/srv/irclogs.ubuntu.com/2013/03/08/#ubuntu-unity.txt

bschaefermterry, i've filed a FFe, so lets just leave it out of trunk for now00:03
hasselmmmhall119, Saviq: well. new plan. instead of ranting, try to get UnityNext running on wayland :-D00:03
bschaefermterry, thanks!00:03
hasselmmmhall119, Saviq: is there public code already?00:03
mhall119hasselmm: you mean on weston, or make Mir fluent in the Wayland protocol?00:06
mhall119hasselmm: public code for what?00:07
mhall119FWIW, I like this new plan better :)00:08
hasselmmmhall119: well. raof said weston and its protocols don't fit at all. so i expect things to end somewhere in the middle between wayland core and weston.00:10
mhall119so do you want mir code or Unity code?00:10
hasselmmmhall119, the unity branch that's supposed to run on top of mir00:11
mhall119I think https://launchpad.net/unity/phablet is the closest thing to that atm, as we re-org around one converged codebase00:11
hasselmmok. thank you.00:12
=== me4oslav_ is now known as me4oslav
mhall119np00:13
mhall119happy hacking00:14
hasselmmwell. for now good night. (stupid early daily scrums... kills all hacker romance)00:16
=== rsalveti- is now known as rsalveti
=== _salem is now known as salem_
=== rsalveti_ is now known as rsalveti
=== salem_ is now known as _salem
TheMusoDoes Qt provide any framework for sound events, i.e incoming phone call/message etc? If not, we should look at tying in libcanberra via QML/Qt bindings.04:32
TheMusoQuick check of stuff in Ubuntu says that KDE uses it, but not Qt.04:39
=== mmrazik|eod is now known as mmrazik
sil2100didrocks: hello! Are the jenkins autopilot builders still down?08:58
didrockssil2100: it seems so…08:58
sil2100didrocks: since I see the ps-indicators-autopilot-release-testing job waiting for autopilot-nvidia ;/08:58
didrockssil2100: veebers told me he would see with mmrazik08:58
didrocksmmrazik: any news? were you pinged on that?08:58
mmrazikdidrocks: I was. Wasn't it the discussion you had with gema on #qa?08:59
mmrazikdidrocks: srry. I had an impression they are aware and we are waiting for jcollado.08:59
mmraziklet me check08:59
mmrazikoh... I see. the problem is something else then wha tI thouhgt09:00
didrocksmmrazik: TBH, nothing is running atm on the machine09:00
didrocksmmrazik: can't we shutdown jenkins?09:00
didrocksand restart it09:00
mmrazikdidrocks: nothing running on magners you mean?09:01
didrocksmmrazik: yeah09:01
didrockswell, only the stuck job :p09:01
mmrazikdidrocks: I don't know TBH. I might be able to fix the autopilot machines and I don't really want to restart jenkins where I'm onlye "guest"09:01
mmrazikdidrocks: jibel is on vacation, right?09:01
didrocksmmrazik: right, he is09:01
didrocksmmrazik: do you mind coordinating with #qa?09:02
didrocksmmrazik: then, we can relaunch the daily, that's not an issue09:02
mmrazikdidrocks: I'm not sure a restart will help anyway. The problem is some of the nodes are offline09:02
mmrazikwe just need to bring them online again09:02
mmrazikrestarting just jenkins most likely is not going to help with these09:03
didrocksmmrazik: what's the node?09:04
didrocksmmrazik: the machine is working09:04
didrocksisn't the node attached to the jenkins instance?09:04
mmrazikdidrocks: magners. For some reason it fails to authenticate. It looks like its authenticating with jibels credentials.09:04
didrocksinteresting…09:05
mmrazikdidrocks: AFAICS the configuration of the slaves got corrupted09:10
mmrazikfixing it09:10
didrocksmmrazik: \o/ thanks09:11
sil2100\o/09:12
mmraziksil2100, didrocks, veebers: they should be all up09:14
didrocksok09:14
mmrazikgoing to write jibel an e-mail as I don't know what the original configuration was...09:14
sil2100Excellent! Thanks09:14
didrockslet me try restarting indicators09:14
didrocksthanks mmrazik :)09:14
mmrazikdidrocks: they are building09:14
didrocksmmrazik: oh great :)09:14
