=== chihchun_afk is now known as chihchun [00:53] zyga: you'd have to talk to beuno for changes in the store, is there a reason you want to use that different namespace rather than com.ubuntu.? === chihchun is now known as chihchun_afk === _salem is now known as salem_ === salem_ is now known as _salem === chihchun_afk is now known as chihchun === _salem is now known as salem_ === salem_ is now known as _salem [05:57] Good Morning all. [05:57] balloons: are these pep8 formatting errors on AP for calendar, http://91.189.93.70:8080/job/ubuntu-calendar-app-utopic-amd64-ci/235/console ? [06:26] I'm trying to download the SWORD library on qt but I'm not having any luck. [06:26] I just can't seem to install it [06:58] Has anybody had a hard time installing other API's or engines in Qt or even attempt to? [07:42] good morning [07:42] good morning dholbach [07:43] hi justCarakas [07:51] Good morning :-) [07:53] nik90, so I merge mzanetti's branch and pushed it [07:53] but it makes the build fail as well [07:54] dholbach: I think you pushed to the wrong branch [07:54] dholbach: balloons yesterday created a common branch that we all can push stuff into -> https://code.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/reboot-packaging/+merge/230530 [07:54] dholbach: this has jenkins passing :) [07:55] hey [07:55] mhall119: thanks for the answer [07:55] mhall119: on com.ubuntu vs com.canonical, not sure, it's not a community app really [07:55] beuno: hey [07:55] beuno: mhall119 told me you can fix my account's namespace setting for publishing click apps [07:56] nik90, oh? [07:57] dholbach: balloons did that since jenkins failed on your branch due to the find_local_path() function and other small issues in the __init__.py file. So he reverted that file alone to the trunk version and we added more fixes on top of that. [07:58] dholbach: so as of yesterday night, the debian and click packaging works correctly [07:59] nik90, I don't think the build works [08:00] dholbach: What do you mean? The last comment is by jenkins which successfully ran the tests after installing the debian package [08:00] let me retry with a fresh branch [08:02] ah no [08:02] it passes \o/ [08:02] woohoo :D [08:02] you scared me for a moment there (insert angry face here) [08:03] let me do a couple more tests [08:03] ok [08:10] dholbach: nik90: hey [08:10] I'm not sure if my branch is really the right thing to do [08:10] but this makes it compile and run on the phone again for me [08:10] cool [08:10] however, I've no idea if it breaks other stuff [08:10] I'm just checking a couple of things and let you know how it works :) [08:11] mzanetti: After doing http://bazaar.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/reboot-packaging/revision/80 it seems to run on phone for me. [08:11] it seems on the phone it was unable to find the c++ plugins that we wrote. The above fixes that [08:11] ah ok... you're disabling the whole tests/autopilot dir for click mode [08:12] which should make my patch obsolete I guess [08:12] hmm... for me it was only an issue with finding the python include dir. the rest was fine [08:13] sergiusens came to the conclusion that since we do not run the tests in the click mode might as well not include it in the click package. [08:13] +1 [08:13] I thought you guys wanted to have that [08:13] I for one would not package them up either, no [08:14] if anything, create a separate package just for the tests [08:14] yes, we have a separate autopilot debian package [08:16] so I don't know how the autopilot tests are supposed to be run [08:16] but run as part of autopkgtest, they still fail: http://paste.ubuntu.com/8034122/ [08:16] maybe autopkgtest is not the way to go, I don't know [08:16] I'll leave this for the gods of testing to decide [08:17] or maybe "adt-run -U -o test-output --source /tmp/ubuntu-clock-app_1.2.dsc --- qemu adt-utopic-amd64-cloud.img" is the wrong way to go about it [08:17] but yeah, making jenkins happy is fantastic [08:17] I guess my branch can be dropped then [08:18] yes [08:18] dholbach: can I add more past release entries into the debian changelog? [08:19] dholbach: I want the release landing on the phone to be 3.0 [08:19] nik90, sure, feel free to [08:19] with "dch -i [08:19] with "dch -i" you can add a template new entry [08:19] or just modify the current one [08:19] I will just modify the current one [08:19] so I will add past release below the current 1.2 version for instance [08:21] sure sure [08:21] good work everyone [08:21] thnx daniel [08:57] morning nik90, I've added i18n support to build the .pot file and take into account the .desktop file too to clock reboot, if you've got time to look at the MPs, then I can set up translations in LP once approved. Let me know if you have any questions. Thanks! [08:59] Good morning all; happy International Lefthanders Day! :-D [09:03] Good Morning all o/ [09:05] hi nerochiaro, around? [09:05] dpm: yes [09:05] hi [09:05] hey :) [09:06] nerochiaro, I just wanted to check, as folks start going on holiday on these dates :) - I sent you an e-mail about the translations for the .desktop files for camera, dialer and messaging not being loaded in the dash. Do you think you could have a look at it when you've got a minute? [09:07] dpm do you know if there is already a system to use translations in HTML5 apps [09:08] dpm: when i did that work back then i basically copy-pasted what had been done in other apps for translating these .desktop filtes. i'm not actually sure what the dash does to read them [09:08] justCarakas, daker was working on one, but I'm not sure about its status [09:09] nerochiaro, do the translations get added correctly to the final .desktop files? That'd be the first thing to look at [09:09] dpm: i checked that they did, but i'll double check that. i'm in the middle of something right now but i added it to the todo list [09:10] thanks nerochiaro [09:37] dpm: hey, yeah I got the notifications about the Mps by email when I woke up. [09:37] dpm: we are in the process of landing the debian and click packaging fixes in the trunk. Once that happens jenkins can autoland stuff from here on. After that branch lands, I will review your branch since there might be some code conflicts. [09:38] ok, cool [09:51] dpm: hey [09:51] dpm: what's the recommended way for a canonical team to create and upload click packages to the store? [09:52] dpm: can we all do it or does it have to be owned by one specific person? [09:52] zyga, there is no support for teams in the store, so the pragmatic way is to have a shared account where everyone who needs upload permissions has the password [09:55] dpm: thanks [09:55] dpm: is that how other teams do it? [09:55] dpm: any SOP pages that document that (I'm ok to create them if they are missing) [09:57] dpm: last question, is there any magic regarding namespaces, can anyone (e.g. me) just create apps in the com.canonical.* namespace/ [09:57] zyga, that's how other teams do it, yes. I'm not sure what SOP is [09:58] dpm: thanks :-) [09:58] dpm: Statement of Practice or something like that (on the wiki, lots of SOP pages document how we do things) [09:58] zyga, I don't think there is anything special about the com.canonical.