[08:14] <dholbach> good morning
[09:17] <nerochiaro> oSoMoN: hey, good morning. safely back from Florida ?
[09:17] <oSoMoN> nerochiaro, morning!
[09:17] <oSoMoN> yup, back home
[09:30] <nerochiaro> oSoMoN: can you have a look at https://code.launchpad.net/~amanzi-team/gallery-app/gallery-app-swipe-desktop/+merge/204741 when you have time ? CI fails but not because of my stuff I think, and I triggered a re-run since I saw on the ML there were fixes in the infrastructure
[09:30] <oSoMoN> nerochiaro, yes, will take a look
[09:34] <JamesTait> Good morning all; happy Clean Out Your Computer Day! :-D
[11:55] <nerochiaro> oSoMoN: are you running ubuntu on dual boot ?
[11:56] <oSoMoN> nerochiaro, no, not tried yet
[11:56] <nerochiaro> ah ok, was going to ask if it's normal that it takes ages to update the ubuntu image
[12:38] <dpm> nerochiaro, it does
[12:38] <nerochiaro> dpm: is anyone on the case ? or is it considered normal ?
[12:38] <dpm> nerochiaro, it takes about 30min-1h for me to download it
[12:39] <dpm> I don't consider it normal, but I'm not sure what's causing it
[12:39] <dpm> I talked to ondra about it a while ago, and he seems to think it's because of the load on the download servers
[12:40] <dpm> but I'm surprised to see it's got _that_ much load
[12:40] <dpm> especially as most people with an Ubuntu phone would not download the full image, but the image updates instead
[12:43] <nerochiaro> dpm: for me it takes ages to install the image, not downloading it
[12:45] <dpm> oh, installation was relatively quick for me, about a couple of minutes
[12:48] <nerochiaro> dpm: on what device ?
[12:49] <dpm> Nexus 4
[12:57] <nerochiaro> dpm: galaxy nexus here
[12:57] <nerochiaro> ondra: have you tried running the ubuntu dual boot installation (or update) on a galaxy nexus ? It takes hours to update on my device
[13:06] <ondra> nerochiaro: what is taking long, download or actual installation?
[13:07] <nerochiaro> ondra: actual installation, download is slow but it's still ok. installation is really slow
[13:07] <ondra> nerochiaro: yep, it's due to slow hw, if you use phablet-flash it will be same
[13:08] <nerochiaro> ondra: actually no, it is significantly faster
[13:08] <ondra> nerochiaro: you might enable in developer options to stay alive when plugged, to make sure powermanager does not halt install when it runs on background for too long
[13:08] <nerochiaro> ondra: i never had to wait more than 15-20 minutes to finish an install with phablet-flash, it took more than an hour with the updater from android
[13:09] <nerochiaro> ondra: oh, that thing with it stopping in background might be the problem. i'll check that
[13:09] <ondra> nerochiaro: but actual unpacking of tar.xz is slow and then creating swap is dead slow, as ogra_ pointed out
[13:10] <nerochiaro> ondra: but it shouldn't be slower than phablet-flash, should it ?
[13:10] <ondra> nerochiaro: dunno, ogra_ said it took him ~45 mins to update GNexus
[13:10] <ogra_> the unxz on the maguro is extremely slow ...
[13:10] <ondra> nerochiaro: install not really, it does same thing as phablet-flash, just download is done on the phone, rather than pc
[13:10] <ogra_> (thats the nature of this HW)
[13:11] <ondra> nerochiaro: once files are in the phone, it's identical to phablet-flash
[13:14] <nerochiaro> ondra: i'll try to see if keeping it alive will make things better, it could be it was being stopped when in background for too long. i'll report again if it's still slow
[13:20] <ondra> nerochiaro: thanks!
[13:20] <nerochiaro> ondra: thanks to you
[14:31] <elopio> ping renato. I have some questions about EDS.
[14:39] <renato> elopio, hi
[14:39] <elopio> renato: hello!
[14:39] <elopio> renato: we need to start autopilot tests with a fresh database. Is there a way we can do that?
[14:40] <renato> elopio, you can check the unit tests for eds plugin
[14:40] <elopio> renato: here? https://launchpad.net/qtorganizer5-eds
[14:40] <renato> yes
[14:40] <renato> there is some code on makefile to do that
[14:41] <elopio> renato: looking at it.
