=== jono is now known as Guest50754
=== alan_g is now known as alan_g|afk
=== ted is now known as tedg
=== alan_g|afk is now known as alan_g
=== tedg is now known as ted
=== ted is now known as tedg
[06:48] <Saviq> Cimi, mterry, please have a look at bug #1322233 - that's the last failing one
[06:48] <ubot5> bug 1322233 in Unity 8 "Greeter tests fail and can't be discerned" [Critical,Triaged] https://launchpad.net/bugs/1322233
=== charles is now known as charles``
=== charles__ is now known as charles
[06:55] <tsdgeos> Cimi: https://bugs.launchpad.net/unity8/+bug/1322233
[06:55] <ubot5> Ubuntu bug 1322233 in Unity 8 "Greeter tests fail and can't be discerned" [Critical,Triaged]
[06:56] <tsdgeos> Cimi: https://bugs.launchpad.net/unity8/+bug/1322233
[06:56] <tsdgeos> mzanetti: do you remember about https://bugs.launchpad.net/unity8/+bug/1297308 ?
[06:56] <ubot5> Ubuntu bug 1297308 in Unity 8 "new-scope's scopes-scope cards lack leftMargins and text-vCenter if no mascot is set" [Medium,In progress]
[06:59] <mterry> Saviq, tsdgeos, Cimi: those tests *may* be fixed by split greeter
[07:05] <mhr3> dednick, tsdgeos closeby?
[07:05] <mhr3> make him appear here pls
[07:06] <dednick> mhr3: he's fiddling with unity8-desktop
[07:06] <mhr3> dednick, aight, just tell him that it's pushed
[07:06] <dednick> mhr3: he says thanks
[07:07] <dednick> mardy: ping.
[07:08] <mardy> dednick: hi!
[07:10] <dednick> mardy: hi. so i was looking at the signon-ui, and yeah. it's going to be a bit of a pain in the arse.
[07:11] <dednick> I haven't really got a solution yet.
[07:14] <mardy> dednick: yes, it's quite complex. I actually have a couple of files which implement a simple IPC based on stdin/stdout, but there's much more than that to be done, in order to split the UI into another process
[07:15] <mardy> dednick: do you think it would be difficult to export the mir connection from QGuiApplication? I guess that that would only involve changes to the QPA plugin, or is there more than that?
[07:15] <dednick> mardy: qpa uses platform-api, not mir directly.
[07:16] <mardy> dednick: ah, right, you even told me that yesterday :-)
[07:17] <dednick> mardy: actually, i'm just thinking now. I may have a solution. give me a minute.
[07:19] <dednick> mardy: we could create a separate process for the trust helper rather than the signon-ui, and add by pid.
[07:20] <paulliu> tsdgeos, http://paste.ubuntu.com/7534738/
[07:20] <mardy> dednick: right, that would be much easier
[07:21] <dednick> 1) dbus invoke requestAccess.
[07:21] <dednick> 2) create trust helper process (argv=originating app pid , signon-ui pid)
[07:21] <dednick> 2.1) create mir_connection
[07:21] <dednick> 2.2) start trust_session(originating app pid)
[07:21] <dednick> 2.3) trust_session_add_pid(signon-ui pid)
[07:21] <dednick> mardy: ^
[07:22] <dednick> 3) signon-ui process: create view for request
[07:22] <mardy> dednick: and for adding more clients later on (at the moment we don't need that, but we'll have for 14.10), it could just print the address of the mir socket to stdout, and signon-ui could read that and pass it to the new clients
[07:23] <dednick> mardy: um. that might be a bit more work...
[07:23] <mardy> dednick: yes, but it's not important now, and maybe by then we'll even have a better solution
[07:24] <mardy> dednick: anyway, thanks for your suggestion, it sounds indeed like something we could use now
[07:24] <dednick> mardy: cool.
[07:25] <dednick> mardy: the only problem may be that the signon-ui might be rejected by mir connection authorization. But we might be able to just add it to a whitelist.