* namespaces, but beuno would know more (he's away today, though, I think) [09:59] ok, I'll double check with beuno [10:03] zyga: we have a few teams who are uploading multiple apps, for example the core apps are one 'team' (shared account) and webapps are another, perhaps speak to dbarth to see what he did [10:04] popey: thanks! good idea [10:17] justCarakas: yes we have two https://code.launchpad.net/~daker/ubuntu-html5-theme/i18n-pojson [10:17] https://code.launchpad.net/~daker/ubuntu-html5-theme/i18n-po [10:18] the json is still WIP and po one is a working prototype that's still wait for a review [10:18] ok, I'm gonne update BE Mobile soon and atm I'm roling my own for it but it would be nice to integrate [10:26] popey: do you have access to that shared account? [10:26] popey: we'd rather not create another account and just share this one [10:33] zyga: what's to be uploaded? [10:35] popey: we want to upload under-development versions of checkbox [10:35] popey: it will be able to run simple device tests soon and the store is how we want to distribute it [10:35] ok, so what we do for most apps is we have one shared account which multiple people can upload with, including Mirv, xnox, sergiusens and balloons [10:36] probably easiest to just provide a click to one of them, and ask them to upload to the store [10:36] then the semi-automated review and publishing will take care of the rest [10:36] popey: can we upload it ourselves or do we need to go through a middle-man each time? [10:36] you could, but then it's tied to your account [10:36] I mean, could we just use the same shared account directly/ [10:36] we block on you if there's an emergency upload required [10:37] I don't know what the "rules" are on sharing that account. I don't manage it. [10:37] I don't have access to upload (on purpose) [10:37] ok, do you know who I should talk to (apart from dbarth)? xnox, etc? [10:38] to what? upload yourself? [10:38] to know "rules" and maybe get the access [10:38] yes [10:40] how many apps is this for? are they expected to be baked into the device? [10:40] or are we just talking about effectively 3rd party apps [10:40] popey: just the one [10:41] popey: not sure, maybe, maybe not, that's up in the air (baked or not) [10:41] popey: this is a system-level kind of app [10:41] popey: right now it's just a public prototype [10:41] popey: though there are some non-public derivatives that might be baked in later [10:41] popey: I don't manage that part as it's not really a tangible thing yet [10:44] popey, there are 2 clock branches blocked on data missing from geoname-lookup.u.c - have you followed with ev or someone else on how to add additional data to the service? http://people.canonical.com/~dpm/core-apps-sponsoring/ [10:51] dpm: am looking at it myself [10:53] zyga: as dpm says I think beuno is the best person to speak to, but in the short term, the quickest way to get your click in the store is to ask dpm / balloons or mirv to upload [10:54] hey Kaleo, do you know if camera-app can move to content1 instead of content0.1? [10:55] thanks popey, if you could update the clock MP with the results, that'd be great [10:55] actually I don't know what this build dependency is for, maybe tests? [10:55] will do when I figure it out [10:56] popey: thanks, I'll follow up with dpm and beuno when he's around :) [10:57] dpm, popey: Is there a person with debian packaging know how to do a quick review of https://code.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/reboot-packaging/+merge/230530. It was done by Daniel, so he can't review it :) [10:57] dpm, popey: I reviewed the cmake bits and think it is okay [10:58] Mirv, do you think you could help us with a quick packaging review? It's about adding .deb packaging support for the new clock app -> https://code.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/reboot-packaging/+merge/230530 [11:00] Laney, you mean the import, or deb deps? or what? [11:02] Kaleo: ah yes, I see the imports now, I mean both I guess [11:03] content-hub doesn't even build the old one any more [11:23] dpm: I'm looking at it, is it meant for archive upload? (like, to be used on desktop) [11:23] Mirv, it's only for Jenkins purposes, to be uploaded in the core apps PPA, so that Jenkins can run desktop tests [11:24] i.e. not for the archive [11:32] Mirv: any idea why the deb packages created (as shown in jenkins) at http://91.189.93.70:8080/job/generic-mediumtests-utopic-python3/12/? is named ubuntu-clock-app-autopilot_3.0+autopilot0_all.deb instead of ubuntu_clock_app_3.0_all.deb? [11:32] Mirv: not sure why it appends autopilot to the name when we have a separate package for that. [11:33] nik90: just fyi, I'm fiddling with the geonames stuff, and will have it up soon on a server, can we test your clock, with it pointing at my geonames server once done? [11:33] popey: yup, I need to make a tiny change to the branch when the server is ready [11:33] k [11:33] * popey goes back to battling postgres [11:34] * popey gets coffee [11:34] dpm: ping [11:35] Elleo: did you try building the gallery in a chroot last week ? [11:38] nerochiaro: oh no sorry, only natively; although I did hit the QtDbus issue there too. artmello also ran into it and was going to submit a branch I think (adding a missing build-dep instead of removing the test) [11:39] Elleo: the test has been removed in trunk already [11:39] ah, okay [11:39] Elleo: can you please try building in the chroot ? I'm having other issues, like cmake not being able to find Qt5 [11:40] nerochiaro: okay, I'll setup a chroot now [11:40] Elleo: thanks [11:42] hey nik90 [11:43] dpm: Hi, at the moment the debian/control files are point to https://code.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/trunk which is the old branch. Should we change this for the clock reboot? [11:43] nik90: not sure, possibly related to how they mangle the versions in jenkins and the fact that the package has native version (instead of what's 'normally' used ie distro version appended) [11:43] zbenjamin: do you know what might be causing this when building gallery-app in a chroot from qtcreator ? http://pastebin.ubuntu.com/8035417/ i updated the chroot from qtcreator and also used "maintain" to do "apt-get build-dep gallery-app" but it's not getting better [11:43] dpm: or are we planning to keep the reboot branch as it is? or copy it all over to the other main branch? [11:43] nik90: I just pushed some more debian/ changes in, please pull [11:43] Mirv: ah ok. sure [11:45] Mirv: can you change the description in debian/control to Ubuntu Core Clock application which provides Alarm, Clock and World Clock functionalities. [11:45] we no longer have timer and stopwatch for RTM [11:45] Mirv: and lastly, why do we have ubuntu-clock-app (3.0-0ubuntu1) UNRELEASED; urgency=medium as UNRELEASED rather than say utopic? [11:46] nik90, I'd say we update debian/control to point to the reboot branch, so it's got the correct branch, and then we have a separate discussion on how we'll swap or branch off. Does that make sense? [11:46] dpm: +1 [11:46] ok, cool [11:46] dpm: I guess the same applies then to the manifest.json files as well [11:47] right [11:47] dpm: in the email you sent me you say that for camera-app I need to remove from the .desktop file template the X-Ubuntu-Gettext-Domain=camera-app' line. However I did that, as you can see here: https://code.launchpad.net/~phablet-team/camera-app/camera-app-desktop-translation/+merge/223229 [11:47] ack [11:47] nik90: ok, changing. UNRELEASED it would be as long as it's not in ubuntu archives (I just wrote a comment to the MP), but I can revert that back since not all distro policies are really needed if you're not aiming there but the core-apps PPA [11:48] nerochiaro, I'm perhaps looking at the wrong branch, but I can still see it here: http://bazaar.launchpad.net/~phablet-team/camera-app/trunk/view/head:/camera-app.desktop.in [11:48] dpm: you're looking at trunk, the branch i submitted for review is lp:~phablet-team/camera-app/camera-app-desktop-translation and it hasn't been reviewed yet I think [11:49] dpm: jenkins was rejecting it so i merged trunk and pushed it again [11:49] nik90: pushed another commit [11:50] Mirv: ok. I will make a few other changes now (vcs-bzr needs to point at the correct bzr branch for instance). But nothing that should break the debian build. [11:50] Mirv: thnx [11:50] Mirv: can you give me 5 mins before approving the branch for debian code review? [11:50] it surely looks good enough for the PPA [11:51] nik90: yeah, no problem, and maybe dpm or someone can do the top-approve. I'm fine enough with the packaging. [11:51] ack === MacSlow is now known as MacSlow|lunch [12:01] nerochiaro, thanks, I've just added a fet comments. Do you think someone else could review it to get it landed? [12:02] dpm: i'll fix your comments, then i think you can review it and we can check with oSoMoN to get it landed [12:03] cool [12:05] dholbach: hey, ready to merge https://code.