[01:28] hi [02:03] anybody working on the proposed dev of ubuntu touch? [03:29] Hello? [03:29] Hello World! [05:50] TS hey [07:05] o/ good morning all [07:28] mzanetti, hey, not bad, Tagger already translated into 15 languages ;) https://translations.launchpad.net/tagger [07:30] what does tagger do ? [07:31] because it isn't translated to dutch, I wouldn't mind helping with that [07:32] justCarakas: https://launchpad.net/tagger [07:32] app for QR code reader [07:34] nice [07:44] dpm: you around ? [07:52] dpm: nice :) [07:56] ahh that's nice , at what stage the app is ? [07:56] nik90: ^^ [07:56] no idea [07:57] I just read the project description [07:57] okay :) [07:57] mihir, I am, but I'm a bit busy for the next couple of hours. If you've got any question, please feel free to ask it directly instead of pinging first [07:58] mzanetti, hey, I'm reviewing your branch. Looks good, just some little type here and there, I'll leave inline comments. But one thing is broke the count() of tags: add a tag from website and assign to a note, refresh with pulldown, the new tag appear in the list, but the tag count reports 0, and I have to restart the app to have '1' [07:58] dpm: question was regarding splash-screen , are we unblocked , if we remove _X-Ubuntu-Splash-Title= ? [07:58] mihir, what do you mean by unblocked? [07:58] dpm: https://code.launchpad.net/~mihirsoni/ubuntu-calendar-app/calendar-splash/+merge/237319 [07:59] mihir, I still don't understand [07:59] dpm: the splash merge was blocked because of bug https://code.launchpad.net/bugs/1377896 , translation issues [07:59] mzanetti, tldr; the tags.refres() doesn't refresh the count property [07:59] well, NotesStore.refreshTags(); [08:00] mihir, that bug does not block the merge. Neither the current splash nor the splash in that MP are translated, but they will be once the fix for that bug lands [08:00] that is, there is nothing blocking the MP to land [08:00] dpm: okay popey ^^ [08:01] dpm: thank you :) [08:01] mihir, and afaik, we can remove _X-Ubuntu-Splash-Title= if it's the same as Name [08:01] but that's the theory, I guess we'll have to try :) [08:02] popey, music remix branches now on http://people.canonical.com/~dpm/core-apps-sponsoring/ [08:02] dpm: okay thank you . i'll do that. [08:13] rpadovani: ah, thanks for the testing! [08:13] rpadovani: yeah, note things down in the MP [08:30] mzanetti, rpadovani, I'll be there in 2 minutes [08:30] cool [08:41] zbenjamin, hi [08:42] davidcalle: hey [08:43] zbenjamin, I know this is very fresh, but when can we expect the qtcreator "reviewtasks" fixes to be released? [08:44] davidcalle: its in landing already, today or tomorrw [08:44] zbenjamin, awesome :) [08:48] Good morning all; happy Fire Prevention Day! :-D [09:12] hey oSoMoN, could you ping me when you have 5 minutes for a query, please? :-) [09:16] rpadovani, ping [09:44] Hi [09:55] balloons, fginther: Will the jenkins focus issue for clock app be resolved today? (pretty pls :) ) [09:56] rpadovani: hey, I remember you having university classes this afternoon, Can you help me with some code reviews when you are back home? [09:56] nik90, sure, after 18:00 I think [09:56] nik90, or 2night [09:57] rpadovani: cool, thnx..just give me a ping then [09:57] sure [10:16] dpm: hey, I approved the bouncing progress indicator [10:16] dpm: can you test this one? https://code.launchpad.net/~rpadovani/reminders-app/multipleAccounts/+merge/237682 [10:20] mzanetti, sure, I'll give it a go as soon as I've fixed UOA on my phone [10:49] mihir, ping [11:02] hey, what does this 'ota' & 'ota-1' mean that i'm seeing in bugs and bug-tags? I presume it's a post-rtm thing? [11:03] DanChapman, hmmm good question [11:03] popey: ^^ care to enlighten me with your wordly knowledge :-) [11:04] over the air update post-RTM [11:04] things we expect to ship really soon after release [11:05] popey: ahh ok then, makes more sense now. Thanks !! :-) [11:31] hi zbenjamin, it seems I can no longer run Reminders on the emulator due to some click review tools check failures - I think you had a fix for this about to land IIRC? Do you know of any workaround in the meantime? http://i.imgur.com/jOitaqG.png [11:33] dpm: yes if you don't need debug output [11:33] dpm: are you on trusty or utopic? [11:33] dpm, I presume you know how to install them manually? [11:33] dpm, I know how to get around that; nik90 showed me [11:33] dpm: if you are on utopic, the silo7 contains your solution [11:33] zbenjamin, I'm on trusty on this PC [11:34] dpm, http://paste.ubuntu.com/8506701/ [11:34] dpm: or that https://launchpad.net/~ubuntu-sdk-team/+archive/ubuntu/tools-development [11:34] * akiva-thinkpad is not sure if that is what you need. [11:34] dpm: i think bzoltan pushed the newest fix to it [11:35] akiva-thinkpad, I know how to manually install click packages, but that's not what I'm trying to do - I appreciate the help, though [11:35] dpm: yeah the development ppa should have the fix [11:35] Okay sorry :P Its probably to you like me telling you how to use sudo apt-get! of course you know!!! [11:35] :D [11:35] zbenjamin, hm, I thought I was using the development PPA already and I was up-to-date, let me re-check [11:36] dpm: the package is only 50 mins old [11:36] * nik90 is on trusty