[14:41] <elopio> nik90: ^
[14:42] <nik90> hmm I am terrible with C++ and unit tests
[14:44] <dpm_> hi mzanetti. We want to upload a new reminders app version to the store. If you've got a minute, could you have a look at https://code.launchpad.net/~dpm/reminders-app/bump-to-0.4/+merge/205602 ? The MP is quite trivial, but it'd be nice to sort out the issue with the bzr revision explained in the description. Do you have any ideas?
[14:44] <renato> elopio, nik90 the big trick is on this file: http://bazaar.launchpad.net/~phablet-team/qtorganizer5-eds/trunk/view/head:/tests/unittest/CMakeLists.txt
[14:46] <mzanetti> dpm_: can do in a bit. not right now
[14:46] <WebbyIT> dpm, sorry, I don't understand your comment here. mzanetti and I are speaking about a button to choose with wich account do the login in reminder app, why it be should part of Online Accounts? After a login with an account, user maybe wants to switch  account, and IMO is non-sense to go to online account to change actual user....
[14:46] <WebbyIT> https://code.launchpad.net/~rpadovani/reminders-app/multiple-accounts/+merge/205066/comments/480650
[14:47] <dpm_> np, thanks mzanetti
[14:47] <nik90> renato where is the database file actually stored in the system? Cant I just move that database to a backup location and create the alarms. Then delete the temp database and restore the original database back?
[14:48] <renato> nik90, I do not recommend that if the user is using the database this will cause a problem, if the test crash or the user kill the test the database will not be restored
[14:48] <renato> and a lot of other problems
[14:48] <renato> in my opinion the test should never touch on user data
[14:49] <nik90> renato: okay
[15:15] <WebbyIT> dpm_, sorry, I don't understand your comment here. mzanetti and I are speaking about a button to choose with wich account do the login in reminder app, why it be should part of Online Accounts? After a login with an account, user maybe wants to switch  account, and IMO is non-sense to go to online account to change actual user....
[15:15] <WebbyIT> https://code.launchpad.net/~rpadovani/reminders-app/multiple-accounts/+merge/205066/comments/480650
[15:15] <mzanetti> :D
[15:16] <WebbyIT> well, I dunno if dpm_ see it, cause I pinged wrong dpm :P
[16:10] <dpm_> WebbyIT, sorry. I did see it, but I haven't had the chance to reply
[16:10] <WebbyIT> dpm_, np, when you have time :)
[16:10] <dpm_> WebbyIT, I'll reply on the comment as soon as I can
[16:10] <WebbyIT> dpm_, thanks, as said np, when you'll reply I'll update the branch :-) Thanks!
[18:16] <m-b-o> balloons: ping
[18:22] <balloons> m-b-o, just thinking about you :-)
[18:22] <msvb-lab> Anybody know if the <MyProj>.qmlproject.user from new Ubuntu-SDK projects should be checked in to revision control?
[18:22] <msvb-lab> ...or if these .qmlproject.user files should NOT be checked in?
[18:22] <m-b-o> ballons: fine :)
[18:22] <msvb-lab> ...or if it's optional?
[18:23] <msvb-lab> Wondering about these files due to the 'user' part that seems to me could include a password or something?
[18:23] <msvb-lab> ...and that a new project by default adds many files to revision control, but not this one.
[18:23] <balloons> m-b-o, so what's up?
[18:23] <m-b-o> balloons: I've tried to test keyboard-shortcuts but had no luck with jenkins/autopilot-sandbox-run
[18:23] <msvb-lab> ...yet there isn't a default 'ignore' policy either.
[18:23] <balloons> m-b-o, what'
[18:23] <m-b-o> balloons:  on the desktop all test are fine
[18:24] <m-b-o> balloons: see https://code.launchpad.net/~martin-borho/ubuntu-weather-app/keyboard-shortcuts the tests with shortcuts are skipped at he moment
[18:25] <balloons> m-b-o, we should land the feature with the test
[18:25] <balloons> m-b-o, let me help figure it out with you
[18:26] <m-b-o> balloons: cool :) MP got approved minutes ago I saw
[18:27] <balloons> m-b-o, can we wait to land it? have time now to fix?
[18:27] <m-b-o> ballons: I have time! Is it bad, if it gets merged without proper tests?