mmrazikdidrocks: they were queued09:14
mmrazikjust the machines were not online so the jobs were waiting09:14
didrocksmmrazik: ok, perfect, thanks!09:15
* sil2100 wants to use the indicators job to nasty experiments later09:15
didrockssil2100: so I confirm the issue btw09:17
didrockssil2100: fresh session, first super + A doesn't work09:17
didrockssil2100: what we can do is workaround for now, like press super, then super + A in this test09:18
didrockssil2100: and having a separate test for Super + A :)09:18
sil2100didrocks: yes, well, I discussed that with andyrock yesterday, but we're still not sure if it's the same issue - and what's the cause of it09:19
didrockssil2100: I think we need a separate test for super + A anyway, but ok, good hunting! :-)09:19
sil2100didrocks: since ok, indeed the first super+a does not work as I noticed yesterday, but the dash is being opened many times before the test already09:19
didrockssil2100: ah ok09:20
sil2100didrocks: I think indeed we'll have to workaround this breakage anyway for now ;p09:20
didrocksso not that one maybe in this case :)09:20
didrockssil2100: yeah :/09:20
sil2100hmmmm09:49
sil2100No idea what's happening, now I can't even reproduce that 'first super+a on a clean session does not work' bug - even though I have not upgraded anything!09:52
sil2100Still, for future reference:09:52
sil2100https://bugs.launchpad.net/unity/+bug/115251709:52
ubot5Launchpad bug 1152517 in Unity "Lenses shortcuts not working in singular cases (e.g. Super+A)" [Undecided,New]09:52
sil2100And I'll anyway workaround the problem, since it's still failing on jenkins09:52
didrockssil2100: you have new test results10:32
MCR1So you are not interested in Compiz fixes anymore and want contributors to go away so you have full control ? Or what is the reason for making those stupid Compiz changes ? Do I have to beg you twice now per fix to accept it, so your users won't experience bugs anymore ?11:06
MCR1I am pretty pissed as contributor... and I will not accept the death of Compiz because of Canonical policies...11:07
MCR1ignorance is bliss ?11:09
sil2100didrocks: testing the workaround in a jenkins environment right now ;)11:11
didrockssil2100: hum, using the jobs?11:11
didrockssil2100: or something else?11:11
didrockssil2100: I would want to finish waiting on unity to publish first if possible :)11:11
sil2100didrocks: veebers has a fun job called autopilot-run-custom-branch on another jenkins which can be used for building given tests from given branches ;)11:12
didrocksok :)11:13
sonnegreetings!11:36
sonnereading the shopping lens source code for ringtail, i see that the https handling is delegated to the vala libraries... i'm wondering how strict is their certificate authenticity check, anyone knows about it?11:38
sil2100veebers: ping!11:41
sil2100sonne: hello, maybe pstolowski would know more?11:42
sonnesil2100, i've been suggested to ask him actually... :)11:42
pstolowskisonne, sil2100: no idea, sorry, you may want to check how the handle ssl_strict flag there... mhr3?11:43
mhr3sonne, all i know is that our security team is happy with it11:44
mhr3mdeslaur, ^11:44
sil2100;)11:44
mhr3sonne, also, it's no "vala libraries" it's standard libsoup11:45
sonnemhr3, you may have noticed that my vala knowledge is nowhere beyond "vala is a language"... forgive my imprecision there :)11:46
mhr3sonne, just so you don't think that we have our own very special https implementation11:47
sonnemhr3, that's a good point11:47
sonnei thought there was some kind of stdlib, like there is in other languages...11:47
sonnemhr3, pstolowski, in case you're interested: i checked libsoup and it creates strict https connections by default :)11:57
didrocksTrevinho: hey, I'm seeing an interesting bug12:11
didrocksTrevinho: maximize a window12:11
didrocksyou have a one pixel "free" on the the edge12:11