and checks as well [11:36] zbenjamin: it is in [11:36] zbenjamin, do you know the cause of the failures, though? Why does it not recognize i386 as a valid arch, and where does that debug policy (which is not in my .apparmor file) comes from? [11:37] aha, yes, apt-get update shows some new qtcreator packages :) [11:39] dpm: the debug policy is injected by QtC so it can reroute stdout and stderr [11:39] dpm: the i386 arch i don't know, the review tools consider it as invalid arch [11:39] ack [11:40] zbenjamin, ok, I've installed the new QtC, but it still does not launch the app. The only difference now is that 2 of the previous errors are demoted to warnings, but still it doesn't install and run the click [11:40] dpm: the problem with the debug policy is, it would be much nicer if i would inject it after the package has built. But its not so trivial and would make the app startup even slower [11:41] dpm: thats weird [11:41] dpm: how is the branch called again? [11:42] zbenjamin, bzr branch lp:~rpadovani/reminders-app/multipleAccounts [11:42] zbenjamin, dpm: Clock had the same error msg for the i386 issue, but after the update it works as expected. I got a dialog asking if I wanted to ignore the message and then it worked [11:42] dpm: did you see tht dialog? [11:43] nik90: but you get the dialog because of the policy you use, there should be still a error listed [11:43] nik90, no, no dialog after updating QtC, closing it, restarting it and rerunning the app [11:44] zbenjamin: the i386 issue is listed as a warning, not a error anymore [11:44] nik90: ok, i had to filter that in the plugin ;) [11:44] the only real error is the use of "calendar" which is available only for vetted applications [11:44] nik90: can you check the new devices page? does everything look alright to you? [11:44] (which clock is) [11:44] nik90, yeah, I saw that too, I get a warning now, but no dialog [11:45] dpm: and it still doesn't run on the emulator or phoen for you? [11:45] nope [11:45] dpm: building [11:45] zbenjamin: wow you changed the device details page quite a bit [11:45] nik90: yep :) [11:46] zbenjamin, also, the devices page has some black squares on my PC: http://i.imgur.com/71KKvb4.png [11:46] nik90: do you see black squares also ? [11:46] zbenjamin: no I see no icon at all, https://imgur.com/5q1dToe [11:47] zbenjamin: so at first I didn't know they were clickable [11:47] https://imgur.com/wzS5mUW [11:47] weird ... probably i should include the icon then [11:48] zbenjamin, oh, it lost the kit. I've got no kit assigned to my emulator. Let me see if that fixes it [11:49] Hey there! [11:50] I'm working for Spotify, trying to improve the Linux client and implement proper media key support [11:50] But I'm hitting a bug in dbus-cpp documented here: https://bugs.launchpad.net/dbus-cpp/+bug/1378807 [11:50] dpm: it starts for me but i had to choose the reminders2 run configuration [11:50] dpm: since it creates a executable the default cmake plugin also creates a (sadly) invalid runconfig [11:51] Has anybody any experience with that library? Knows the bug? Has any idea on how I could fix the issue? [11:51] seb128, perhaps you can point Orphis to the right person? ^^ [11:51] That would be awesome [11:52] Orphis, tvoss is your man :) [11:53] zbenjamin, yeah, I'm familiar with that bug. I seem to remember fixing that was blocked on not all project types using kits, but I'm not sure if I recall correctly. Are we in a better position now to fix that "extra run configs" bug? [11:53] dpm: no, fixing it would break all other cmake projects [11:54] dpm: or refactoring the QtCreator API [11:54] dpm: right now there is no way to choose tell the QtC API that the cmake runconfigs are not compatible with our kits [11:55] dpm: i probably can take another look at it later .. maybe i missed something before [11:55] dpm: nik90: do you get any errors on the console because of the icons? [11:56] zbenjamin: 1 sec, let me open qtc via console [11:58] nik90: just paste the log somewhere i'll sort it out [11:58] zbenjamin, thanks. I think it'd be good to talk about it at the sprint. It currently makes for a bad developer experience, as running apps is a bit of a hit and miss. It happens to me all of the time, but because I know I have to switch run configs, I just do it. But the first time it happened to me I was quite confused and thought the SDK was broken. I could imagine most app developers don't know about the issue and are equally confused [11:58] dpm: agreed [11:58] dpm, Orphis: what library? [11:58] dpm: but most app devs will not hit it because we have only qml based templates, it happens only when you build your own main.cpp [11:58] seb128: dbus-cpp; tvoss is on the issue already :) [11:59] zbenjamin: http://paste.ubuntu.com/8526422/ [11:59] dpm: but it sucks that is true [11:59] Orphis, right, that's the right guy for your questions I think ;-) [11:59] nik90: looks like https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1371509 [11:59] Ubuntu bug 1371509 in Ubuntu UI Toolkit "Icons black in gallery from krillin #45" [Critical,Fix released] [11:59] zbenjamin, well, I don't have stats as to how many devs write QML vs C++. I think QML still wins C++ at this time, but we might have a different picture when we