[18:28] <balloons> so if the keyboard shortcuts work on the desktop, where else do they need to work but don't?
[18:28] <m-b-o> balloons: ah! :)
[18:28] <balloons> I don't think it's intended they work on the phone right?
[18:28] <m-b-o> yes
[18:28] <balloons> m-b-o, so we can make them desktop only test
[18:29] <m-b-o> yes, I think so
[18:32] <m-b-o> ballons: I guess I have to set somehow another scenario?
[18:33] <nik90> msvb-lab: you shouldn't add .qmlproject.user to the version control. It is generated in every user's computer for a project
[18:33] <m-b-o> balloons ↑
[18:34] <msvb-lab> nik90: Okay, makes sense. But wouldn't the tab and style settings for a project be useful to duplicate across developers?
[18:34] <msvb-lab> Are there eventually passwords, paths, or other identifiers that will be stored in the .qmlproject.user file without the developer's knowlege?
[18:36] <nik90> msvb-lab: yes and no..settings like tab, style setting etc are personal. Every developer has their own. They wouldn't want their own settings to be replaced with yours, wouldn't you agree?
[18:36] <balloons> m-b-o, we can add a check for what device, and just return if it's not desktop
[18:36] <nik90> msvb-lab: I don't think passwords or other identifiers are stored in it.
[18:36] <msvb-lab> nik90: But it seems that paths do eventually get in there.
[18:37] <msvb-lab> ...and that's bad enough leakage.
[18:37] <m-b-o> balloons: ok, let me try
[18:37] <balloons> m-b-o, like if platform.model() != "Desktop" then return
[18:37] <msvb-lab> nik90: I agree that these files shouldn't be put in revision control, now the question is why aren't there entries by default in the 'ignore' file of a new revision controlled project.
[18:38] <m-b-o> balloons: same thought :)
[18:38] <nik90> msvb-lab: all core apps just add that file to their bzrignore file
[18:41] <msvb-lab> nik90: Do you mean the developer of a new core app manually adds the file to bzrignore?
[18:41] <msvb-lab> It doesn't seem that the Ubuntu-SDK generates a bzrignore file at all for new projects.
[18:41] <msvb-lab> ...neither for mercurial '.hgignore'
[18:41] <msvb-lab> So I wonder why not, since that would be useful wouldn't it?
[18:42] <nik90> msvb-lab: I guess it would be
[18:42] <m-b-o> balloons: autopilot-sandbox-run is also "Desktop".... so this alone won't make it
[18:42] <nik90> msvb-lab: I will file a bug about it
[18:43] <msvb-lab> nik90: Okay, please tell me the bug number since I might work on it (or at least monitor its progress.)
[18:48] <nik90> msvb-lab: https://bugs.launchpad.net/qtcreator-plugin-ubuntu/+bug/1278538
[18:48] <ubot2> Launchpad bug 1278538 in qtcreator-plugin-ubuntu "Creating a new QML project should add sensible defaults to the .bzrignore file" [Undecided,New]
[18:48] <msvb-lab> Okay, thanks.
[18:55] <balloons> m-b-o, autopilot sandbox isn't an issue
[19:07] <balloons> m-b-o, so with that change it should run fine through jenkins; it runs fine on the desktop and will continue to run on phablet
[19:08] <balloons> m-b-o, we could also add a check for the proper behavoir on the phablet devices. What should happen if I try a keyboard shortcut? I guess I'm not sure how I could even invoke it
[19:08] <m-b-o> balloons: pushed!
[19:09] <m-b-o> balloons: hmm, you should connect a real keyboard then
[19:10] <m-b-o> ballons: but with a bluetooth-kb connected to a tablet...
[19:10] <m-b-o> balloons: but with a bluetooth-kb connected to a tablet...
[19:15] <m-b-o> balloons: do you have to top approve to trigger jenkins?
[19:18] <balloons> m-b-o, sure
[19:18] <balloons> m-b-o, no, don't need to top approve to trigger
[19:25] <m-b-o> balloons: nope... failed again
[19:26] <balloons> m-b-o, oO The buildbot is considered desktop
[19:26] <balloons> let me look closer
[19:26] <balloons> ok, so statenotfound errors is likely a timing thing
[19:29] <m-b-o> balloons: in that particular case the sheet wasn't opened
[19:49] <balloons> m-b-o, I'm pulling your branch
[20:00] <balloons> m-b-o, I would simply return, instead of "return None"
[20:11] <m-b-o> balloons: yes, but that wouldn't fix it, I guess
[20:19] <m-b-o> balloons: going off now, will ping you tomorrow!
