=== chihchun_afk is now known as chihchun [07:29] good morning [08:51] good morning [09:12] hello! my qmake for ubuntu armhf refers to a directory "/var/lib/schroot/chroots/click-ubuntu-sdk-15.04-armhf/usr/lib/arm-linux-gnueabihf/qt5/libexec" but it does not exist [10:29] bartbes, did you figure out what that xcb error is? "This application failed to start because it could not find or load the Qt platform plugin "xcb"." [10:31] zbenjamin, I can't run apps on my 14.04 laptop with the sdk installed - i get the above error http://imgur.com/VHaWITm - which I know bartbes had over the weekend too.. seen it? [10:32] I didn't, I made sure I had all xcb packages, I did a clean all and a rebuild, no success [10:32] reboots, no luck [10:32] and then suddenly it worked again [10:33] gah [10:35] popey: hm weird never saw it [10:36] popey: do you use the ubuntu-sdk-ide package? or the old one from archive? [10:36] hmm, lemme see [10:38] zbenjamin, Installed: 3.5.0+15.10.20151024-0ubuntu1~97+201511170914~ubuntu14.04.1 [10:38] re-launch and now I get a different error :) [10:40] popey: actually, I noticed that launching outside of the sdk worked [10:41] and then after I accidentally also managed to launch it from within it [10:53] zbenjamin, seen this? "module "QtQuick" version 2.4 is not installed" [11:04] so why does the ubuntu-sdk-ide require opengl anyway? [11:15] well, adding the nvidia drivers to the docker container does allow it to start [11:16] it doesn't actually run though === boiko__ is now known as boiko === _salem is now known as salem_ [11:39] hello! does anyone know what "/var/lib/schroot/chroots/click-ubuntu-sdk-15.04-armhf/usr/lib/arm-linux-gnueabihf/qt5/libexec" is for? [11:40] (I found this path in the qmake for ubuntu armhf, but it don't exist) [12:04] popey: you probably need to use the new Kit that comes with the IDE [12:04] zbenjamin, oh, is that documented somewhere? [12:05] popey: probably not yet. Its still pretty new and works only for apps atm [12:05] ok [12:05] popey: but you get the most recent UITK even on LTS [12:05] \o/ [12:05] where from? Can I get that now? :) [12:05] or is it still in progress? [12:05] popey: do you have ubuntu-sdk-ide and ubuntu-sdk-dev installed? [12:05] yes [12:06] popey: then you should just be able to add it to your project as a new target [12:06] how? [12:06] popey: in the projects tab, add Kit i think [12:07] popey: should be named "Ubuntu SDK Desktop Kit" [12:07] got it [12:07] popey: its still very basic and has only the basic Qt + UTIK modules. So if you use advanced APIs they might not be available yet [12:08] yeah [12:08] file:///home/alan/Development/music-app/app/music-app.qml:23 module "Ubuntu.MediaScanner" is not installed [12:08] :( [12:08] popey, yup, i had that :-) [12:08] popey: but we can add more as long as they can be compiled on LTS [12:09] ahayzen, do you have a hilight on music-app perchance? :) [12:09] popey, 'music' ;-) [12:09] ah [12:09] ouch [12:09] :-) [12:10] zbenjamin, is the desktop kit not like the arm one where you can just apt-get stuff ? [12:11] ahayzen: no, its installed alongside the system qt, we are aiming to get more and more independent from the system installed qt so we can support multiple ubuntu versions with the same SDK [12:11] ah ok [12:11] ahayzen: so the goal is: system env != dev env [12:12] cool :-) [12:12] popey: oh, I did run cmake again in Debug mode, maybe that helps? [12:19] is there a framework for 16.04 yet? [12:20] or will not OTA-9 be based on Xenial? [12:29] ahayzen: hey :) [12:29] faenil, o/ [12:30] ahayzen: been looking for your last week :D [12:30] ahayzen: https://code.launchpad.net/~faenil/ubuntu-ui-toolkit/listitemlayout_fixes_and_docs/+merge/281919 [12:30] faenil, hah, sorry i wasn't around :-) ooo [12:30] ahayzen: np, just wanted to share the MR ;) [12:31] so that should work when text wraps now ? [12:31] and i guess will be in the OTA10 UITK ? [12:32] ahayzen: no more string overlap, + wrapping changed to avoid early word cuts, + docs