start selling devices [12:00] dpm: agreed [12:00] seb128: Yup, he gave me some awesome support already :) [12:02] zbenjamin: hmm that fix is going to be released into trusty though :/ [12:02] hey check this out http://2buntu.com/touch/device-art-generator/ [12:02] nik90: yeah i most likely will have to work around by putting the icons into the qrc === _salem is now known as salem_ [12:18] nik90: ping, are you destroying any Alarm data that you get using AlarmModel.get() function? [12:19] zsombi: destroy? Not that I can recall.. I don't remember calling alarm.clear() except for in the qml tests [12:20] nik90: alarm.reset() only clears the alarm data, does not delete it [12:20] zsombi: to delete an alarm I do, var alarm = alarmModel.get(i); alarm.cancel() [12:20] nik90: that's fine [12:20] zsombi: ok, other than the 2 above methods, I don't do anything to destroy alarm data [12:21] tbh I am not how to do it either way [12:21] nik90: :) [12:21] nik90: alarm.destroy() dould do the job.... however you shouldn't do that EVER!!! I'll put that in teh doc ;) [12:21] zsombi: lol [12:22] zsombi: if it should never be used EVER, why are you exposing it in the docs, or were you sarcastic about it? [12:23] nik90: did I say should NEVER be used? [12:24] zsombi: isn't shouldn't do that EVER and NEVER the same thing in this context? :P [12:24] nik90: can you check if libqt5svg5 is installed for you? [12:24] zbenjamin: sure [12:24] 15:21 zsombi: nik90: alarm.destroy() dould do the job.... however you shouldn't do that EVER!!! I'll put that in teh doc ;) [12:24] nik90: from this I'd understand that I should never destroy the returned object... === salem_ is now known as _salem [12:24] zbenjamin: yup I have it, Installed: 5.2.1-1 [12:24] t1mp: ^ [12:25] nik90: and /usr/share/icons/suru/actions/scalable/view-expand.svg exists for you? [12:25] * nik90 checks [12:26] t1mp: nope I don't have it [12:26] nik90: is suru-icon-theme installed? [12:26] zbenjamin: yup [12:26] zbenjamin: I have the suru folder and the icons in it [12:26] ok then those files are new :/ [12:26] but that particular icon I cannot find [12:26] means i need to ship em [12:27] zbenjamin: or the icon theme needs to be updated on trusty [12:27] zbenjamin: wouldn't backporting ubuntu-suru-icons be easier? [12:27] after all it is just a icon theme [12:27] t1mp: i would vote for that ! [12:27] t1mp: but it requires a SRU still [12:27] nik90: otoh, a delegate can use the "model" role to access the Alarm object, so in cases where the actions are performed on the delegate, there's no need to use the get() function to retrieve the Alarm reference [12:28] zsombi: oh yeah, I could just have done model.cancel() and that would work as well? [12:28] nik90: and that is even documented [12:28] t1mp: except we backport it ourselfes in the ppa [12:28] nik90: yes [12:28] zsombi: tbh I don't remember why I did what I did there [12:28] nik90: perhaps because in the early phases teh "model" role wasn't there [12:29] I suppose [12:30] zsombi: I noticed I can access the alarm properties in a delegate by just doing "type, name, date" etc...couldn't I hence just do cancel() instead of model.cancel() ? [12:31] nik90: no, because a model can only export roles, cannot export slots [12:31] zsombi: ah .. ok [12:31] nik90: those you can access now are roles [12:31] nik90: for convenience the roles are named same way the Alarm properties [12:32] nik90: "model" role is a QML-is way to expose teh Alarm object itsemf ;) [12:32] ah yes true === _salem is now known as salem_ === salem_ is now known as _salem [12:39] bfiller, have there been any issues with the camera recently that are pending landing? It used to work pretty well, but on the last couple of weeks, starting or using the camera on mako either freezes the whole UI or reboots the device === rmescandon is now known as rmescandon-lunch [12:43] jdstrand, it is done. [12:50] jhodapp, oh hey the music-app designer is here [12:51] ah shame ahayzen and vic arent online [12:52] jhodapp, question: Ahayzen and I were discussing the choice to not have a album label on the now playing page, and we decided that it would lead to a bug in cases where no bug was defined. [12:52] what say you? [12:53] * DanChapman just sat here and did all the en_GB translations for dekko. "Apparently" you have to click "Save & Continue" after each page and not just "Next". [12:53] * DanChapman starts again [12:55] jhodapp, this is a comparison between really the two styles. [12:55] http://i.imgur.com/k15ifrw.png [12:57] jhodapp, this one is better http://i.imgur.com/daNXXMQ.png [13:00] rpadovani, good work with the multiple accounts branch! I've reviewed and added a commend with some design suggestions: https://code.launchpad.net/~rpadovani/reminders-app/multipleAccounts/+merge/237682 [13:01] beuno: thanks! [13:02] dpm, thanks! I haven't freeze on desktop switching account. About the dropdown, I'm not sure to understand what you mean: you *need* to go to online account settings to authorize an account, for security reason [13:02] it's impossible that an app could choose to use an unauthorized account [13:03] rpadovani, yeah, I understand that. What I mean is that when you tap on the option, it should go to online accounts, rather than using