[20:20] <balloons> m-b-o, cu
[20:20] <m-b-o> cu
[21:37] <daker> is there any precise ppa for qtdeclarative5-usermetrics0.1 ?
[22:36] <ybon> I've this message in OSMTouch logs: "Unable to write tile cache file "/home/phablet/.tilecache/queue1" Is there any config to for the app to be able to write on /home/phablet? Can this be some apparmor policy missing?
[22:41] <sarnold> ybon: check /var/log/syslog for DENIED messages to see if apparmor is related
[22:42] <ybon> thanks sarnold
[22:42] <ybon> checking :)
[22:43] <ybon> Many of this Feb 10 23:20:17 ubuntu-phablet kernel: [ 2129.702242] type=1400 audit(1392070817.580:649): apparmor="DENIED" operation="open" parent=1785 profile="me.yohanboniface.osmtouch_OSMTouch_0.1.2" name="/home/phablet/.config/me.yohanboniface.osmtouch.conf" pid=8259 comm="qmlscene" requested_mask="r" denied_mask="r" fsuid=32011 ouid=32011
[22:43] <ybon> but nothing about the previous path
[22:44] <popey> you should write to one of the directories you're allowed to via the apparmor profile
[22:44] <popey> ~/.tilecache doesn't look XDG compliant
[22:44] <ybon> ah, interesting
[22:44] <popey> https://lists.ubuntu.com/archives/apparmor/2013-August/004183.html
[22:44] <ybon> this is the QtLocations default behaviour, btw
[22:45] <ybon> but I will check how I can take control
[22:45] <ybon> (I've many thing which I want to override, but for this I need some time to do C++, for now I'm only dealing the qml side)
[22:46] <popey> jdstrand knows all about this and may have a handy link for reference
[22:47] <popey> http://askubuntu.com/questions/160813/where-should-i-save-temporary-files-of-the-app-i-am-developing maybe
[22:47] <ybon> thanks :)
[22:48]  * ybon reading both links
[23:26] <jdstrand> ybon (and popey): http://developer.ubuntu.com/publish/apps/security-policy-for-click-packages/
[23:27] <popey> aha! that's the one, thanks jdstrand
[23:27]  * popey embookmarkens
[23:27] <jdstrand> np
[23:28] <ybon> nice, thanks jdstrand :)
[23:35] <ybon> I'm not sure I'm able to fix this qml side, though
[23:40] <sarnold> ybon: hrm, I thought everything was just supposed to work. Perhaps it's in how you're launching the application? see if this is related / useful: https://wiki.ubuntu.com/SecurityTeam/Specifications/ApplicationConfinement#Launching_applications
[23:40] <ybon> ah, intesting, thanks
[23:41] <ybon> here is my desktop file, in case: http://bazaar.launchpad.net/~yohanboniface/osmtouch/trunk/view/head:/OSMTouch.desktop
[23:41]  * ybon reading
[23:47] <ybon> I do see a ~/.cache/me.yohanboniface.osmtouch/ folder in my phone
[23:47] <ybon> so I guess in some way the settings are good
[23:47] <ybon> BUT the QtLocation tilemanager seems to want to do its own way
[23:48] <ybon> ah, damn
[23:48] <ybon> seems to be fixed in QT 5.2.0
[23:48] <sarnold> ooo
[23:49] <ybon> just as many of the bugs I see in QtLocation
[23:50] <ybon> seems that I need to employ a bit of my energy to help jumping to QT 5.2.0
[23:50] <ybon> (instead of taking time to workaround)
[23:50] <sarnold> I'm surprised moving to 5.2 is as much effort as it sounds.
[23:51] <ybon> there was a list of bugs tagged 5.2.0 but I can't remember where
[23:51]  * ybon still a bit lost in all the repositories
[23:52] <ybon> https://bugs.launchpad.net/bugs/+bugs?field.tag=qt5.2
[23:53] <sarnold> wowwwweee
[23:56] <ybon> ah but some have "fix committed"
[23:57] <ybon> "only" 26 remaining