didrocks(like if it's chromium, you can't scroll for instance)12:11
didrocksif you have an application behind12:11
didrocksthis one will become focused12:11
=== MacSlow is now known as MacSlow|lunch
=== _salem is now known as salem_
mdeslaursonne, mhr3: the lens does set ssl_use_system_ca_file and ssl_strict. If you suspect there's an issue, please file a security bug, and we'll take a look.12:38
om26erwhere can i get libbamf3-0 0.4.0 ??12:53
om26erdidrocks, you might know ? ^^12:58
didrocksom26er: are you using raring?12:59
didrocksraring has  0.4.0daily13.03.07-0ubuntu112:59
om26erdidrocks, its quantal, i am installing qml-phone-shell and it seem to require that version12:59
didrocksom26er: not sure about the qml-phone-shell on quantal though12:59
didrocksbecause to ask on #ubuntu-phone12:59
om26erdidrocks, i'll just download the raring daily and try there then13:00
om26erwhy is autolanding on compiz not working ?13:24
sil2100om26er: for which branch?13:25
om26ersil2100, https://code.launchpad.net/~mc-return/compiz/compiz0.9.9.merge-plugin-freewins/+merge/15096013:25
om26eror this as well https://code.launchpad.net/~brandontschaefer/compiz/lp.1075207-fix-ubuntu-super-p-patch/+merge/15107913:25
sil2100Maybe the merger got moved to lp:compiz/raring and it's no longer enabled for lp:compiz? hmmm, fginther ^ ?13:26
fginthersil2100, yes, the auto-merge was moved to lp:compiz/raring13:28
om26erah13:28
sil2100fginther: morning!13:30
sil2100:)13:30
fginthersil2100, good morning13:30
luvhi again13:31
luvI asked where to get in touch with ubuntu online accounts devs13:31
luvand was pointed here (?)13:32
didrocksmardy: ^13:32
mardyluv: hi! I'm one of them :-)13:32
luvgreat13:33
luvI would like to add a feature to log out (that is not just disable/enable an account) but discard the token and all so a new login is required13:33
luvkinda regression not having this feature in but patching apps to use UOA instead of they auth mechanism (because the apps usually support logouts)13:33
luvobviously not for raring, though sounds like something you would be interested in merging?13:34
mardyluv: yes, definitely13:35
mardyluv: discarding the token is something definitely possible (at the framework level), but forcing the logout is more complicated13:35
mardyluv: do you intendo to do so that when the logout button is pressed, all applications using the account immediately stop using it?13:36
luvwell, that sounds ideal, though i wouldn't really mind personally closing an app explicitly after logging out (I use only shotwell afterall)13:37
luvbut apps stopping using it imemediately sounds like the behaviour the user would expect13:38
bregmafginther, does that (lp:compiz automerge move) mean someone with appropriate privs will need to manually merge approved MPs to lp:compiz now?13:46
fgintherbregma, we could setup a new automerge job for lp:compiz13:51
fgintherbregma, but without that someone would have to do it manually13:51
fgintherbregma, I can set that up without any trouble, just let me know13:54
bregmafginther, having an automerge job would be nice for trunk compiz, but since it's not the branch being used for Ubuntu going forward, would it be using resources best focused elsewhere?13:56
sil2100I wonder how much resources does such an automerger eat up13:57
bregmaI don't know, that's why I'm asking here13:57
bregmasurely it takes a builder slot, and the tests get run, so it's non-trivial13:58
fgintherbregma, sil2100, If we only configure it to build amd64 and i386, the resource usage is not a problem. We're only limited on armhf right now13:58
luvmardy: can you just throw at me package names I should look at ? I guess I will have a look at the code and ping you sometime next week13:59
bregmaI dunno, I still see 6 hour waits when I upload to my PPAs...13:59
=== MacSlow|lunch is now known as MacSlow
fgintherbregma, would lp:compiz need to be dput into a ppa?14:00