about aliasing labels properties [12:33] coolio [12:33] and yes, too late for OTA9, so OTA10 :) [12:33] zbenjamin, also, https://plus.google.com/u/0/101489925843882654647/posts/3L3HTVJ2cKs - others having sdk problems [12:35] ahayzen: but it will be soon on rc-proposed [12:35] (that is, next landing, I don't know when that is happening, bzoltan_ knows) [12:36] faenil, yup :-) i'll have a play on my laptop later, as i have a branch of the uitk [12:36] cool [12:37] ahayzen: note you don't have to set left and right anchors of listitemlayout, save those bindings :D [12:37] :-) [12:38] faenil, also if jenkins would work, then the first round of listitem layout stuff would be in music-app trunk already :-) [12:38] hehe [12:39] faenil: ahayzen: The next UITK landing is expected after the OTA9 is released [12:39] cool :-) [12:39] bzoltan_: so rc-proposed will be on OTA9 uitk until end of Jan? is that correct? [12:40] faenil: Yes [12:40] ok, thanks [12:44] bzoltan_: any idea? https://plus.google.com/u/0/101489925843882654647/posts/3L3HTVJ2cKs [12:46] zbenjamin: there was no change in the package [12:47] zbenjamin: is he using the official relese? [12:47] release [12:47] bzoltan_: at least he is running "ubuntu-sdk-ide" [12:50] zbenjamin: but hopefully not from tools dev or staging [12:50] popey: is he on IRC too? [12:51] don't think so. [12:54] is there a release schedule somewhere? [12:57] bartbes: for OTA release of UbuntuTouch? [12:57] releases* [12:57] yeah [12:57] bartbes: there are daily emails that the landing team sends to ubuntu-phone mailing list [12:58] bartbes: roughly every 6 weeks, we are behind on this one due to the Xmas Shutdown [12:58] whenever the schedule for a new OTA is decided, the landing team will communicate that using that mailing list [13:00] I see the next one's scheduled for the end of the month? === rickspencer3_ is now known as rickspencer3 [14:54] hey, I'm wondering how many app developers are writing autopilot tests for their apps, and using the UITK autopilot helpers (formerly: custom proxy objects) for that? [14:55] and how do you run these tests? In an automated way (jenkins), or locally on your computer or device when you make changes? [14:56] I am wondering because I am working on scrolling in the header sections, and quite some work is needed to add proper autopilot proxy objects to deal with the scrolling [14:56] I am adding the CPOs anyway, but I am curious who will use it :) [14:59] ahayzen, renatu, popey: ^? [14:59] o/ [15:00] autopilot _should_ be running for all core apps merges but it broke when we moved from one jenkins instance to another, but should be re-enabled (soon, balloons ?) :D [15:00] it's running right now [15:00] hence things aren't landing, hah [15:00] t1mp, you mean like MainView, UbuntuUIToolkitCustomProxyObjectBase, UCListItem ? [15:01] balloons, o/ [15:01] popey, as in, we fixed the issue with IS last week that made it blow up [15:01] popey, balloons: ah, cool. Will each core app have their own jenkins? And also a device attached to that jenkins for testing? [15:01] ahhh [15:01] one for all core apps [15:01] t1mp, I am trying to use most of SDK proxy objects as possible, but I faced some problems while trying to use it with the new AdaptiveLayout [15:01] t1mp, http://core-apps-jenkins.ubuntu.com/. You can request a build of anything you like [15:01] there is only one device shared amongst them all === chihchun is now known as chihchun_afk [15:02] balloons, https://code.launchpad.net/~ahayzen/music-app/fix-1526274-use-layouts/+merge/281757 is failing is jenkins...but passes on device/desktop :-/ [15:02] there's going to be a bit of that. I'm not sure if popey wants to back off the stance of failing merges on the backs on test results or not [15:02] balloons, so me and Victor aren't sure what is going on, as it is specific to one view as well :-/ [15:02] ahayzen: in python, you'd select_single to get a MainView, and then on that object you'd get functions like get_header() from the CPO. [15:03] btw, trv