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/reboot-packaging/+merge/230530 ? [12:05] dholbach: can you top approve? [12:07] nik90, can we change 3.0ubuntu1 → 3.0? [12:07] nik90, sorry [12:07] nik90, can we change 3.0-0ubuntu1 → 3.0? [12:07] dholbach: it was done by Mirv just now [12:07] oh ok [12:07] Mirv: is that okay ^^ [12:07] Mirv, ^ is this something with daily landing I don't understand? :-) [12:09] dholbach: well we don't have non-distronative packages as native packages normally? [12:09] dholbach: it's irrelevant for PPA usage though and can be reverted [12:09] dpm: i don't understand why the file should be named .in.in.h, in your filemanager branch it is named .in.in, not .in.in.h [12:10] Mirv, I don't have a hard opinion on it - it just surprised me somewhat [12:11] dholbach, Mirv: May be we do what the camera app does? https://bazaar.launchpad.net/~phablet-team/camera-app/trunk/view/head:/debian/changelog [12:11] as far as I understand it is also pushed as a click to the store [12:12] nerochiaro, it's the name of the file that intltool-extract generates. It's .in.in.h in L20 of https://code.launchpad.net/~dpm/ubuntu-filemanager-app/desktop-i18n/+merge/230591 [12:12] top-approved [12:13] dholbach: jenkins should automerge right? [12:13] nik90: it'd be now the same for clock app if it would be released through CI Train (which adds the date) [12:13] I don't know [12:13] dholbach: alrite I will wait and see [12:14] Mirv: ah ok [12:14] excellent [12:14] so the "3.0.0-0ubuntu1) UNRELEASED" is what CI Train would expect, except that we're not doing this for CI Train now, or archives [12:14] ack [12:19] dpm: but if i change CMakeFile.txt to configure_file(${DESKTOP_FILE}.in.in.h ${DESKTOP_FILE}.in) and then run cmake it tells me that the .in.in.h file doens't exist [12:20] Elleo: news with the chroot ? [12:20] nerochiaro: still downloading packages [12:20] think it's almost done as its on to the qt packages now [12:22] dpm: i also change the po/CMakeLists.txt to create a .in.in.h but i get the same problem [12:24] dholbach: the autolanding failed :/ [12:25] hmm I see the error could not import package ubuntu_clock_app: No module named ubuntu_clock_app [12:26] nik90: TimeZoneID is the only extra field you need, right? [12:27] zbenjamin: ping [12:27] popey: one sec, let me check [12:28] popey: https://bugs.launchpad.net/ubuntu-weather-app/+bug/1326918/comments/5 [12:28] Ubuntu bug 1326918 in Ubuntu Clock App reboot "API key for geonames.org" [High,In progress] [12:28] popey: essentially admin2, admin3 and timezoneID which are also required by weather app [12:28] hm, ok [12:28] * popey fiddles with postgres queries [12:29] popey: admin2, admin3 are basically used to distuinguish cities with the same names. [12:29] so it is not essential right away [12:29] ok === _salem is now known as salem_ [12:32] nerochiaro, you don't have to configure .in.in.h to .in. the .in.in.h file is a product of running intltool-extract [12:32] dpm: oh, i see [12:34] so intltool-extract generates the .in.in.h file, which is added to the list of files to pass to xgettext [12:34] balloons: ping [12:35] nerochiaro: chroot creation failed, will try again [12:36] Elleo: failed how ? [12:37] nerochiaro: it just gave a generic error saying that the chroot creation had failed and then cleaned up (using the Qt chroot creation stuff) [12:38] Elleo: bleh, we really need some help from zbenjamin here [12:39] nerochiaro: restarted the sdk and letting it have another try; if that fails I'll try doing it using the click tools at the command line and see if that spits out any more info [12:41] dpm, thanks for review my MR. I have fixed your comments could you check it again? https://code.launchpad.net/~renatofilho/address-book-app/update-translation/+merge/230556 [12:41] Elleo: thanks [12:43] renatu, done, thanks! [12:44] popey, does the accounts button in system settings work for you? [12:44] i saw davmor2 mention aproblem with it earlier... [12:44] I've noticed since a couple of images ago that it does nothing for me [12:44] davmor2, are you having the same problem? ^ [12:44] works here [12:45] back button doesn't though, black screen [12:45] dpm, thank you [12:45] yeah, I had that too, I'd be happy to see the accounts at all, nevermind going back one screen and crashing :) [12:45] you want the moon on a stick! [12:46] :) [13:02] dpm: no it's not crashing it's just not opened the normal way so the close isn't handled quite correctly. [13:03] fginther: quick question, how does something that passes jenkins continous-integration fail autolanding? [13:03] dpm: if you got bored of waiting and swiped the accounts page closed that used to stop it from opening only way to get around that was to reboot [13:03] davmor2, no, in my case it's not opening at all [13:05] is it allowed to provide things like this https://developer.cdn.mozilla.net/media/uploads/demos/v/i/viliusl/66924b4d3bff8b1c0abecfd52e72407f/minipaint_1375196685_demo_package/index.html as a web app ? [13:05] hm, and I can't find a log file for the the system settings app [13:07] dpm: i'm still quite confused by the other MRs you point out. the filemanager does explicitly configure .in.in.h for example, if you look at http://bazaar.launchpad.net/~dpm/ubuntu-filemanager-app/desktop-i18n/view/head:/CMakeLists.txt [13:07] nerochiaro, I can only see "configure_file(${DESKTOP_FILE}.in.in ${DESKTOP_FILE}.in)" [13:07] no .h [13:09] nik90, hum, not sure [13:10] nerochiaro: qtcreator crashed partway through :/ attempting from the command line now [13:11] dholbach: I kicked it again since after the failed autolanding, it said continuous integration said it passed. Not sure what changed in that mean time. [13:11] * nik90 waits for few mins [13:11] Elleo: thanks. we really need to report all these bugs. I don't know how you manage, but I'm wasting a ton of time trying to build stuff for the device these days [13:11] dpm: try online-accounts it's a separate application [13:11] nik90, it looks like it still hasn't landed in lp:ubuntu-clock-app/reboot [13:12] nerochiaro: I tend to mostly build stuff directly on device, as its a lot less hassle for me that way [13:13] Elleo: i wish i could, it doesn't work for me. maybe you can give me an hand setting that up, if it works for you ? [13:13] Elleo: or tell me your process [13:13] dholbach: I know since the autolanding failed [13:14] nerochiaro: I basically