the Add account button for that [13:03] dpm, ah, understood! [13:04] sorry, that was perhaps not too clear on my drawing [13:08] mzanetti: hi [13:08] hi mivoligo [13:09] mzanetti: just wanted to let you know I've added 5 new levels and sent a merge proposal [13:09] mivoligo: cool, thanks [13:09] will look at it tonight [13:10] mzanetti: feel free to rearrange them :) === rmescandon-lunch is now known as rmescandon [13:21] beuno, sergiusens, bzoltan: hi! do you guys know the status of fat packages these days? like, does the store handle them right? does the sdk? does installing on the device? [13:22] jdstrand, the store does, yes [13:23] beuno: what will the store do if it gets an i386 only package (assuming that the click-reviewers-tools says that is ok). it will be accepted but devices won't see it because the arch doesn't match? [13:23] jdstrand, correct [13:24] the reverse case is common, where the emulator is i386 and a lot of packages aren't visible because they are armhf [13:24] beuno: right now the scripts don't consider i386 as valid, but it was pointed out to me that makes it difficult for the sdk to deploy on the emulator, so I was thinking of adjusting that [13:24] yeah, true [13:24] dpm: it's related to the location service [13:24] * beuno nods [13:25] (and then that reminded me of fat packages in general) [13:25] dpm: camera is trying to instantiate it at startup and it's either hanging or crashing [13:25] dpm: working on a fix [13:25] s/in general// [13:26] jdstrand, a while back, I saw them in use, but had to be built manually [13:27] jdstrand: it is handled [13:27] I guess stats by arch is something I should add [13:27] sergiusens: and specifying 'multi' in the click manifest is the way, correct? [13:28] jdstrand: I have a long overdue task to make click buddy auto build for the aches in the manifest (creating a fat if necessary) [13:28] * jdstrand is verifying the click-reviewers-tools [13:28] jdstrand: arch: ["i386", "armhf"] [13:28] jdstrand: or was it "architectures"? (but you get the point I hope :-) [13:29] I do [13:32] ok, thanks for the update, thanks bfiller [13:33] kalikiana, how do I override the translation domain if I want to use something different than app id? [13:33] where and how should I specify i18n.domain? [13:33] dpm: either set i18n.domain to set the default, for example in the MainView's Component.onCompleted [13:34] or use i18n.dtr("bla", "domain") [13:34] kalikiana, thanks. The second one is not an option, so I think the first should work. mzanetti ^ [13:34] kalikiana, can i18n.domain be set in main.cpp too? [13:37] dpm: UbuntuI18n* i18n = &UbuntuI18n::instance(); i18n->setDomain("domain"); [13:37] jdstrand, it seems that the c-r-t have started throwing errors not detecting i386 as a valid architecture. As an example, this click: http://people.canonical.com/~dpm/click/com.ubuntu.reminders_0.5.latest_i386.click - is this a known bug? [13:40] zbenjamin, a question: does QtC injecting the debug policy in the click package mean that if I publish a click generated with QtC to the store, it will fail the upload? I.e. it will have the 'debug' policy in it, which is not allowed [13:40] dpm: thats why we have the publish page, if you create it there it will be clean [13:41] zbenjamin, I don't quite understand. Do you mean that QtC creates two types of click packages, depending on whether you press the Play button or use the Publish page? [13:41] dpm: thats what i mean [13:42] oh [13:48] akiva-thinkpad, I'm not the music-app designer :) [13:49] akiva-thinkpad, I'm the backend guy for media [13:49] jhodapp, didn't you do the remix design? [13:49] akiva-thinkpad, nope [13:50] o_o sorry :P [13:50] ... who the heck was it then... [13:50] I could have sworn it was you :P [13:56] akiva-thinkpad, I forget, but ahayzen should be able to tell you [13:57] mihir, nik90 , ok we got approved from SDK guys (thanks zsombi ) https://code.launchpad.net/~renatofilho/ubuntu-ui-toolkit/save-alarms-in-local-time/+merge/237598 [13:58] mihir, nik90 could you test eds branch again, and check if all still working: https://code.launchpad.net/~renatofilho/qtorganizer5-eds/fix-1311165/+merge/237142 [13:58] nik90, you will need SDK branch to test alarms ^^ [13:59] renato___: could you confirm that eds backend does not send itemChanged() signal when the eds data is changed due to non-organizer initiated change? [14:00] zsombi, what do you mean? For example if I change the item with evolution app you do not get the change on QtOrganizer?] [14:01] renato___: yes [14:01] zsombi, it should works (, I will test it [14:02] zsombi, after my current meeting [14:02] renato___: ok, just drop me a mail, I [14:02] I'll be away soon [14:03] renato___: aaah, no, it does work!!!!!!!!!!!! [14:04] renato___: awesome ;) [14:05] renato___: :woot: awesome [14:05] renato___: thanks a lot :) [14:05] * mihir hugs renato___ :D [14:07] \o/ === mpt_ is now known as mpt [14:17] zsombi, renato___ : Awesome. Please give me a an hour or two to thoroughly check both the MPs together with the clock app. [14:31] nik90, frantically working on it now [14:32] fginther: sry for the rush, [14:32] nik90, no worries, it needs to be fixed [14:37] kenvandine: is there something blocking