is broken on xenial now.. not sure why [15:03] I'd rather we had AP running than not, and I'd rather the tests pass than not, unsurprisingly :) [15:03] balloons: cool. That makes sense. For UITK we have our own jenkins though so personally I don't need that one. [15:03] balloons: will there be a jenkins for community apps too? [15:03] ahh brillant [15:03] no [15:03] not for non-core community apps [15:04] t1mp, yup we do that sortof stuff :-) trigger_trailing_action etc :-) [15:04] In my experience non-canonical non-core don't use Autopilot [15:04] Unless I'm completely mistaken [15:04] popey: I guess you are mostly right (but it is my guess). [15:04] and if community app developers want that, they can ask here, right? :) [15:05] ahayzen: okay, great :D [15:06] Thanks for the information people. You convinced me that what I'm doing right now will be useful for apps too :) [15:06] :-) [15:06] (I don't think community developers want autopilot) [15:06] (call me a filthy heathen, but I doubt anyone outside canonical wants/uses it) [15:06] * ahayzen wants autopilot \o/ [15:07] * mcphail chortles [15:07] * t1mp looks up chortle [15:08] * popey sniggers [15:09] * t1mp suddenly feels like eating a snickers bar [15:09] but I don't have one... thanks popey! ;) [15:09] yw [15:17] balloons, any ideas how we can fix the jenkins issue, as it passed here https://core-apps-jenkins.ubuntu.com/job/run-ap-tests-ci/78/ ... then when it came to autoland (no changes in the branch) it then fails https://core-apps-jenkins.ubuntu.com/job/run-ap-tests-autolanding/2/ ... and we can't replicate the failures on our devices yet :-/ [15:18] so ahayzen, popey I can turn off failing to merge when tests fail for now. But at some point, we'll need to clean these up [15:18] Right now it seems to be causing havoc as expected. The question is, are we ok with trying to sort through all the apps at once? [15:19] :( [15:19] balloons, but why is it failing then? if it passed a few minutes before on the test run via a commit ? [15:19] what's the difference ? [15:20] you can see in the subunit everything passes for #78 then in autolanding #1 and #2 it fails on two specific tests [15:21] gang65, see conversation and chime in here [15:22] ahayzen, ohh, sorry, trying to look at calendar, calc and you too [15:22] balloons, haha :-) no worries [15:23] so what I'm looking for is some feedback on if you want to block landings on this or not. For instance, one thing we could do is still have it fail during normal merges, but don't stop on landings (top approval) [15:24] in other words, the test run job doesn't have to cause jenkins bot to issue a fail, even if it fails. We can run it and simply record the results [15:25] i would prefer it to block landings, as it ensures our test suites pass [15:26] if the bot itself could be consistent and pass/fail correctly, unlike at the moment how it says pass one minute then fail the next :-) [15:26] sorry gang65, we're here, not #ubuntu-app-dev. I can't type some days [15:26] :-) [15:26] hi [15:26] ahayzen, well let's look at your specific case then as to what happened [15:26] the jobs are the same [15:26] balloons, the jobs are the same, nothing specific in the logs [15:26] passes on our device [15:27] passes on the run-ap-tests-ci [15:27] ahayzen, what does subunit show? [15:27] fails on run-ap-tests-autolanding [15:27] that it couldn't find a CardView to then get a count from it [15:29] ahayzen, you found a bug [15:29] and it's both my fault, and sneaky [15:29] let me confirm, then i'll explain [15:30] :-) [15:30] indeed.. sad panda. the autolanding run used *961 version of the click. You can see yourself [15:31] it's an old click -- not sure were it's copying it from. Let me fix the job [15:31] balloons, i thought that was because it used 960 then merged my branch in and committed...but maybe not ;-) [15:31] (960 is trunk) [15:31] balloons, or should it be the same version as the latest from the proposed branch ? [15:31] ohh.. interesting. I just