[07:25] <dednick> for now.
[07:31] <mardy> dednick: why would it be rejected? Can't one add just any client to the session?
[07:34] <tsdgeos> paulliu: http://paste.ubuntu.com/7534797/
[07:34] <dednick> mardy: because it's not an "application" (not started with upstart/dekstop_file_hint)
=== josharenson1 is now known as josharenson
[07:43] <mzanetti> Wellark: enhanced PIN lockscreen now coming to an image near you
[07:44] <mhr3> Saviq, is u8 now supposed to be x-buildable?
[07:45] <mhr3> Saviq, i suppose no is the answer? http://paste.ubuntu.com/7534866/
[07:46] <Saviq> mhr3, not until SDK releases http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/staging/revision/1075
[07:46] <tsdgeos> paulliu: can you add the bug number you created for indicator-session at https://code.launchpad.net/~paulliu/unity8/logout/+merge/216373 ?
[07:49] <paulliu> charles, https://bugs.launchpad.net/ubuntu/+source/indicator-session/+bug/1323992
[07:49] <ubot5> Ubuntu bug 1323992 in indicator-session (Ubuntu) "Logout doesn't call com.canonical.Unity.Session.RequestLogout" [Undecided,New]
[07:49] <charles> paulliu, thanks. I'll investigate today
[07:52] <tsdgeos> bregma: in which room are you?
[07:53] <bregma> tsdgeos, ballroom
[07:53] <tsdgeos> bregma: do you have 2 minutes for us?
[07:53] <bregma> tsdgeos, it depends....
[07:53] <tsdgeos> :D
[07:53] <tsdgeos> so you have
[07:53] <bregma> yes
[07:53] <tsdgeos> there in a sec
[07:54] <mhr3> Saviq, hm, was looking through qtdeclarative, i didn't find a way to the override image provider for all http stuff
[07:54] <Saviq> mhr3, https://code.launchpad.net/~saviq/unity8/cache-network-data/+merge/221079 is a first step - at least for the network
[07:54] <Saviq> mhr3, but UITK does put all images through the @foo mechanism
[07:54] <Saviq> mhr3, so it must be possible there
[07:56] <mhr3> Saviq, @foo mechanism? what's that?
[07:56] <Saviq> mhr3, when you have a blah@20.png asset, it would get rescaled to the current GU automagically
[07:56] <Saviq> mhr3, @20 means the asset was prepared for a GU of 20px
[07:57] <mhr3> ah, interesting
[07:57]  * mhr3 looks at uitk too then
[08:12] <Mirv> I wonder if someone could land https://code.launchpad.net/~mzanetti/unity-api/fix-launcher-tests-with-5.3/+merge/220623 and https://code.launchpad.net/~afrantzis/qtubuntu/fix-1321189/+merge/220613 so that no manual builds would be needed for those in the Qt 5.3 PPA:s?
[08:12] <Mirv> in other words, they now regress every time they're updated without those branches in the archives
[08:12] <mzanetti> tsdgeos: can you pick the first one? ^
[08:13] <Mirv> it would also help to get https://bugs.launchpad.net/bugs/+bugs?field.tag=qt5.3 down two notches :)
[08:13] <mzanetti> meh... already approved. sorry
[08:13]  * mzanetti will see how to land it asap
[08:13] <mzanetti> thanks Mirv
[08:13] <Mirv> thanks mzanetti!
[08:14] <Mirv> for anyone craving for more polished Qt 5.3 packages, https://launchpad.net/~ci-train-ppa-service/+archive/landing-005 is the place - but on device you currently lose keyboard because of FTBFS bug #1323962
[08:14] <ubot5> bug 1323962 in maliit-framework (Ubuntu) "maliit-framework fails to build on Qt 5.3" [Undecided,New] https://launchpad.net/bugs/1323962
[08:14] <tsdgeos> mzanetti: sure
[08:14] <tsdgeos> done!