https://code.launchpad.net/~ken-vandine/content-hub/lp1373086/+merge/237459 from landing? [14:37] mhall119, yes... it FTBFS in the silo [14:37] because it now has translations... imported from LP :) [14:37] Elleo has a branch [14:37] mhall119, so soon [14:42] nik90, is bug 1376513 something that needs to be fixed in the clock app or somewhere else on the platform? [14:42] bug 1376513 in Ubuntu Clock App "Alarm volume should use the stream-restore dbus api available in pulseaudio" [High,Triaged] https://launchpad.net/bugs/1376513 [14:42] dpm: clock-app [14:42] dpm: I will try to get that implemented this weekend [14:42] ok, thanks [14:42] awesome :) [14:44] kenvandine: ah cool [14:50] zsombi, renato___: Works good. Tested clock app with both the MPs. I commented in the MPs. [14:50] nik90, thanks === oSoMoN_ is now known as oSoMoN [15:13] nik90, balloons, https://code.launchpad.net/~nik90/ubuntu-clock-app/sync-bottomedge-sep08/+merge/236604 passed \o/ [15:13] oO [15:13] nik90, balloons switched out the WM from fvwm to xmonad [15:13] fginther: woohoo yay..thnx [15:14] fginther, awesome. Yea, I found xmonad worked great locally in tests, as it autofocused the window on launch [15:14] fginther, balloons: I presume the xmonad change is only for clock jenkins? or for jenkins as a whole? [15:14] balloons, I wish I had discovered that fvwm was in use a few days ago, became a real simple update [15:15] nik90, this does apply to all testing at this time [15:15] fginther, ohh.. heh, so a wm was already running, just not a useful one [15:15] either way, this is good stuff [15:15] fginther: cool, this should make the testing environment more comparable to the user environment [15:15] yep yep [15:16] balloons, exactly :-), although I have some fondness for fvwm [15:17] balloons, thanks for doing the investigation on the WM to use === gatox is now known as gatox_lunch [15:37] popey: hmm... this is interesting: http://i.imgur.com/EFBoFAH.png [15:37] i have had that [15:37] its submitted and you can go in and manually to the rest of the steps [15:37] bueno ^ [16:00] DanChapman: Hi, are you around? [16:12] m-b-o, hey ,how are you! [16:13] balloons: fine, thanks and yourself? [16:13] excellent === _salem is now known as salem_ [16:15] balloons: started the weather app with this new cpp extension for TimeZones. No way to get the app started out of QtcReator, do you know? [16:17] m-b-o, what do you mean get the app started? [16:18] m-b-o, this? https://code.launchpad.net/~ubuntu-weather-dev/ubuntu-weather-app/timezone-plugin/+merge/237613 [16:18] m-b-o: You need to open the CMakeList.txt file in Qtc, and if the cmake file has been properly configured, you should be able to run it fine by qtcreator. Qtc will automatically build the plugin when you press the green run button. [16:19] balloons: wll, that didn't happen. I'Ve followed the steps in http://bazaar.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/utopic-3.0/view/head:/README.developers to get it run === gatox_lunch is now known as gatox [16:23] nik90, ping for review [16:23] rpadovani: hey, let me grab the MPs [16:24] m-b-o, so you are a-ok now then? [16:24] rpadovani: https://code.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/utopic-3.0/+activereviews [16:24] rpadovani: You need to start from 10-alarm-change onwards.. [16:24] rpadovani: ignore the top 2 [16:25] balloons: if there would be a way to star the app from QtC, it would be cool to do so [16:25] nik90, gotcha [16:25] rpadovani: Also you don't need to review the qmltest files in those MPs. They have already been looked at by elopio. [16:25] thanks elopio :D [16:25] rpadovani: and finally, these Mps are for the release tomrrow..so not required to complete the review tonight if you are busy [16:26] nik90, nah, itΕ› ok [16:27] cool, thnx mate === karni is now known as karni-afk [16:34] rpadovani: don't top approve yet, since I need to update the debian changelog and also merge trunk (trivial stuff) which I am doing one by one. I will top approve once I do it. [16:35] nik90, gotcha [16:35] nik90, https://code.launchpad.net/~nik90/ubuntu-clock-app/11-remove-utils/+merge/237047/comments/583292 [16:35] rpadovani: cool, will fix === salem_ is now known as _salem === karni-afk is now known as karni [16:48] rpadovani: fixed the comment in the above MP ^^ [16:51] nik90, https://code.launchpad.net/~nik90/ubuntu-clock-app/12-alarm-list-design-spec/+merge/237177 [16:55] rpadovani: ah I did the code so since if the text is hidden, it still calls alarmUtils.get_time_to_next_alarm(model.date - localTime) which is unnecessary cpu wastage. hence when the alarm is disabled, I give it a static text. [16:56] nik90, makes sense [17:05] nik90, done [17:05] rpadovani: btw no need to code review https://code.launchpad.net/~nik90/ubuntu-clock-app/14-sync-listitem-actions-oct5/+merge/237189. It just syncs up with upstream code which was already reviewed. If you want you can just test on phone to see if it caused any regressions [17:05] rpadovani: cool, thnx a lot [17:06] rpadovani: I am merging them one by one as they are dependent on each other and so jenkins cant merge them all at once [17:06] nik90, you're welcome. I left a little comment on 13 about another way to implement an if, that IMO is better, but they are