just ssh in and treat it like any other linux system [13:14] Elleo: so for gallery you build a deb with debuild and then install it with dpkg ? [13:14] nerochiaro: no I use the click build script [13:15] good morning everyone [13:15] Elleo: what's the command you use ? [13:15] Im writing a ubuntu touch app with u1db as db, but im getting troubles with query and index. Im saving a doc with sub-fields, but at retrieve point with index and query i wont get the expected resutls. I need the list to show two rows one for the account "Ahorro" another one for the account "Corriente" but only one row show, the first one. [13:15] nerochiaro: there's a script called click-build.sh that's part of the gallery repo, that has everything in it already [13:15] nerochiaro: you'll need to install all the build-deps first though [13:15] here is the code http://pastie.org/9470062 [13:16] can anyone helpme? [13:16] nerochiaro: I haven't built gallery in a couple of months though, so can't guarantee it still works okay [13:21] nerochiaro: pong, sorry for the delay i was in my lunch break [13:22] zbenjamin: both me and Elleo are having problems building things in chroots (or in his case building the chroot itself) [13:23] Elleo: try to use click chroot -a -f ubuntu-sdk- create on the commandline and look if you can find any errors there [13:23] zbenjamin: yep, that's what I'm doing now :) [13:24] nerochiaro: whats your problem? [13:24] dholbach: nope still fails. I guess I will have to wait for balloons or fginther to come online [13:24] ok, or somebody in #ubuntu-ci-eng? [13:24] nik90, hello, what's the MP? [13:24] fginther: https://code.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/reboot-packaging/+merge/230530 [13:25] also im trying to desing an app that have relation between documents, there is a best practice for do that on on u1db? [13:25] zbenjamin: when building gallery in the chroot i get this: http://pastebin.ubuntu.com/8035417/ i tried updating the chroot and installing build-dep of gallery-app in it, but no luck [13:26] fginther: I see, [13:26] could not import package ubuntu_clock_app: No module named ubuntu_clock_app [13:26] Loading tests from: [13:26] Did not find any tests [13:26] nik90, I found the problem. [13:26] looks like an autopilot error [13:27] fginther: ? [13:27] nik90, I didn't update the testing for the autolanding job to use python3. I have that fixed and will re-approve that MP [13:27] fginther: ah :) [13:27] fginther: hope that fixes the issue [13:29] nerochiaro: ugh that does not look good, is this a new chroot? [13:29] zbenjamin: i created it some weeks ago [13:30] zbenjamin: should i try a new one ? do they bitrot that quick ? [13:32] wait [13:33] dholbach: you and I both reviewed dinko's app at the same time, you beat me by 1s and I got a server error ⍨ [13:34] \0/ [13:34] nerochiaro: its weird, can you try to delete all build files and try again? [13:35] zbenjamin: already tried [13:36] nerochiaro: the weird thing is apt-file does not find those cmake files either [13:37] nerochiaro: the weird thing is apt-file does not find those cmake files either [13:38] zbenjamin: i think they are called something different, that's the name CMake use [13:39] nerochiaro: yeah the problem is that cmake is still looking for those files [13:39] nerochiaro: or something in the click chroots changed that broke cmake, which is also possible [13:39] zbenjamin: anything i can do to help debug the problem ? [13:40] * zbenjamin updates his chroot [13:40] nerochiaro: i have not enough knowledge about how cmake searches for qt now [13:40] nerochiaro: let me try if i can reproduce [13:41] zbenjamin: i'm on 14.10 if it changes anything, but it shouldn't [13:41] dholbach, fginther: Autolanding PASSED! [13:41] fginther: thnx for the hotfix [13:42] nik90, you're welcome, sorry it was borked in the first place :-) [13:42] np [13:42] yeehaw! [13:42] dholbach: got a ton of branches to land.. [13:42] * nik90 gets busy busy [13:43] :-D [13:43] please help. I have a Java Swing project in Netbeans. I need him to create the Debian Source package. Can someone help me. thanks [13:44] nerochiaro: hm, it works for me even with a up2date chroot. [13:45] nerochiaro: you _could_ try to recreate it [13:45] nerochiaro: and pray that the problem is gone [13:46] zbenjamin: well, i'm trying with a fresh chroot, but i'm not a religious person [13:46] so no praying to technology [13:50] zbenjamin: i'm going t try this with the new chroot. if it doesn't work if you don't mind it might be a good idea to get on an hangout and talk about it by voice, will be quickler [13:50] quicker [13:50] nerochiaro: you can also pray to the FSM , makes no difference ;) [13:51] nerochiaro: we can try this, but when its still not working we probably should have a click developer there [13:52] zbenjamin: who would that be ? [13:53] nerochiaro: good question [13:53] dholbach: need your opinion on the debian versioning. So right now the latest version is 3.0-0ubuntu1. If I fix a bug and update the changelog, it seems a bit strange to release a new version just for that 1 bug fix. So do I keep appending to this existing version? And secondly, for new versions, does 3.0.1-0ubuntu1 look good? [13:54] nerochiaro: seems mvo is the one to ask [13:54] nik90, I personally would just have gone with 3.0 and then 3.1, etc - adding something like "-0ubuntu1" usually indicates that you do "upstream releases" (like tarball releases) or your piece of software [13:55] dholbach: ack [13:55] and then "-0ubuntu1" means: not been in Debian yet, one revision of this particular upstream version has been uploaded to Ubuntu [13:55] zbenjamin: i get this when trying to create the new chroot: http://pastebin.ubuntu.com/8036366/ [13:55] for a small fix like a dependency change you'd then go with 0ubuntu2 [13:55] Elleo: is it the same problem you were having ? http://pastebin.ubuntu.com/8036366/ [13:55] but maybe there's a good reason in the autolanding scheme of things to use -0ubuntu1 [13:56] nerochiaro: check with mount if you old chroot is still mounted [13:56] dholbach: ok. I am going to add 3.1 and then keep appending bug fixes to it before I consider 3.2 worthy. [13:57] mzanetti, would you mind if we move today's Reminders meeting to tomorrow same time? [13:57] nerochiaro: if it is try schroot --all-sessions -e [13:57] nik90, maybe have a chat with Mirv as well to confirm [13:58] nerochiaro: possibly, I didn't get as much output as that only the "There was an error creating the click target, cleaning up" line onwards in qtcreator [13:58] dholbach: ok [13:58] zbenjamin: i asked qtcreator to delete it. that should've unmounted it no ? [13:58] zbenjamin: to delete the old one [13:58] Mirv: hey, when you are free, can you ping me. Just a really low priority question. So no hurry :) [13:59] dpm: works for me [13:59] please help. I have a Java Swing project in Netbeans. I need him to create the Debian Source package. Can someone help me. thanks [14:00] mzanetti, thanks. Actually, it'd be half an hour earlier, if that still works [14:00] dpm: sure [14:00] excellent [14:00] nerochiaro: its chroot, sometimes it fails [14:01] dpm: can you merge trunk to your branches. we just landed a mega branch into trunk :) [14:01] zbenjamin: interestingly enough when i try to build on the device (manually, not via qtcreator) i have the same problem [14:01] zbenjamin: of qt5 libs not being found [14:02] nerochiaro: the dev packages are not installed on the device [14:03] zbenjamin: i installed build-dep of gallery-app, that