[08:14] <tsdgeos> that was fast!
[08:14] <mzanetti> :D
[08:15] <Mirv> functionally it should be ~identical to qt5-beta2 because it's the same code, but the abi:s are bumped, symbols are there so you actually catch these kind of dependency problems like what's needing a rebuild
[08:15] <Mirv> albert is teh fastest
[08:15] <mzanetti> he is
[08:15] <mzanetti> Mirv: I'm not the right one to review alf's branch
[08:15] <mzanetti> but will take care about the unity-api stuff
[08:16] <Mirv> mzanetti: ok
[08:16] <Mirv> rsalveti: would alf's branch https://code.launchpad.net/~afrantzis/qtubuntu/fix-1321189/+merge/220613 fall on your turf or is there anyone else to review/land it?
[08:20] <mhr3> tsdgeos, can you pull again? changed when endResetModel is emitted, looked wrong
[08:20] <tsdgeos> mhr3: sure
[08:20] <rsalveti> Mirv: I can take a look
[08:20] <Mirv> rsalveti: thank you
[08:23] <tsdgeos> mhr3: still looks good
[08:23] <mhr3> tsdgeos, coolio, separating into unity-api then
[08:23] <mhr3> tsdgeos, will ping you with mps once done
[08:28] <Cimi> Saviq, it's likely that tests fail for SDK change
[08:29] <Cimi> Saviq, there is no altered code in unity8 on those tests failing, but there was a big rework on textfield etc
[08:29] <Saviq> Cimi, k
[08:30] <Cimi> Saviq, two of them could be races, will check after break
[08:36] <mhr3> sil2100, ping?
[08:40] <sil2100> mhr3: pong, something's happening?
[08:40] <mhr3> sil2100, have some issue with hud, can we meet for a coffee and talk about it?
[08:41] <sil2100> mhr3: depends on what problems you have, since I don't know if I'm the right person to help ;)
[08:41] <Mirv> sil2100: in case needed: http://people.canonical.com/~tjyrinki/shorts/
[08:41] <mhr3> sil2100, you want a coffee anyway, dont you? :)
[08:42] <sil2100> Mirv: thanks! I guess I have access to s-jenkins today it seems, but this is more convinient
[08:43] <sil2100> robru: ^
[08:43] <sil2100> mhr3: ok, let's meet around coffee in 10 minutes :)
[08:43] <mhr3> sil2100, k
[08:51] <mhr3> sil2100, heading there
[08:53] <robru> sil2100, hey, 100% (all 7) tests pass for shorts app, you are go for launch
[08:54] <apw> can anyone point me at the python3 library to talk to the messaging indicator
[08:58] <larsu> apw: it's in the indicator-messages source package (libmessaging-menu)
[08:59] <larsu> libmessaging-menu-dev installs docs for devhelp
[08:59] <larsu> sadly, someone took down the python docs from the website, so those are only in C
[09:01] <apw> larsu, hey thanks, will look at that
[09:07] <tsdgeos> pstolowski: we have something that "works" for departments now (i.e. mhr3's plugin + my qml code + mhr3's test scope work)
[09:07] <tsdgeos> pstolowski: is there any "non test" scope that we could use?
[09:13] <pstolowski> tsdgeos, yes and no.. i've a branch for click scope, but it needs you to run fake click server (i've a python script for that as well).
[09:13] <pstolowski> tsdgeos, I can try to test all that if you point me to your branches
[09:15] <davidcalle> pstolowski, on that topic, is there a decision on how/when remote scopes should declare their departments?
[09:15] <davidcalle> (also, hi :) )
[09:16] <tsdgeos> pstolowski: so it's https://code.launchpad.net/~mhr3/unity-scopes-shell/department-support and https://code.launchpad.net/~aacid/unity8/departments/+merge/221072
[09:16] <pstolowski> davidcalle, hi! not yet, we'll hopefully get to that very soon (I had a very brief discussion with facundobatista about that)
[09:16] <tsdgeos> pstolowski: and you need a newer libunity-scopes-api or something, but i guess you have that
[09:16] <tsdgeos> pstolowski: you need to be on utopic, are you?