tastes :-) Anyway, I approved it [17:07] rpadovani: I am taking a look atm [17:08] popey: shall we start the clock meeting? [17:08] nik90: sorry, didn't realise we had an overlap! lets [17:10] m-b-o hey :-) have you managed to get it to run? sorry i havn't had a chance to get back to looking at dholbach's packaging changes. It was working before that just jenkins wouldn't have any of it. I'll branch it now if you are still having issues [17:11] danChapman heyho! :) Yes, got it run. But not out of Qtc, though. [17:12] ahayzen, ah how goes it? [17:12] m-b-o, what's the issue you're having with getting it to run? How are you trying to get it to run? [17:13] akiva-thinkpad, hey good thanks afternoon lectures today :) lol [17:14] dpm: Using the TimeZone Extension in QML and run it from within QtC fails. [17:14] dpm DanChapman " module "TimeZone" is not installed" [17:14] m-b-o it's just run straight off the bat from QtC for me. Ahh... i still have the TimeZone plugin installed from last time. [17:15] * DanChapman gets rid of that [17:15] I'Ve followed http://bazaar.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/utopic-3.0/view/head:/README.developers than and that works, but outside of QtC [17:16] * dpm reads doc [17:21] m-b-o, ah, so the issue you're having is to run the weather app + plugin from QtC! Sorry, I thought it was the other way round. What's the error you are getting? [17:21] dpm " module "TimeZone" is not installed" [17:22] balloons, in response to your comments this mp was waiting for someone to review it... at the time my url-dispatcher had decided music:// should open facebook lol and victor was/is unable to use adt ... https://code.launchpad.net/~andrew-hayzen/music-app/add-url-dispatcher-tests/+merge/233840 [17:22] m-b-o, ok, so it's not finding the module in the expected location. Are you running it on a desktop kit or on the emulator? [17:23] dpm desktop [17:23] m-b-o: dpm ok so looking at http://bazaar.launchpad.net/~ubuntu-weather-dev/ubuntu-weather-app/timezone-plugin/view/head:/plugins/CMakeLists.txt#L26 seeing as no -I plugin/path/TimeZone is being passed to qmlscene apart from the $@ then would it not be looking for the plugin in ${QT_IMPORTS_DIR}/TimeZone and not ${QT_IMPORTS_DIR}/qt5/qml/UbuntuWeather/TimeZone/ [17:25] m-b-o, DanChapman, the other issue is that with reminders, terminal and file manager, where we've got this same setup, we have a rule in cmake that copies over the compiled plugin to the build directory, and adds the build directory to the list of locations where the plugin is looked for, so that you don't need to install it on the system. I'm not sure if that branch has that rule, I've not looked at it [17:30] rpadovani: addressed all your comments [17:31] mzanetti, see, I'm on it. Nevermind for layout of delegate, let do it in another branch [17:33] mzanetti, about the message, I wrote the wrong thing: if you don't have any tag, and you go to the tab of tags, there isn't any message [17:33] I think you have to add something here [17:41] mhall119: ping [17:42] snizzo: pong [17:43] mhall119: I'd like to know if I'll receive my tshirt for being a pioneer :) [17:51] ... [17:51] I'm losing faith... [17:52] brb [17:54] rpadovani: done [18:18] popey: you wanted me to revoew ubutu_title from splash MR , correct? [18:20] mihir: no, i think it should still be there for now. [18:21] popey: okay i recalled it, need to test from Calendar to Month view how does it go :| [18:21] yeah [19:54] security_template_exists (apparmor-push.json): specified unsupported template 'ubuntu-push-helper' , need to install anything? [19:56] mihir: what version of click-reviewers-tools do you have? you need 0.15 [19:56] jdstrand: how do i check that? [19:57] apt-cache policy click-reviewers-tools [19:58] jdstrand: http://paste.ubuntu.com/8528803/ [19:58] okay i need to upgrade my click-reivew tools [19:58] yeah [19:58] this one too regarding that correct, πŸ‘Ž error: security_template_push_helper(apparmor-push.json): template is not 'ubuntu-sdk' ? [19:59] if I understand your question, yes, you want to use the ubuntu-push-helper template in apparmor-push.json [20:00] * mihir upgrades [20:10] rpadovani, ping [20:10] balloons, pong [20:11] rpadovani, calculator doesn't seem to save calculations anymore between runs. It should be saving them right? [20:12] balloons, after you swype them up, yes [20:13] balloons, r329 works well on mako devel, where are you testing? [20:13] balloons, do you have any output log? [20:16] rpadovani, I'm playing with it on the desktop atm [20:16] rpadovani, ahh I see.. heh [20:16] * balloons notes we should have a test for that [20:17] balloons, mhh, what's wrong? I'm able to save them also on desktop [20:17] rpadovani, right.. I am too. I was just confused [20:17] balloons, popey says that the app will be redesigned soon :-) [20:18] fginther, ping [20:18] fginther, check out the giant calculator :-) http://91.189.93.70:8080/job/generic-mediumtests-utopic-python3/814/artifact/ubuntu_calculator_app.tests.test_simple_page.TestSimplePage.test_swipe_to_delete_calculation.ogv [20:19] mzanetti: I forgot to mention I didn't update enemies waves in level.json files but you probably noticed that :) [20:20] fginther, so did you set the size for