noticed that the other run used *965 [15:32] one of them isn't right, hah [15:32] they should be the same, it's the same MP [15:32] :-) [15:32] so anyways, I'll fix that [15:32] so to gang65's and calendar, etc, problem. Should we be blocking landings or not? [15:33] I know ahayzen wants it; should we do it on a case by case basis at first [15:33] balloons, what are the other problems for other coreapps ? [15:33] popey? [15:33] balloons, simply that trunk fails to pass ? [15:33] well, my idea in doing this was to force everyone to get tests cleaned up [15:33] I mean, we can tackle them as they come up one at a time. Right now, 3 apps are blocked [15:34] msuic should be solved in a moment, so that leaves 2 more [15:34] balloons, sure if they don't pass they won't go into the store anyway, don't QA fail it if AP doesn't pass ? [15:34] *surely [15:35] they do now [15:35] so yes, we should fix them [15:35] \o/ :-) [15:37] ok, so gang65 in order to land, we need to fix your tests. Make sense? [15:51] yes [15:57] gang65, so let me fix the issue ahayzen found, then I can help with your tests. [15:57] it's ok that these merges will be held up for a little bit? [15:57] Thanks [15:58] We adopted our tests to pass on Desktop [15:58] We will need to readapt it to phone again [15:59] (on Desktop Calculator is running in Landscape mode by default) [16:01] right, so hopefully it shouldn't be too bad to re-adapt ;-) [16:04] ahayzen, I believe your issue should be sorted. Jenkins re-enabled a option I didn't want but only on that autolanding job [16:04] Let's see if it's a-ok now [16:04] gang65, let's look at your stuff [16:04] balloons, ok thanks :-) [16:06] balloons, something went pop https://core-apps-jenkins.ubuntu.com/job/run-ap-tests-autolanding/13/ :-/ [16:07] hah [16:07] I like to imagine a "pop" sound in the data centre [16:07] piped directly to balloons' head [16:07] +1 [16:10] one thing further snafu found and fixed [16:10] :-) [16:10] balloons vs jenkins :-) [16:11] nice.. a little bug in every autolanding config. Lovely [16:11] * balloons fixes the others [16:15] ok, everything looks fixed and proper now [16:16] https://core-apps-jenkins.ubuntu.com/job/run-ap-tests-autolanding/15/console [16:17] as balloons says that 'psutil.NoSuchProcess: no process found with pid 30531' appears in the console [16:17] ohh, lol.. That actual run is bombing hah [16:18] apparmor="DENIED" operation="open" profile="com.ubuntu.music_music_2.3.965" name="/custom/xdg/data/dconf/profile/custom" pid=29360 comm="qmlscene" requested_mask="r" denied_mask="r" fsuid=32011 ouid=0 [16:18] the kernel hates us [16:18] hah \o/ [16:21] heh [16:24] are you sure this works? [16:24] jenkins is running it now [16:25] well it passed here https://core-apps-jenkins.ubuntu.com/job/run-ap-tests-ci/78/ [16:25] but the autolanding failed again [16:25] I'm running the old -ci job [16:26] ah :-) [16:26] perhaps I really fixed things for you [16:26] but seriously, jenkins is all good now hehe [16:26] it was working on mako over the weekend via adt-run :-) [16:28] looks like it's running fine [16:28] https://core-apps-jenkins.ubuntu.com/job/run-ap-tests-ci/83/console [16:29] top approving again -- let's just see what happens [16:46] balloons, 'Status: Approved => Merged' \o/ [16:47] balloons, sweet, jenkins seems happier thanks :-) [17:37] ahayzen, yw.. Thanks for noticing the little snafu. I couldn't test automerges as well naturally before rollout [18:06] hey [18:06] will Ubuntu Touch be available in the US soon? [18:21] AbuDhar, that would be good, but depends on your definition of "soon" [18:21] balloons, am I right in saying on jenkins everything builds in a chroot? [18:21] AbuDhar: wrong channel for that question, but ubuntu is available worldwide. there are no US retail phones with ubuntu pre-installed yet though. it's up to manufactuers to announce if/when they will release any such devices. canonical/ubuntu don't manufacture devices [18:22] balloons, I re-installed my laptop and lost my