[09:16] <davidcalle> pstolowski, ok thanks
[09:17] <pstolowski> tsdgeos, yup, no worries
[09:17] <dandrader> mzanetti, qmake PREFIX=/usr CONFIG+=debug
[09:24] <tsdgeos> pstolowski: ping me if you come across something that you think doesn't work
[09:25] <pstolowski> tsdgeos, sure
[09:25] <pstolowski> compiling..
[09:26] <sil2100> mhr3: so, strange thing, as from what you mentioned I was expecting a new version of hud in trusty
[09:26] <mhr3> Saviq, can we talk about the caching once you have a moment?
[09:26] <Saviq> mhr3, sure, should be free before lunch
[09:26] <sil2100> mhr3: but actually I don't see any version like that, there's just that the version that was in UNAPPROVED got removed somehow
[09:27] <sil2100> mhr3: did I misunderstand something?
[09:27] <mhr3> sil2100, hmmm, maybe i did when talking to pete about it, let me look at it
[09:27] <tsdgeos> mzanetti: what's wrong in the tests?
[09:27] <mzanetti> tsdgeos: which tests?
[09:28] <tsdgeos> mzanetti: qmluitests
[09:28] <sil2100> mhr3: anyway, we'll have to just republish it somehow, since I don't see any direct upload at least
[09:28]  * mzanetti checks
[09:28] <sil2100> So it might have been removed and need a new push
[09:28] <dandrader> greyback, solved the blocked qt app issue!
[09:28] <tsdgeos> mzanetti: well you said it was very broken :D
[09:28] <tsdgeos> mzanetti: but i see now that lightdm didn't start :S
[09:28] <mzanetti> oh that one. not sure if it was a temporary issue or not
[09:28] <tsdgeos> seems it is
[09:29] <greyback> dandrader: \o/
[09:29] <tsdgeos> mzanetti: https://jenkins.qa.ubuntu.com/job/unity-phablet-qmluitests-utopic/ looks "better"
[09:29] <greyback> dandrader: what was it?
[09:32] <dandrader> greyback, well... maybe not. it fixed my simple qml app but system-settings is still getting kinda stuck.. but the situation improved a bit anyway
[09:32] <mhr3> sil2100, hm, so there's some changes in the changelog in the branch
[09:32] <mhr3> sil2100, i think it'll be best to rebuild it
[09:33] <pstolowski> tsdgeos, will the departments be also displayed in scope-tool?
[09:33] <tsdgeos> pstolowski: yes
[09:33] <sil2100> mhr3: in trusty trunk you mean?
[09:33] <pstolowski> cool
[09:33] <mhr3> sil2100, no, in the mp
[09:33] <mhr3> sil2100, pete pushed them an hour ago trying to sync everything
[09:34] <mhr3> sil2100, so, rebuild + publish again?
[09:37] <mhr3> sil2100, seems the misunderstanding was that the missing changelog entries were coming from proposed, not release
[09:37] <mhr3> although i thought we already had them there...
[09:37] <sil2100> mhr3: yeah, the stuff in -proposed was the previously verification-failed SRU...
[09:37] <mhr3> merge weirdness
[09:37] <sil2100> mhr3: you can rebuild and we can republish then
[09:38] <mhr3> sil2100, k, pushing teh button
[09:42] <tsdgeos> Cimi: https://jenkins.qa.ubuntu.com/job/unity-phablet-qmluitests-utopic/95/testReport/junit/%28root%29/qmltestrunner/Shell__initTestCase/ in your lightdm branch
[09:42] <tsdgeos> err
[09:42] <tsdgeos> infographics branch i mean
[09:42] <tsdgeos> Cimi: looks like that is a regression
[09:43] <Cimi> tsdgeos, is this running with the required branch?