should've brought them in [14:04] nerochiaro: true [14:04] zbenjamin: wait, trying again now installs some more packages [14:04] nerochiaro: wtf [14:05] zbenjamin: yeah, wtf... but it's building now, on device, let's see what happens. in the meantime, i'm trying these chroots [14:06] zbenjamin: this is the result of the commmand you suggested before: http://pastebin.ubuntu.com/8036416/ [14:08] nik90, sure. In about 2 hours I should be able to [14:08] dpm: ok [14:09] nerochiaro: as i thought, schroot did not unmount correctly [14:10] nerochiaro: check if the /var/lib/schroot/chroots/click-ubuntu-sdk-14.10-armhf is deleted [14:10] nerochiaro: if its gone, restart your command [14:13] zbenjamin: the dir is not there but if i try to create a new chroot from qtcreator it gives me the error i pastebin'd you before [14:14] nerochiaro: click chroot -a armhf -f ubuntu-sdk-14.10 destroy [14:14] nik90: http://146.185.188.87/?query=Farnborough now has timezone.. any good? [14:14] popey: oh yeah that's awesome [14:14] popey: that should unblock my branches [14:15] zbenjamin: "A chroot for that name and architecture does not exist." [14:15] zbenjamin: if i try "create" instead of destroy i get that error [14:15] popey: so should I point the clock app to your server link or are you merging this into geonames lookup [14:16] zbenjamin: "that error" i mean the one i put on pastebin [14:17] zbenjamin: so it seems like a problem with debootstrap, who's responsible for that ? [14:18] nik90: I'll prepare a patch, no guarantees when it will land. [14:18] nerochiaro: Cannot install into target '/var/lib/schroot/chroots/click-ubuntu-sdk-14.10-armhf' mounted with noexec or nodev thats the error and it tells you that its still mounted [14:18] popey: ok [14:18] nik90: the box that is running on is a vps I'm personally pay for, not canonical infra.. [14:18] nerochiaro: you can try to reboot or force unmount them [14:19] nik90: I'll speak to the guys in our community call and see what we can do to get it landed, but feel free to test on that IP for now [14:19] zbenjamin: well the real cause seems "/usr/sbin/debootstrap: 1290: /usr/sbin/debootstrap: cannot create /var/lib/schroot/chroots/click-ubuntu-sdk-14.10-armhf/test-dev-null: Permission denied [14:19] " [14:19] popey: ok. I can test my branch with your server but cannot land it before your patch lands. [14:20] zbenjamin: shit, /var/lib/schroot does not actually exist [14:20] nerochiaro: i would try a reboot [14:20] zbenjamin: i created that dir instead [14:20] nerochiaro: wut [14:25] nik90: ok [14:26] zbenjamin: ok, seems to be finally building the chroot [14:26] nerochiaro: how did you fix it? [14:30] zbenjamin: created the missing dirs [14:31] zbenjamin: not sure why they were not there [14:32] dholbach: myapps completely gone for you? [14:32] The requested URL / was not found on this server. [14:32] woah, its back [14:32] yep, wfm [14:32] nerochiaro: good that its working now [14:32] maybe a deployment in progress? [14:32] zbenjamin: well, it's creating the chroot [14:32] zbenjamin: we'll see if it can build in there [14:33] nerochiaro: true ;) [14:33] zbenjamin: also i built the click on the device. i get this when i try to install it: [14:33] ** (process:24247): CRITICAL **: ubuntu_app_launch_get_primary_pid: assertion 'con != NULL' failed [14:33] ** Message: database.vala:342: Removing /opt/click.ubuntu.com/com.ubuntu.gallery/2.9.1.1034 [14:33] zbenjamin: any idea who can i ask to ? [14:34] dholbach: i always seem to catch the deployments! [14:35] nerochiaro: you can try to verify the packet you created on the device , with the click review tools [14:35] they often tell you when something is wrong [14:35] popey: my branch works perfectly with your server :) cities are returned fast with the correct times [14:36] awesome! [14:39] nik90: should the field name be timezone or timezoneid? [14:40] nerochiaro: with the content-hub integration one little extra thing that might not be obvious, when the content type is set to unknown you'll need to allow picking both videos and pictures (this is the case when the user hasn't requested a specific content-type, so all are valid) [14:40] the column in the db is timezoneid, so probably should be that? [14:40] nerochiaro: haven't seen your branch yet so you might already be doing that, but thought I should mention it since it's not an immediately obvious usecase [14:40] popey: I am fine with either since it is just a simple change on my end. I think that timezoneID fits better since it is indeed the timezoneID of that location. [14:40] (just preparing the merge but don't want to have to re-do it) [14:40] its all lower case, so will be timezoneid [14:40] thanks [14:40] ok [14:41] np [14:41] Elleo: i accounted for that, the only thing i haven't tested yet is what is being passed to me when the content typei is not specified. an empty string ? [14:41] popey: ping me the MP links when you are done so that I can include that in MP for easy tracking [14:42] nerochiaro: ah, you're using the C++ API in gallery, so yeah, it might be an empty string; might want to check with kenvandine [14:42] nerochiaro: for QML I think it gets an explicit ContentType.Unknown value set, but I'm not sure about the C++ API [14:43] ugh [14:43] right [14:43] so that only gets set to a valid value when the transfer is created in the qml bindings [14:43] nik90: https://code.launchpad.net/~popey/ubuntu-geonames/fix-1326918/+merge/230650 [14:44] kenvandine: should i just go ahead and spend some extra time making it all work in pure qml ? [14:44] kenvandine: it seems such a pain in the neck to do it in c++ [14:44] nerochiaro, that would be ideal... [14:44] kenvandine: but what do i get now in c++ when no type is specified ? empty string ? that will work for me [14:44] we really don't want to support the c++ api... that's not part of the platform definition [14:45] you should get "unknown" [14:45] i think [14:45] ok, i'll debug it i guess [14:47] kenvandine: this is unrelated to the above, but i'm trying to install build-dep of gallery-app in an armhf chroot and i get "libcontent-hub-dev : Depends: libcontent-hub0 (= 0.0+14.10.20140806.2-0ubuntu1) but it is not going to be installed" [14:47] kenvandine: any idea what's amiss there ? [14:48] not off hand... what if you try to install both? [14:49] does it resolve the dep right? [14:49] * kenvandine wonders if something libcontent-hub0 deps on is broken [14:49] ah, the bindings i bet [14:49] if you do a dist-upgrade it'll fix it [14:50] or include qtdeclarative5-ubuntu-content1 in your install [14:52] nerochiaro: you sure it installs the armhf version? [14:53] nerochiaro: try apt-get install libcontent-hub-dev:armhf [14:54] nik90, did you talk with timp about the PageHeader state "" [14:57] zbenjamin: i'm just doing "apt-get build-dep gallery-app" in the chroot [14:57] nerochiaro: if that does not pull the right arch [14:57] you are doomed ;) [14:58] zbenjamin: well, looks like the case, or some dependency mix up. but i need to get to the bottom of it [14:58] kenvandine: ^ [14:58] renatu: oh no not yet..I got busy with some other stuff. I will ask him now. [14:59] nerochiaro, i suspect it's because of the package rename for qtdeclarative5-ubuntu-content1 [14:59] not sure if build-dep will do the right thing there [14:59] t1mp: Hey, while using PageHeadState in the clock app, it was not possible to set the PageHeadState name to "" instead of "default". Is there any particular why this is the case? [14:59] does your chroot had