chroots (forgot they were omitted from backups) so am re-making them and need the package list to enable me to build filemanager and docviewer (as we had to jump through some hoops to get python2/3 installed in various ways - if you remember) [18:22] balloons, so if you could "dpkg -l" in the chroot and pastebinit I'd appreciate it :) [18:22] popeym ahh [18:22] ok, give me a second. if you'd like, jenkins can build any branch of those you wish [18:23] AbuDhar: if you want a phone in the US running ubuntu right now, best option is buying a nexus 4 off ebay or somewhere, and flashing it yourself [18:24] balloons, nah, I need to build them locally [18:24] popey, soon means within a year [18:25] I have it running on a OnePlus One which isn't quite daily-driver quality, but performs well [18:25] and is nice and fast [18:30] I wonder when manufacturers will make one that works in US [18:32] There are phones that work in the US :) [18:32] Just not sold in the US with Ubuntu pre-installed. [18:33] yeah I was looking for a phone that had Ubuntu pre-installed. :) [18:34] but I don't live in the US. I was asking for someone else actually. [18:46] popey: oh, re chroots, those apps have debian packages in the archive as well, right? [18:50] popey, https://core-apps-jenkins.ubuntu.com/view/Maintenance%20Jobs/job/list-chroot-packages/1/consoleText [18:50] heh, that's one way to do it! [18:51] dobey, pass? [18:51] popey, you should have permissions to edit the job / re-run it as neded [18:51] multipass ! [18:51] heh [18:51] thats perfect balloons [18:51] popey, you can see there's chroots for 2 frameworks and all arches [18:52] Leeloo Dallas Multipass! [18:52] popey: was just going to say, enable the deb-src lines for the archive and overlay ppa (if a vivid chroot) and then apt-get build-dep filemanager-app [18:52] sadly not that easy [18:52] but if it's not packaged in the archive too (like gallery-app is), that won't work [18:53] popey, though I will say; as with the other core apps I made tasks for, a writeup of readme's containing build info would be an excellent addition to those projects [19:07] Muuuuuulllllltttttttiiiiiiiiippppppaaassssssss [21:12] balloons, remember how we got past this? http://paste.ubuntu.com/14472515/ [21:12] * popey goes for fresh air, this is doing my head in (again) [21:14] I'd be interested if that can be circmvented. I've never managed to deal with python in a hybrid click chroot, and have had to use a full armhf chroot [21:14] i got it working once but it took some doing [21:14] and i could never update it [21:27] popey, ahh the fun bit [21:27] yes I have a script for that [21:29] popey, see: https://core-apps-jenkins.ubuntu.com/view/Maintenance%20Jobs/job/maint-chroot/74/consoleText [21:30] here's the madness behind it: http://paste.ubuntu.com/14472627/ [21:32] gosh [21:33] popey: why are you trying to install armhf python? [21:34] popey: you need to install the native python, otherwise /usr/bin/python won't run [21:34] and thus, configuring will fail [21:47] dobey, because libsmbclient [21:47] or samba-libs - one of the two [21:47] heh, thanks balloons [21:49] dobey, yes it's really so crazy packaging [21:49] oh [21:49] archive depends don't install cleanly [21:49] so honestly, I suppose it should be fixed [21:49] i guess samba needs fixed so you can cross-compile things that depend on it [21:49] yeah [21:49] there's a debian bug on it IIRC which is wontfix [21:49] right [21:55] balloons, do you have a separate / similar one for 15.04? [21:56] or is it identical bar the s/15.10/15.04/g [21:56] ^^ this [21:56] ta [21:57] * popey saves fix_chroot_nonsense.sh [21:59] ran it twice to be sure ;) === salem_ is now known as _salem [22:31] balloons, I note your package list shows you don't have bzr installed... [22:32] balloons, if that's the case how does it do the bzr_revno stuff? [22:53] Successfully built package in './com.ubuntu.filemanager_0.4.512_multi.click'. [22:53] \o/ [22:53] now I sleep. [22:53] thanks for the help balloons :)