are still available and such
[16:20] <mterry> Saviq, I didn't notice bad background, but I may have just been distracted by fact that it worked at all :)
[16:21] <dednick> kgunn: what is the mir staging ppa used for?
[16:22] <kgunn> dednick: its meant to be a recent build of mir-devel & unity-mir/papi/usc to go with it...
[16:23] <dednick> kgunn: i c. so CI for mir depended devel branches (unity-mir/devel, usc/devel) should be using that ppa?
[16:23] <Saviq> I'd say so, yeah
[16:24] <Saviq> mterry, it's actually bad background inside session, too
[16:24] <mterry> Saviq, and that's specific to the silo?
[16:25] <Saviq> mterry, yeah, only common bug is bug #1171918
[16:25] <Saviq> WRONG
[16:25] <Saviq> bug #1308011
[16:26] <Saviq> mterry, ↑
[16:26] <mterry> k...
[16:26] <mterry> I will look at its background
[16:26] <Saviq> mterry, it should be the wallpaper
[16:27] <Saviq> mterry, "Shell.qml:406: ReferenceError: greeter is not defined" sounds relevant
[16:27] <Saviq> it's probably looking for the background there
[16:27] <Saviq> aaanyway
[16:27] <mterry> hmm in Shell.qml?  Sounds like a missed reference in a merge
[16:28] <mterry> Shell.qml is always trying to sneak in references to greeter  :)
[16:28] <Saviq> mterry, I'll sort my devices out and do a proper functional review tomorrow morning
[16:28] <mterry> Saviq, yeah, I'll fix this 10% cpu issue and background thing
[16:34] <Saviq> mterry, some update from distro must be causing the added system and radio users
[16:35] <Saviq> as I don't get those when upgraded just the silo packages
[16:36] <mterry> Saviq, but you do if you do a full dist-upgrade?
[16:36] <Cimi> do-release-upgrade -d finally works /me upgrades
[16:36] <Saviq> mterry, yes
[16:36]  * Saviq tries again
[16:36] <mterry> sigh
[16:38] <Saviq> mterry, yeah, accountsservice got an upgrade
[16:38] <Saviq> mterry, looks related
[16:38] <mterry> son of a
[16:39] <Saviq> mterry, when it happens, I can tap on users behind the edge demos
[16:39] <mterry> Saviq, noted...  will look into that
[16:40] <robru> mterry, gonna need to rebuild unity-mir in silo 2, you ready for that?
[16:40] <mterry> robru, uh sure
[16:40] <robru> mterry, heh, k. just published unity-mir from silo 18, will rebuild once it lands
[16:40] <Saviq> black screen :|
[16:40] <mterry> Saviq, behind the greeter?~
[16:40] <Saviq> mterry, whole
[16:40] <mterry> Saviq, after logging in?
[16:41] <Saviq> mterry, upstart and greeter .crash
[16:41] <mterry> great...
[16:41]  * mterry climbs to the roof to fiddle
[16:41] <Saviq> wait, those are old
[16:42] <Saviq> so I won't get the new crash files then...
[16:43] <Saviq> mterry, confirmed, accoutnsservice upgrade adds users
[16:44] <Saviq> aaand /me out
[16:45] <mterry> Saviq, bye
[17:18] <mterry> How do I connect QtCreator to my device?  I press the "Autocreate a kit" button in QtC but it doesn't do anything
[17:55] <Saviq> mterry, for profiling you don't need that
[17:56] <mterry> Saviq, aren't you supposed to be gone?