qtdeclarative5-content1 or qtdeclarative5-ubuntu-content0.1 ? [15:00] nerochiaro, which depends on libcontent-hub0 [15:00] kenvandine: it has -ubuntu-content0.1 [15:00] ok [15:01] install qtdeclarative5-ubuntu-content1 [15:01] that'll remove the old one [15:01] it's the quickest solution [15:01] a dist-upgrade would work too [15:01] t1mp: if I set the PageHeadState name to "", and the assign the default page to "", it does not load the header action buttons that were defined in the "" pageheadstate. [15:02] kenvandine: dist upgrade doesn't do anything. the other solution you just suggested is now removeing a lot of packages and installing others [15:02] kenvandine: let's see if it works [15:02] ugh [15:02] renatu: oh btw I overcame that limitation (or bug) by adding when: !alarmListView.isInSelectionMode in the default pageheadstate :) [15:02] dist-upgrade should... [15:03] renatu: this way they switch back to the correct state based on if the selectionMode is active or not [15:03] kenvandine: it just says nothing to do [15:03] kenvandine: it's a fresh chroot so it's already as up to date as possbile [15:03] kenvandine: ok, the other solution worked and i can instal build-dep of gallery-app now [15:03] nik90, could you report a bug about that? I am reporting another bug that I facing with PageHeaderState [15:03] cool [15:03] nik90, with a example [15:04] renatu: sure, creating one now [15:05] but zbenjamin now i have the fresh chroot, i installed the build-dep gallery-app successfully. removed all kits from the project except the armhf one. trying to configure the project still fails as it can't find the qt5 libraries it seems [15:06] bzoltan: we probably broken chroots ^^^ [15:06] +have [15:13] nik90, this is another bug that I found: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1356420 [15:13] Ubuntu bug 1356420 in Ubuntu UI Toolkit "[PageHeaderState] Does not work in a state with other PropertyChanges" [Undecided,New] [15:14] renatu: yeah I hit that bug as well, So I just did contents: Label {} where I set the label [15:14] kkk [15:15] nerochiaro, do you have any idea how much work it would be to refactor gallery-app to use the qml bindings? [15:15] renatu: btw in the MultiSelectionView.qml file, the functions startSelection() and other, what state do they change? The PageHeadState or the listview state? [15:15] nik90, listview [15:15] renatu: I don't see any reference to which state they are changing except the assignments like state = "selected" [15:16] hmm [15:17] renatu: ok, I better understand the bug. [15:18] kenvandine: probably not more than a day [15:19] boiko: for when you have some time, this is ready for a review: https://code.launchpad.net/~canonical-platform-qa/dialer-app/fake_url-dispatcher/+merge/230403 [15:20] we've been bit a couple times by gallery already, but i guess it's up to bfiller :) [15:20] nerochiaro: could you give us a branch what we can try out with zbenjamin? [15:22] bzoltan: lp:gallery-app === chihchun is now known as chihchun_afk [15:42] popey: do you mind top approving https://code.launchpad.net/~nik90/ubuntu-clock-app/fix-alarm-label-behavior/+merge/230387. it is ready to go [15:43] nik90: will do in 15, on a hangout right now [15:43] popey: np [15:43] popey: what is the oposit of a hangout :p [15:48] renatu: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1356434 [15:48] Ubuntu bug 1356434 in Ubuntu UI Toolkit "[PageHeadState] Cannot create pageheadstate with name """ [Undecided,New] [15:50] nerochiaro: for me it finds the qt stuff but fails at the end when linking to expat [15:51] nerochiaro: i sadly have to run now, its my dads birthday [16:16] nerochiaro: I have installed the deps to my chroot and cmake went fine... now compiling [16:18] bzoltan1: how did you install the deps ? [16:26] dpm, hey, I see the email for tomorrow meeting... Thanks, it definitely works for me :-) [16:28] nerochiaro: apt-get install [16:30] bzoltan1: install whta ? [16:32] nerochiaro: libmediainfo-dev libkexiv2-dev libcontent-hub-dev [16:32] nerochiaro: I made it on an i386 target ... let me do for armhf too [16:35] nik90: yeah I noticed a problem with PageHeadState.. it seems like all properties of the standard State get broken :( [16:35] I'll need to find a solution for that, but there is a lot of other stuff to do first :( [16:35] nik90: I gotta go now, let's discuss it tomorrow [16:39] nerochiaro: the libmediainfo-dev:armhf was trivial ... other other too are not [16:40] nerochiaro: I am afraid one of the reason for your broken chroot was that you have installed libcontent-hub-dev:armhf and that has removed some essential libs [16:41] nerochiaro: zbenjamin: for questions like this :) http://pastebin.ubuntu.com/8037566/ my answer is _NO_ [16:43] rpadovani, awesome, thanks! [16:45] nerochiaro: so in short, the gallery app is not supported by click chroots ... good point to file a big against the click package [16:51] bzoltan: hi, had any luck with the failing builds in the silo? [16:53] popey, balloons: If either of you are free, can you top approve https://code.launchpad.net/~nik90/ubuntu-clock-app/update-innerclock-texture/+merge/230669 and https://code.launchpad.net/~nik90/ubuntu-clock-app/sync-upstream-fastscroll/+merge/230666. Both these MPs require no code review since one replaces an image and the other syncs up with upstream code [16:53] which is already tested. [16:54] I win! [16:54] balloons: hi , Good morning [16:54] lol [16:54] :-) [16:54] morning mihir [16:55] balloons: i did write an AP for editing an event and jenkins failed it [16:55] https://code.launchpad.net/~mihirsoni/ubuntu-calendar-app/1188819/+merge/230554 [16:55] balloons: are these errors regarding pep8 formatting issues? [16:56] nerochiaro: zbenjamin: https://bugs.launchpad.net/ubuntu/+source/click/+bug/1356481 [16:56] Ubuntu bug 1356481 in click (Ubuntu) "Can not install build deps for gallery-app in an armhf click chroot" [Undecided,New] [16:57] mihir, yep.. you can use autopep8 to fix it if you wish, or do the edits yourself [16:57] bzoltan: i actually recreated the chroot from scratch then installed the armhf deps explicitly as Kaleo suggested and it worked [16:57] python-pep8 [16:57] balloons: sure i'll do that [16:57] then run pep8 name_of_fille.py [16:57] nerochiaro: where can I see Kaleo's suggestion? [16:58] balloons: if you have got my previous messages, i gave one link to paste.ubuntu , which contains errors while adding guest. [16:58] bzoltan: apt-get install libexpat1-dev:armhf libcontent-hub-dev:armhf libexiv2-dev:armhf libmediainfo-dev:armhf [16:59] balloons: actually i have lost the link :| and on web IRC client. [16:59] nerochiaro: that command removes half of the APIs from the chroot [16:59] mihir, not sure I have anything else from you.. [17:00] bzoltan: didn't do that in mine [17:00] bzoltan: Kaleo ? ^ [17:00] nerochiaro: OK.. let me create a fresh chroot [17:02] nerochiaro: anyhow.. if the gallery app is meant to be built with our SDK and these libs are part of our public API set then these packages should be present on the click chroot by default [17:03] bzoltan: i've been told the dependencies of an app need to be installed manually isn't what we're doing here ? [17:03] nerochiaro: who told you that? [17:04] bzoltan: benjamin [17:04] nerochiaro: I mean clearly, you do not have other choice [17:04] bzoltan: i think [17:05] nerochiaro: but optimally the builder sysroot should be static. Because if the developers start to add there APIs then the risk will grow that they build and package apps what will not