[09:43] <tsdgeos> Cimi: no
[09:44] <Cimi> let's run it locally then
[09:45] <tsdgeos> pete-woods: you have a few <<<<<<< in https://code.launchpad.net/~unity-team/libusermetrics/file-based-infographics/+merge/214020
[09:46] <tsdgeos> pete-woods: can you fix them?
[09:46] <Saviq> Cimi, it doesn't look like has anything to do with infographics
[09:46] <Cimi> Saviq, Unable to assign [undefined] to QUrl
[09:47] <pete-woods> tsdgeos: sure
[09:48] <pstolowski> tsdgeos, yay, mostly works :), there is an issue with leaf that i'm currently looking into, could be something wrong on my side
[09:48] <Cimi> Saviq, I haven't touched that code
[09:48] <pete-woods> tsdgeos: done
[09:49] <tsdgeos> pstolowski: nice, tell me if you need something
[09:49] <tsdgeos> pete-woods: cool tx
[12:18] <mzanetti> dandrader: o/
[12:22] <mterry> Saviq, mzanetti: so...  word is that the kernel fix that split silo is waiting on is winding its way through Ubuntu process.  Which will be done tomorrow apparently.   Do we want to consider landing the silo in the meantime, with the expectation that both will land about the same time?  There's the risk that the cgmanager update will still be delayed by tomorrow for random reasons...
[12:23] <Saviq> mterry, we could put it through the testing and mark it tested: true
[12:23] <Saviq> mterry, with a note to not publish it before the kernel thing lands
[12:23] <mzanetti> +1
[12:23] <Saviq> wdyt?
[12:23] <mterry> sure?
[12:24] <Saviq> mterry, I really don't want to break smoketesting :|
[12:24] <Saviq> mterry, although we did determine that it shouldn't?
[12:24] <Saviq> mterry, so the fallback is that it won't boot every 20 times or so?
[12:24] <mterry> Saviq, theory is it shouldn't
[12:24] <Saviq> fallout
[12:25] <Saviq> mterry, let's just do what I said, we won't land anything unity8 in between
[12:25] <mterry> Saviq, did not test by running unlock-and-reboot script 40 times.  But maybe I should today to test that
[12:25] <Saviq> mterry, it also needs to go through QA signoff, so it will take until tomorrow anyway
[12:26] <mterry> Saviq, ok, well let's start that train then
[12:26] <Saviq> mterry, yup
[12:26] <Saviq> mterry, should we rebuild the silo yet again or are we good with what's there?
[12:26] <mterry> Saviq, it's good
[12:26] <mterry> Saviq, I merged and rebuilt unity8 today
[12:31] <Saviq> mterry, I need to sort out the silos, qtubuntu, ubuntu-touch-session and unity-mir are all in other silos as well
=== _salem is now known as salem_
[12:48] <tsdgeos> pstolowski: having problems seting up the thing as per your email, can you come over and help me?
[13:03] <tsdgeos> mhr3: have a sec?
[13:07] <mhr3> tsdgeos, sure
[13:07] <mhr3> tsdgeos, should i come over?
[13:08] <tsdgeos> sure
[13:42] <mhr3> tsdgeos, the changed signals we have in unity-api are pretty inconsistent
[13:42] <mhr3> tsdgeos, some of them do have param, others don't
[13:42] <mhr3> tsdgeos, shouldn't we consolidate it?
[13:42] <mhr3> tsdgeos, and what's the desired signature for those signals?
[13:54] <tsdgeos> mhr3: wops, coming over
[13:59] <mhr3> tsdgeos, lp:~unity-team/unity-api/departments
[14:02] <cwayne1> ooh departments
[14:04] <tsdgeos> bregma: with the help of Saviq we ended up with https://code.launchpad.net/~aacid/unity8-desktop-session/fix_logout/+merge/221238
[14:04] <tsdgeos> bregma: can you review?