mardyluv: https://launchpad.net/gnome-control-center-signon (libaccount-plugin/oauth-plugin.c) <- for the UI14:00
mardyluv: signon-plugin-oauth2 is the package which implements the OAuth2 method. IIRC it already supports clearing the token14:01
fgintherbregma, and would it need an armhf build?14:02
luvso for the starters it _might_ be enough to add ui for that?14:02
bregmafginther, yes it needs a PPA since it's not landing in Ubuntu (https://launchpad.net/~compiz-team/+archive/compiz works), but lets punt on the armhf build until someone demands it14:02
didrockswho would use this ppa?14:03
didrocksbregma: you guys should rather focus on what we are delivering I guess14:03
bregmadidrocks, it's for trunk compiz developers14:03
didrocksbregma: yeah, but who are compiz developers?14:03
didrockssam is always using the source, not the package14:03
bregmaI don't want my guys spending time doing manual merges to trunk compiz, so an automatic merger benefits me14:04
didrocksbregma: who would they build lp:compiz?14:04
didrocksbregma: we told we are using lp:compiz/raring14:04
didrocksit doesn't make sense for our resource to test/run something else than what we are shipping/focussing on14:05
bregmaI don't think setting up an autolander for a community project that we're taking from is a very big price to pay, certainly a smaller price than contibuting to the reputation for taking and giving nothing back14:08
didrocksbregma: I'm talking about ppa build14:09
didrocksbregma: we already have too many commit ppa builds14:09
didrocksand that's taking a huge amount of our resources14:09
didrockswe committed to kill the staging ppa, I think fginther is on it for the unity one as we have dailys14:09
om26ermzanetti, can you give the output of apt-cache policy libbamf3-0 ?14:12
bregmaOK, forget the PPA, if someone wants to distribute trunk binaries they can do that manually14:12
mzanettiom26er: Installed: (none)14:13
om26ermzanetti, do you have unity staging ppa ?14:15
mzanettiom26er: I don't think so... however I upgraded to raring and I still have some old quantal packages installed that make it compile...14:16
mzanettiom26er: could be that I had that one enabled some time ago14:16
om26ermzanetti, so what's wrong on raring ?14:16
mzanettiom26er: stuff not released for raring yet14:16
om26ermzanetti, if i install all the dependencies won't it work14:16
om26eraha14:16
om26ermzanetti, namely the HUD stuff ?14:17
mzanettiom26er: not sure what exactly...14:17
om26ermzanetti, it tells me if i upgrade libunity-core-6-0-dev it will remove unity because: Depends: libbamf3-0 (>= 0.4.0) but 0.3.4-0ubuntu1 is to be installed14:19
mzanettiom26er: anyways, jenkins builds on quantal with ppa:phablet-team/ppa ppa:canonical-qt5-edgers/qt5-proper and ppa:ubuntu-sdk-team/ppa14:19
mzanettiom26er: so repository-wise that should be all you need14:19
mzanettido a apt-get build-dep qml-phone-shell once you have those (and I guess only those in regard to phablet stuff) ppas enabled14:19
om26ermzanetti, i didn't have sdk team ppa enabled,14:21
mzanettiom26er: is it working now? or does the remove-unity problem still persist?14:22
om26ermzanetti, apt-get update is running still14:23
om26ermzanetti, i installed bamf from raring, lets see how well that goes14:26
mzanettiom26er: I don't think Unity Next uses bamf any more...14:26
om26ernow it doesn't remove unity but not really sure if unity will work afterwards14:26
om26ermzanetti, but it seems qml-phone-shell from phablet team ppa installs a new version of unity (nux one) so14:27
om26ermzanetti, so the shell runs on the desktop, now trying to compile14:27
om26ermzanetti, it compiles, but the shell crashes as soon as its window shows up14:29
om26erthough the installed version (from the ppa) doesn't crash14:29
om26ermzanetti, now tests are running, so where should i start :)14:34
mzanettiom26er: awesome!14:34