work on the device. [17:06] balloons: balloons: for somehow it is not letting me add guest while creating new event , giving following log http://paste.ubuntu.com/8030498/ [17:06] balloons: ^^^ [17:07] bzoltan: well, so if apps want to use a certian library that's not in the standard image they have to statically link to it ? [17:12] nerochiaro: what else can they do? [17:12] nerochiaro: or package the lib side by to the app [17:12] bzoltan: well i guess it's a limitation/design choice of click === gatox is now known as gatox_lunch [17:13] nerochiaro: it is the same on every mobile platform I know... Sailfish, Maemo, Meego ... [17:14] bzoltan: fair enough. so i guess i should change gallery app at some point to statically link all that stuff [17:14] nerochiaro: if something is not available build time than that library will not be available at runtime either... [17:14] nerochiaro: or we should include those libs to the image and th sdk [17:14] bzoltan: whatever works best. libcontent surely should be in the SDK no ? [17:15] nerochiaro: I would think so [17:15] bzoltan: who would be responsible for making it so ? [17:16] kenvandine: is there any quick way to have an app request specifically video content from the content hub ? [17:16] kenvandine: to test that gallery honors the content type request [17:17] nerochiaro: I am afraid it is me :) at least from the SDK side.. if those libs are expected to be used by other apps then we should consider to include them to our API offering. [17:18] bzoltan: i think only the content hub but i guess it's not up to me to decide [17:18] bzoltan: as i don't know if other apps use exiv etc [17:38] mzanetti: hey you got a minute? Got a question about QTest. [17:39] nik90: shoot [17:40] mzanetti: say I want to test if the back button is disabled when the alarm label is empty. Do I do the usual mouseClick(), type string and check back button status? Or do I just say alarmLabel.text = "value", now check button status? [17:40] mzanetti: oh I just wrote my first QTest, it is nice to write in qml :D [17:41] nik90, ok, https://code.launchpad.net/~dpm/ubuntu-clock-app/reboot-i18n/+merge/230578 merged with trunk [17:41] nik90: yeah... I'd do something like this: [17:42] compare(backButton.enabled, false) [17:42] dpm: thnx :) will check it out [17:42] mouseClick, type etc [17:42] compare(backButton.enabled, true) [17:42] mihir, the guest must be created by the test [17:42] There's code in trunk that should be taking care of it [17:42] mzanetti: but why do the mouseClick and type when you can just set the text value? [17:43] nerochiaro, specifying the type is only used to find the peers that support it [17:43] balloons: hmmm , i just ran existing test, and it failed while entering guest name UX user. [17:43] nik90: well, yes, you can also do that... depends a bit on what exactly you want to test [17:43] balloons, we've landed a fix to the ap tests so could u rerun jenkins for https://code.launchpad.net/~vrruiz/music-app/i18n-strings/+merge/230512 [17:43] balloons, and when u have a free period of time could u ping me so we can discuss the ap refactor \o/ [17:43] mzanetti: ah ok. If it is purely a unit test, then I suppose i could just set text. If I am looking to test user interaction then I do the mouseClick and type stuff. [17:43] ahayzen, victor top approved that so it should be good [17:43] balloons, damn he is fast [17:43] mihir, trunk? let me try [17:44] balloons: i just got latest trunk, am I missing anything on utopic desktop for contacts ? [17:44] kenvandine: sorry, i thought i was supposed to receive the content type in the transfer request as well once my app was selected [17:44] kenvandine: in case my app supports more than one content type (e.g. gallery) [17:44] balloons: i just wrote AP , and formated with pep8 could you just give me your inputs , https://code.launchpad.net/~mihirsoni/ubuntu-calendar-app/1188819/+merge/230554 [17:44] balloons: finally i wrote an AP :D [17:45] mihir, adding a guest works fine here from trunk. tests run properly [17:47] * balloons branches lp:~mihirsoni/ubuntu-calendar-app/1188819 [17:48] nerochiaro, the content type is in the transfer object [17:48] now... just added that [17:48] but it only includes it for transfers created by an app using the QML bindings [17:49] kenvandine: i receive it and it is currently "unknown" [17:49] kenvandine: is that what it is always supposed to be in c++ for now ? [17:49] yeah [17:50] i didn't want to clutter up the c++ API right now, since we aren't really supporting that [17:52] mihir, tried with your branch as well.. tests run fine, adds user fine [17:53] balloons: okay , do i need to create that Guest manually on desktop ? [17:53] balloons: btw, the new test case runs fine ? [17:53] nerochiaro, actually... which app is creating the transfer? [17:53] kenvandine: so why did you even add the property to the c++ API if it's always going to be "unknown" ? [17:53] it should have a valid content type there if it was created by anything other than gallery [17:53] kenvandine: i am testing by creating the transfer from browser [17:53] kenvandine: what other app can i try ? [17:54] oh... is it using ContentType.All maybe? [17:54] the property needs to be there to get it sent over dbus, etc [17:54] kenvandine: sorry, i'm not following you. can you hop quickly on mumble for a minute ? [17:54] sure [17:55] mihir, nope the tests should just work. post a full log where it doesn't. Also check and ensure you have everything installed you need to. my guess is you are missing depends [17:56] mihir, you rneed address-book-service-dummy installed [17:56] https://code.launchpad.net/~ken-vandine/+junk/hub-importer [17:56] nerochiaro, ^^ [18:00] balloons: ping [18:00] kenvandine: when i try to run it from qtcreator on the device it just exits immediately, nothing is displayed. is that expected ? [18:00] m-b-o, pong [18:00] balloons: something seems not right with install instructions on https://wiki.ubuntu.com/Touch/Testing/Autopilot [18:00] nerochiaro, nope... [18:01] nerochiaro, i just build a click for it [18:01] balloons: tried ppa:autopilot/1.5 too, but apt-get update gets me a 404 [18:01] nerochiaro, the content-hub relies on ubuntu-app-launch for switching between the apps, etc [18:01] and it doesn't behave well with apps that aren't installed as click [18:02] kenvandine: i think that's what qtcreator does, installs it as click then launches it [18:02] mzanetti: do you happen to know if it is possible to click on the clear button in a text field? I cannot define its objectName in my qml code since it is a sdk component. So curious how one goes about that. [18:02] balloons: awesome , that was the cause :D [18:02] kenvandine: Installed com.ubuntu.developer.ken-vandine.hub-importer-0.2.all (installed:click,removable=1,app_name=hub-importer) [18:02] balloons: thanks a lot :D [18:02] ok, try launching it from the shell then [18:02] balloons: root cause* [18:02] mihir, :-) awesome. first glance at your test looks good [18:03] balloons: thanks for your review time and thanks a lot for your help to setting this up :) , i'll try to add pending AP listed in Bugs :) [18:04] nik90: well... you could use findChild to find the button inside... but I guess in this case its probably easier to click the label at the right [18:04] nerochiaro, oh.. it might have the wrong policy version specified [18:04] that branch is pretty old [18:04] m-b-o, looking [18:04] or wrong framework [18:04] kenvandine: not sure how to fix the policy [18:04] nerochiaro, fix