the window? [20:24] fginther, testing locally I set the xserver size to 400x600, and that emulated the phone size properly. If I left it as 1024x768 you get weirdness with xmonad on some apps [20:26] balloons, :-( Let me fix that [20:37] fginther, balloons: I saw it for clock as well, but it didnt affect the tests [20:37] balloons: it seems clock test is failing on https://code.launchpad.net/~nik90/ubuntu-clock-app/12-alarm-list-design-spec/+merge/237177 [20:38] balloons: I didn't seem to make any change that part of the code. But I can reproduce the failure locally as well. [20:40] nik90, ok, so how to fix? [20:40] balloons: yeah I am not sure what changed. I need to upgrade my VM after like 3 months and run autopilot vis to see what changed [20:41] nik90, so you feel it's legit or not? [20:41] balloons: its certainly not jenkin's fault :) [20:42] balloons: but the annoying part is that it fails at the very last step of the test http://91.189.93.70:8080/job/generic-mediumtests-utopic-python3/813/testReport/junit/ubuntu_clock_app.tests.test_alarm/TestAlarm/test_add_recurring_type_alarm_must_add_to_alarm_list_random_/ [20:43] nik90, ohh swipe to delete did change [20:43] but it's been some time [20:43] balloons: yeah but I am using custom swipe to delete (due to custom listitem with actions) [20:45] balloons: the function it fails at is https://bazaar.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/utopic-3.0/view/head:/tests/autopilot/ubuntu_clock_app/emulators.py#L519 [20:52] balloons: you mentioned previously that someone (I don't recall who) might be able to help with AP tests for the new tab feature in terminal? https://code.launchpad.net/~flscogna/ubuntu-terminal-app/uri-and-tabs/+merge/237353 [20:52] popey, yes senan was going to work on it [20:52] I've not seen him in a couple days [20:57] jdstrand: hey i am still getting that error while build it , i upgraded my system and it shows it has 0.15 click-review tools. [20:57] jdstrand: my bad error is different ,πŸ‘Ž error: security_template_exists (apparmor-push.json): specified unsupported template 'ubuntu-push-helper' [20:58] mihir: hrmm, try 'rm ~/.cache/click-reviewers-tools/apparmor-easyprof-ubuntu.json' and then try again [21:01] jdstrand: that works , thanks :) [21:04] balloons: quick question, I remember autopilot vis (long time) used to highlight an item in the UI when you click on the object tree items. [21:04] balloons: do you know how to enable it? [21:04] nik90, use autopilot3 vis [21:04] ah thnx [21:04] I should update the readme docs [21:04] yea, I get confused sometimes too [21:05] it's correct in the readme [21:05] btw, I shamelessly modified it and using it for my calc mp :-) [21:05] * nik90 looks silly [21:05] balloons: lol [21:06] balloons: I wrote that guide to help noobs like myself [21:10] balloons: if you got some time, I need help debugging this clock failure together [21:10] nik90, I would love to help, but I need to finish this work on calc first [21:11] ok, I will try some things out meanwhile [21:11] ahayzen_, you about? [21:11] balloons, yep [21:12] ahayzen_, well, heh, technically I'm playing with the old music app so you don't care :-) [21:12] balloons, heh i have the job of syncing it with remix so i do care a little ;) [21:16] balloons, so what are you working on the old app for? [21:16] ahayzen_, so I'm trying to fix https://bugs.launchpad.net/music-app/+bug/1379488 [21:16] Ubuntu bug 1379488 in Ubuntu Calendar App "apparmor denial during test runs for /home/phablet/autopilot/fakeenv/*/.cache/QML/" [Undecided,New] [21:17] ahayzen_, as part of that, I'm going to be dropping support for patching home on the device [21:17] well, I'd like to try and do that anyway [21:17] yep ... balloons note i have branches up ready to put confinement back in (which is yet to be testing with autopilot ;) ) [21:17] balloons, ok so what would it do instead? [21:18] ahayzen_, nothing. IT would be up to the test runner to make sure the device is clean during test runs [21:18] I'll leave basic mocking for the desktop for now [21:18] balloons, so we would copy things directly into ~/Music ? [21:18] ahayzen_, yep [21:18] balloons, so what would happen to all my music on the device? [21:20] ahayzen_, nothing. the tests would fail though as-is [21:20] you would need to run with adt-run to ensure you have a safe env setup [21:21] ahayzen_, we'll be talking more about this in a couple weeks at the sprint, but here's the metabug for it: https://bugs.launchpad.net/ubuntu-app-launch/+bug/1376423 [21:21] Ubuntu bug 1376423 in Ubuntu Application Launcher "There is no easy and future-proof way of starting an app in a clean environment" [Undecided,Invalid] [21:21] balloons, 'safe env setup' what does that involve does it create a new profile? or move all my music somewhere else or what? [21:22] ahayzen_, I'm making these changes now, or thinking about it, because it's causing issues again in CI [21:22] oh yeah i remember seeing that bug [21:22] ahayzen_, I would rather not do anything until we have a better plan in place, but just wanted to give you the heads up [21:22] balloons, is causing all those failures on utopic CI or is that something else? [21:22] ahayzen_, yes [21:22] balloons, but RTM was fine at last