mzanettiom26er: I'd say indicator stuff should be final enough to test it14:35
mzanettiom26er: and since greyback left the team they are orphaned too14:35
om26ermzanetti, indicators are basically not working on the desktop.. i.e. sound and battery indicators are empty, battery indicator is not working on the desktop, and as of now the date/time indicator have all the hard-coded valuues14:39
om26erbut i could still tests very simple things like switching between indicators14:39
mzanettiom26er: you need to start some services...14:39
mzanettiom26er: yeah, start with that14:39
mzanettiom26er: and I will figure how to start the backend daemons for the tests14:40
om26ermzanetti, ok, great14:40
=== dandrader is now known as dandrader|afk
mzanettisomeone: https://code.launchpad.net/~mzanetti/unity/phablet-autopilot-deps/+merge/15241914:51
mzanettiCimi perhaps ^14:51
Cimimzanetti, also the ones I was missing?14:52
Cimimzanetti, qtdeclarative5-dev-tools14:52
mzanettiCimi: huh?14:52
Cimifor unittests14:52
mzanettiCimi: ah... thats for the build-deps14:52
mzanettiisn't it there?14:52
Cimisorry can you check?14:53
mzanettiyep, its there14:53
mzanettiCimi: ^14:53
Cimimzanetti, ok14:54
=== dandrader|afk is now known as dandrader
om26erwhats the default login password of the qml shell ?15:06
Cimiom26er, login as guest, no?15:16
Cimiom26er, I have the paswords though, somewhere :)15:16
om26erCimi, i dont see guest option, i am running on the desktop15:18
om26erCimi, it only shows the password box15:19
Cimimzanetti, ^15:19
cyphermoxsil2100: what's your take on the OIF tests failures this morning ?15:19
mzanettihuh... that's weird15:19
mzanettilike... really weird.15:19
mzanettiguest should always be there15:20
om26ermzanetti, due to that reason already existing tests are failing here as well... i might be something broken on my system though15:21
sil2100cyphermox: looking at oif now15:21
cyphermoxthanks15:21
mzanettiom26er: there seems to be a bug indeed15:22
mzanettiom26er: let me fix it15:22
om26erack15:22
mzanettiom26er: https://code.launchpad.net/~mzanetti/unity/phablet-fix-greeter-guest/+merge/15243615:42
om26ermzanetti, tried, works for me. approved15:48
sil2100cyphermox: seems like a singular failure, some timing issue probably - it won't happen with the next test run probably15:48
cyphermoxsil2100: alright, thanks for checking15:49
cyphermoxdidrocks: it's as I suspected ^15:49
didrockshum15:49
cyphermoxI'm going to check what oif was trying to release, and push the buttons for it and unity?15:49
didrockscyphermox: sounds good to me :)15:50
cyphermoxk15:50
didrockscyphermox: unity has packaging change btw15:50
didrocksas well ;)15:50
cyphermoxyeah15:50
sil2100cyphermox: I might try sewing up that failing test to boost the chances of not failing a bit, but as with timing issues, it's always a roullete15:50
cyphermoxhmmm vpn is really slow today15:50
cyphermoxyup15:51
* cyphermox publishes OIF15:54
* cyphermox publishes Unity15:55
cyphermoxdidrocks: things should depwait, if needed?15:55
didrockscyphermox: sorry, what do you mean?15:56
cyphermoxerr, unity build-depends on oif bits, right15:56
didrocksyep ;)15:56
cyphermoxyeah15:56
didrocksthat's why a stack is depending on the other15:56
cyphermoxso if I publish both in rapid succession, if anything, there will be depwait?15:57
cyphermoxhrmm, I got a launchpad timeout15:57
didrockscyphermox: yeah, they will be blocked in proposed and won't migrate if there is a real packaging dependency15:57
didrocksoh?15:57
cyphermoxand again :'(15:58
cyphermoxI'll wait a bit15:59
cyphermoxoif isn't published yet15:59
didrockscyphermox: ask qa?15:59
didrockscyphermox: they had issues yesterday, I spent part of my morning on getting things back on shape16:00
didrocks(on the DC)16:00
cyphermox*sigh*16:00
fgintherbregma, FYI, I'm setting up the auto merger job, just need to test it before turning it on16:01