[14:04] <tsdgeos> paulliu: ↑↑↑
[14:04] <bregma> OK, I'll add it to my queue
[14:05] <tsdgeos> :)
[14:46] <tsdgeos> mhr3: ping?
[14:51] <Cimi> Saviq, https://code.launchpad.net/~cimi/unity8/workaround_lp1324159/+merge/221248
[14:53] <Saviq> Cimi, tx
[14:54] <paulliu> jhodapp, hi. Do you have the script to resize system.img?
[14:56] <jhodapp> paulliu, not anymore, it was a pastebin that I don't have anymore
[14:56] <jhodapp> paulliu, I can help you a bit later, got a meeting to head to now
[14:56] <mhr3> tsdgeos, pong?
[14:56] <paulliu> jhodapp, ok. thanks a lot.
[14:57] <jhodapp> np
[14:57] <tsdgeos> mhr3: i''ve added docu and the tests to the mr
[14:59] <mhr3> tsdgeos, awesome, thx
[14:59] <mhr3> tsdgeos, so it even compiles now?
[14:59] <tsdgeos> mhr3: yeah!
[15:00] <mhr3> yey! :)
[15:04] <Saviq> mterry, the devices are flashed, but I'm afraid we need to wait for qtubuntu and unity-mir to land from silo 008
[15:05] <mterry> Saviq, I had MacSlow play with them, he got some crashes but I'm trying to see if they are related or not (one was in system-settings, so I doubt it, but the other was in unity8)
[15:05] <Saviq> mterry, kk
[15:11] <Saviq> mterry, so, one thing that we always drop the ball on... package cross-deps (Breaks, Depends etc. ), do we have them proper in silo 002?
[15:13] <mterry> Saviq, I believe so...  We certainly have the breaks/depends right between the unity8 packages.  As for depending on specific versions of other packages in the silo...  Not really.  But mostly because unity8-greeter *can* work alone without the other updates, but it just shouldn't
[15:13] <mterry> Saviq, ubuntu-touch-session will depend on unity8-greeter, to pull it into the image
[15:14] <Saviq> mterry, yeah, sure, if it doesn't *really* Break, we don't need it, just want to make sure that stuff doesn't break if you "apt install $foo"
[15:14] <Saviq> mterry, experience might not be great, but that's expected in that case
[15:14] <mterry> Saviq, is it common that an app crash tears down unity8?
[15:14] <mterry> Saviq, apt-get install foo should be fine
[15:14] <Saviq> mterry, the counter-argument is that we might get a migration from proposed and an image that builds with broken behavior
[15:15] <Saviq> in case things migrate separately
[15:15] <mterry> Saviq, you mean a partial migration situation?
[15:15] <Saviq> yes
[15:15] <Saviq> mterry, re: crash, it happens, but is not common
[15:15] <Saviq> mterry, and if you can get a trace, please do
[15:16] <mterry> Saviq, hmm...  I *think* we'd be mostly fine on silo 002.  Biggest problem would probably be USC landing separately
[15:16] <mterry> Saviq, can't get a good trace yet
[15:16] <mterry> And not reproducable
[15:16] <Saviq> mterry, so yeah, please think about where we should bump/add Provides/Breaks
[15:17] <Saviq> erm
[15:17] <Saviq> s/Provides/Depends
[15:17] <mterry> Saviq, no Breaks should be needed
[15:18] <mterry> Everything can land solo
[15:18] <mterry> Only question is if we might need some depends for expected behavior
[15:18] <mterry> Couldn't hurt to add one for USC
[16:07] <dednick> Saviq: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1324184
[16:07] <ubot5> Ubuntu bug 1324184 in Ubuntu UI Toolkit "UnityThemeIconProvider: Icons fall back to lower priority themes before checking alternative names" [Undecided,New]
=== jono is now known as Guest96631
=== salem_ is now known as _salem