check lol [21:23] balloons, ok so cut to the chase... what do we need to do to music-app autopilot then? [21:26] hey guys ... what's the normal way to create a db-driven ubuntu app with JS/HTML? [21:26] ahayzen_, I'll propose mp's you needn't worry about it [21:26] ahayzen_, we can keep the patch working by doing this: http://bazaar.launchpad.net/~nskaggs/ubuntu-calculator-app/precompiled-qml/revision/333 [21:26] but I'm considering just gutting everything [21:27] balloons, ok thanks :) [21:27] if I can get it to work [21:27] nik90, alright, I can help you out for a bit [21:28] balloons: ok so I am bit confused at the failure message I got [21:28] AttributeError: Class 'AlarmDelegate' has no attribute 'swipe_to_delete'. [21:28] when does that happen? [21:29] I thought it was timing error. So I added a sleep(10) in between that time to confirm it [21:29] but it is not a timing issue [21:30] and the MP I linked shows the code diff to AlarmDelegate.qml which is causing them failure. I am certain that it didn't change anything related to swipe delte [21:30] delete* [21:32] nik90, it's not a timing issue. As it says it doesn't understand swipe_to_delete [21:32] are you sure you have an instance of the object you think you do? [21:33] balloons: I think I found the issue (but do not know the syntax to fix it) [21:33] balloons: here is the old object tree in trunk https://i.imgur.com/L9uq0bx.png [21:33] balloons: here is the new one [21:33] https://i.imgur.com/SBIWRwY.png [21:33] do you see the difference? [21:34] can you expand the new one? but yes my guess is yo udon't have the object you think you do :-) [21:34] so just tweaking a select is probably all we need to do [21:34] yup the listitemwithactions object type got changed into AlarmDelegate object type [21:34] there you are ;-) [21:34] cool I'll assume you are sorted! [21:35] yup :-) [21:36] balloons, how are you forcing 400x600 resolution? X doesn't appear to like this for me [21:39] balloons: quick question, in https://i.imgur.com/uuzd8wv.png, how do I refer to AlarmDelegate as a class type? Previously I did class ListItemWithActions(ubuntuuitoolkit.UbuntuUIToolkitCustomProxyObjectBase): [21:39] fginther, you are using xvfb right? So -screen 0 400x600x24 [21:39] balloons: however doing class AlarmDelegate(ubuntuuitoolkit.UbuntuUIToolkitCustomProxyObjectBase): doesnt seem to do it [21:39] balloons, ahh, these tests have always been using X [21:40] k [21:40] let's figure this out [21:40] in the context of https://bazaar.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/utopic-3.0/view/head:/tests/autopilot/ubuntu_clock_app/emulators.py#L519 [21:42] fginther, so locally I'm running like so http://paste.ubuntu.com/8529327/ [21:43] balloons, thanks, that helps [21:44] balloons, seems your copy and paste has to be improved :P [21:44] https://code.launchpad.net/~nskaggs/ubuntu-calculator-app/drop-fixtures/+merge/237848 [21:46] fginther, I based that on autopilot3-sandbox-run.. not sure if using that script directly would be helpful or not. [21:48] rpadovani, ack, thank you.. pushed the fix [21:49] balloons, good, do you mind if I wait until tomorrow to test it on device? I'm a bit tired right now [21:50] rpadovani: btw I got an email about your webbrowser branch [21:50] rpadovani: did you check with osomon before? [21:50] rpadovani, if you can approve without it, I've been asked to get this fixed asap. Obviously you needn't say you tested on the device :-) [21:50] rpadovani: last time I checked he said he wanted it to first land in the SDK before using it in the webbrowser app. [21:51] nik90, sure, I'm working on implementing some feature for history, and your bugs will be fixed [21:51] nik90, oh, well, we didn't talk of details of implementation, I'll ping him tomorrow [21:52] balloons, ok, but if it will destroy the world is your fault, not mine :P [21:52] balloons, you have a go [21:55] rpadovani, ack [22:02] fginther: hope this is not a bad time to ask, but if you look at http://91.189.93.70:8080/job/ubuntu-clock-app-ci/buildTimeTrend, after the WM fix, the build times have significantly increases from ~6mins to ~50mins. Was this expected? [22:04] nik90, In this case, multiple ubuntu-clock-app-ci jobs all started at the same time, but the only one autopilot job could execute at a time, so most of that time was just waiting [22:05] fginther: ah ok [22:13] balloons, xvfb is setup now and passed the first few tests [22:14] fginther, ack, awesome [22:14] I'll try landing the calc mp [23:24] rpadovani, any chance you are still around? [23:24] balloons, maybe :-) [23:25] rpadovani, one more quick calc mp.. I wanted to toss this in. It fixes the asserts so you get the proper result when there is more than one [23:25] https://code.launchpad.net/~nskaggs/ubuntu-calculator-app/fix-ap-asserts/+merge/237874 [23:25] I want to get it merged and do a release so everything is fixed [23:25] I'm on it [23:27] rpadovani, I reverted the Screen.qml whitespace change hehe [23:28] balloons, yes, your mr lgtm, and the implementation seems very elegant :-) [23:29] balloons, ok, top approved, I think I'll go to sleep now, last 4 days were very intensive :D [23:29] rpadovani, thank you much.. new calc for you in the morning ;-) [23:29] \o/ [23:47] Bye