sil2100fginther: oh, so there will be automerging for lp:compiz?16:03
fginthersil2100, yes16:03
fginthersil2100, just nothing automatically dput to a ppa at this time16:03
sil2100Just automerging is enough I think indeed16:08
bregmafginther, thanks16:08
sil2100Thanks \o/16:09
didrockscyphermox: any news on the timeout?16:17
cyphermoxin progress16:18
sil2100didrocks: I'll approve my workaround branch - it seems to work, but there is no 100% sureness, if it does not, we can always revert it16:19
sil2100It doesn't break anything at least16:19
didrockssil2100: ok16:19
sil2100It's just a test mod16:19
didrockssounds good to me :)16:19
sil2100andyrock already reviewed it ;)16:19
andyrocksil2100, what branch?16:21
andyrockworkaround branch?16:21
=== dandrader is now known as dandrader|lunch
sil2100andyrock: yes16:42
=== dandrader|lunch is now known as dandrader
Trevinhodidrocks: about the new nautilus / sw updater icons... are we also going to set the .desktop file parameter to define the proper background?18:07
didrocksTrevinho: I think it's a question for seb128 ^18:08
seb128Trevinho, didrocks: yeah, question for me18:08
seb128Trevinho, those infos don't belong in the .desktop18:08
Trevinhodidrocks: ah, sorry... I remember you were talking about it with JohnLea18:08
seb128Trevinho, the icon is not the same in e.g the Ubuntu and GNOME themes18:08
seb128so the color should be different depending of the theme18:08
seb128-> the info should be stored in the icon theme18:09
Trevinhoseb128: ah...18:09
seb128you should invent a key file in /usr/share/icons/<theme>18:09
seb128with icon=color18:09
seb128imhp18:09
seb128imho18:09
=== alan_g is now known as alan_g|afk
Trevinhoseb128: that's true.. however at this point we can stil check the theme in unity and use the .desktop parameter if set (considering that it's what we alredy have)18:09
seb128Trevinho, we added it to software-center and nautilus, will add it to update-manager18:10
Trevinhoseb128: it's not a good solution, but it's something that's doable in not that much time18:10
seb128but that's just wrong18:10
Trevinhoseb128: yeah...18:10
seb128Trevinho, it means if we change the icon (as we did for nautilus this week); we need to coordinate .desktop updates in the app as well18:10
Trevinhoseb128: it should have been designed better before18:10
seb128Trevinho, I'm also not going to patch thousand of .desktop for that :p18:10
seb128it seems random to do it for 3 launcher icons18:11
Trevinhoseb128: no, of course... but it's just about three or four for now18:11
seb128Trevinho, well, nautilus (as from today) and software-center has the key, update-manager will also get it in the next upload18:11
seb128Trevinho, I'm fine doing the hack for now18:11
seb128but it's not a proper solution and doesn't scale18:11
seb128so we should think at doing it in a better way ;-)18:12
Trevinhoseb128: totally18:12
Trevinhoseb128: ah, and for that nautilus patches? SHould I fill a FFe for them?18:12
seb128Trevinho, yes please18:13
seb128sorry about that18:13
seb128those release plan changes created issues for everyone18:13
Trevinhono problem... I probably had not the time to backport them before...18:13
Trevinhoseb128: and should I work on 3.6, right?18:13
seb128yes18:13
Trevinhoseb128: i,e the one already in raring..18:13
Trevinhook fine18:13
seb128I doubt we will update to 3.818:14
seb128we don't have gtk 3.7 in raring yet18:14
seb128and 3.8 has UI changes and some new features, and break some our patches18:14
seb128and we are feature frozen18:14
seb128so better to base on 3.618:14
Trevinhoseb128: yeah... but for what it worth, trunk works well even without it (after disabling few lines of code)18:14
=== dandrader is now known as dandrader|afk
=== dandrader|afk is now known as dandrader
=== salem_ is now known as _salem

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