[17:56] <Saviq> mterry, shh
[17:56] <Saviq> mterry, for profiling you just need to select the remote connection from the menu bar... that I can't access now 'cause QtC crashes here due to 5.3
[17:57] <mterry> Saviq, hmm, I don't have a menu bar.  I tried going to Analyze then Qml Profiler (External) but didn't have any luk
[17:58] <Saviq> mterry, yeah, that's it
[17:58] <mterry> Saviq, wasn't working for me, and it had the kit dropdown, so I figured kit was my problem
[17:58] <Saviq> mterry, nah
[17:58] <mterry> something else must be wrong then, I'll try and figure it out
[17:59] <Saviq> mterry, kits are for building
[18:05] <Saviq> mterry, ah, it looks like it changed indeed
[18:05] <Saviq> mterry, before you could just point it at a file
[18:05] <Saviq> mterry, but you can still fake it without a kit
[18:05] <Saviq> mterry, use qmlprofiler on the console to connect to the device
[18:05] <Saviq> mterry, and dump the trace
[18:06] <Saviq> mterry, then you can load it in QtC through Analyze → QML Profiler Options → Load trace
[18:06] <mterry> Saviq, ah cool, didn't know about qmlprofiler
[18:08] <Saviq> mterry, FYI: kicked unity-mir and unity8 rebuild in silo 002 as we landed two silos
[18:08] <mterry> Saviq, cool thanks
[18:16] <Saviq> mterry, in any case, if kit autocreation doesn't work for you, a bug against qtcreator-plugin-ubuntu would be in order
[18:23] <ChrisTownsend> luv: Hi, I have heard that you may be working on a fix for https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/1063617 ?
[18:32] <luv> ChrisTownsend: hiya ...
[18:32] <luv> good to get in touch with you as the issue is assigned to you
[18:33] <luv> I can post my patch to the issue later today (in about two hours time)
[18:33] <luv> I spent about 30 hours in gdb on this one :-D
[18:33] <ChrisTownsend> luv: Whoa!  Thanks for looking at this.  I just want to make sure we don't duplicate effort here.
[18:34] <luv> I guess the patch will be a good starting point.
[18:34] <ChrisTownsend> luv: Ok, cool, thanks.  I'll look out for it and then we can go from there.
[18:35] <luv> Thanks.
[18:35] <ChrisTownsend> luv: btw, I can't reproduce the issue, so I'm kind of going blind here, but it affects enough people to warrant trying to get this fixed.
[18:37] <luv> Strange. Easiest way to reproduce I think is to set a custom shortcuts for closing a window or "toggle maximization" or set "focus follows mouse" and then run "unity" few times or start "ccsm" or click "Advanced Search" in ccsm
[18:38] <luv> the bug is in libcompiz and gnome integration (compiz-gnome) ... running ccsm proved to be the easiest way to reproduce for me (on up to date trusty)
[18:38] <ChrisTownsend> luv: Yeah, I've set some custom shortcuts and have rebooted the machine numerous times and it stays intact.  I even tried a new user with the same result.
[18:39] <ChrisTownsend> luv: I'll try just restarting Unity/Compiz a bunch of times, maybe that'll do it.
[18:40] <luv> The code is clearly wrong (to my eye) though ... there is a function which is setting the gsettings backend keys ... and it compares if the new key and old key are same ... If they are not same, the new one is set. If they are same - it resets the key.
[18:40] <ChrisTownsend> luv: Uh, yeah, that doesn't sound correct.
[18:41] <ChrisTownsend> luv: Well, I look forward to the patch.  Thanks again!
[18:41] <luv> The other bug is in updateSetting (gsettings/src/gsettings.c) - readOption is used when readIntegratedOption should be used (it's correct in readSetting) in the same file.
[18:42] <luv> Yeah - the patch is chaning about 8 lines :-). Looking forward to (finally) post it online :-)
[18:44] <ChrisTownsend> luv: Awesome!  That should also be SRUable too, so we can get this in 14.04.
[18:49] <ChrisTownsend> luv: Ok, just reproduced the issue.  Thanks for the hint.
[22:21] <robru> Saviq, hey, you around? I've been working on a new citrain tool, it does dist-upgrades like you asked. let me know what you think: http://paste.ubuntu.com/7495022/