the framework in the manifest [18:04] mzanetti: hmm..I am tryingn to clear the text inside the textfield [18:04] nik90: mouseClick(textField, textField.width - units.gu(2), textfield.height / 2) [18:04] nik90: assuming the clear button is at the right hand side in the textfield [18:05] m-b-o, ppa:autopilot/1.5 is correct [18:05] ubuntu-sdk-14.10-qml-dev3 [18:05] nerochiaro, ^^ [18:05] it's only published for trusty tho [18:05] mzanetti: hmm yeah that will work [18:05] balloons: yes, but apt-get update gets me a 404 for the Packages [18:06] m-b-o, you are on trusty yes? [18:06] balloons: nope utopic in vm [18:06] kenvandine: still fails. the json file says "policy_version": 1, should i change that ? [18:06] popey: you there ? [18:06] m-b-o, lol.. then you don't need a ppa. just install python3-autopilot and python3-autopilot-vis [18:06] kenvandine: also how would you run it from the console ? [18:06] mihir: yo [18:06] I'll update the instructions to clarify [18:06] m-b-o, ^^ [18:07] popey: did you get time for getting some idea about provider ID ? [18:07] balloons: hehe [18:07] nerochiaro, use ubuntu-app-launch on the console [18:07] and look at the logs in ~/.cache/upstart [18:07] kenvandine: i mean, what id should i pass to ubuntu-app-launch, i always get that wrong [18:08] com.ubuntu.developer.ken-vandine.hub-importer_hub-importer_0.1 [18:08] when in doubt, look in the .desktop file in ~/.local/share/applications [18:08] the APP_ID in put in there [18:09] kenvandine: ah, that's interesting, it's not there. i guess qtcreator uninstalls it after finishing to run it. i'll try building the click manually and installing it [18:09] renatu: ping !! [18:11] mihir, so I'll try to run your test now. it works for you? [18:11] balloons: yeah it works for me :) [18:11] it create new event , edit that event and assert on details view. === gatox_lunch is now known as gatox [18:11] balloons: let me know if I have done wrongly [18:11] mihir: it should work with "google".. [18:12] mihir: have you tested in the emulator? [18:12] popey: okay thanks. [18:12] popey: nope , but balloons did tried on device but it didn't work. [18:12] ☹ [18:13] any apparmor denials? [18:13] balloons: ? [18:13] popey: on desktop , if I write evernote it does open window to add account , but when i write google , it doesn't do anything :( [18:13] mihir, looks awesome. My only thought is that it should probably change all the information. it seems to use the same uuid [18:13] well thats odd [18:13] balloons: while editing an event , ? [18:13] balloons: nope i took another UUID for editing event, if you observe :| [18:13] mihir, yes when you edit, change the details to something different [18:14] mihir, I say that.. but it's weird as the uuid id was the same for me [18:14] kenvandine: the id from the desktop is com.ubuntu.developer.ken-vandine.hub-importer_hub-importer_0.2 and when i launch it i get no feedback at all, and there's no log for it in the logs dir [18:14] mihir, yes I saw test_event = data.Event.make_unique() [18:14] balloons: hmmm , for me all the cases it was diff :| heheh weried [18:14] popey: i'll try to digout this more :) [18:14] hehe.. ok, so yea. I'll try again.. maybe I was crazy [18:14] nerochiaro, weird... do you see it in the app scope? [18:15] mihir, so +1 overall from me [18:15] balloons: awesome , i'll try to finish pending ones :) [18:15] if anything is wrong, ubuntu-app-launch does blindly not tell you :) [18:15] popey: what do you think for this bug, https://bugs.launchpad.net/ubuntu-calendar-app/+bug/1349531 [18:15] Ubuntu bug 1349531 in Ubuntu Calendar App "Delete future recurring events missing" [Wishlist,Confirmed] [18:15] kenvandine: no, doesn't seem to be there [18:15] kenvandine: in the scope [18:15] it must not have properly installed [18:15] you changed the framework right? [18:16] nerochiaro, change policy_version to 1.2 in hub-importer.json [18:16] kenvandine: yes, then i did "click build .", adb pushed the click, then used pkcon install-local to install it [18:16] i bet it failed to generate the click profile for it because of that policy_version [18:17] i need to update these +junk branches :) [18:17] balloons: one question , did we got helper for datetime picker ? [18:18] kenvandine: ok, launches now [18:19] balloons: why did Jenkis failed again :( [18:20] kenvandine: ok, so now if i click on "import pictures" the gallery is not part of the apps that get proposed [18:21] right, you changed to to look for ContentType.Videos right? [18:21] mihir, there's video for jenkins tests.. really handy. I'm watching now to see what happened [18:21] http://91.189.93.70:8080/job/generic-mediumtests-utopic/1512/artifact/calendar_app.tests.test_new_event.NewEventTestCase.test_add_new_event_with_default_values%20%28with%20mouse%29.ogv [18:21] kenvandine: no, not yet. i wanted to test with pictures first [18:21] oh [18:21] what is listed? [18:22] mihir, ohh one thing.. you should add @autopilot.logging.log_action(logger.debug) above def edit(self): in emulators.py [18:22] kenvandine: camera and file manager [18:22] nerochiaro, there's something wrong with your gallery click install [18:23] kenvandine: ok, what should i be looking for ? [18:23] do you have your own build of gallery installed? [18:25] kenvandine: i know what the problem was, i didn't have it properly installed [18:25] kenvandine: galler [18:25] kenvandine: now i fixed that and it does appear [18:25] great [18:26] mihir, ohh you might need to use self._workaround_bug_1350605() as well [18:26] mm.. it's called in _add_event [18:27] balloons: it has to be called in _edit_event ? [18:27] balloons: i did removed it because after we edit we go back to details page and asserting the values right ? [18:27] mihir, I may have spoke too soon [18:28] mihir, mm.. [18:28] mihir, so I see one small bit of logic that might be the cause. I'll comment in the mp [18:29] balloons: hmm okay. [18:35] balloons: one more que, do we have AP helper for datetime picker? [18:35] mihir, yes we do. it's in trunk for calendar [18:35] balloons: okay :) [18:36] mihir, http://pad.lv/1328600 is the issue with using it [18:36] Launchpad bug 1328600 in Autopilot "Autopilot lacks support for large timestamps" [High,In progress] [18:36] balloons: thanks for help :) [18:37] balloons: sorry i lost , this is the cause of failing Jenkins on my NEW AP ? [18:38] mihir, nope, it's not an issue for your mp [18:39] balloons: okay got it :) [18:41] balloons: i'll be off to bed, i'll try to look at that MR tomorrow. let me know your inputs on MR. [18:42] mihir, yep.. almost ready to push. it'll be a branch for you to merge [18:42] just some tweaks [18:43] balloons: Hmm okay. [18:43] balloons: okay waiting , so that we can land this today itself. [18:44] * balloons runs changes [18:50] k, I think it might be good [18:51] balloons: you pushing in saperate MR ? [18:52] m-b-o: nik90 do you have examples of cities which have an admin2/admin3? [18:52] mihir, I reproduced the issue. The trouble is timing [18:53] mihir, no just a branch for you to merge [18:53] hmm okay. [18:54] mihir, anyways, you can see what I did here: lp:~nskaggs/ubuntu-calendar-app/mihir-edit-tweaks [18:54] it's not exactly ready, but the original issue I found so I thought was in _edit_event [18:55] mihir, it can merge tomorrow :-) [18:55] popey: let me check [18:56] balloons: Hmm i see, but i couldn't get why did you add cleanup ? [18:57] nik90: geonames=# SELECT * FROM geoname WHERE admin2 != "" [18:57] geonames-# SELECT * FROM geoname WHERE admin3 != "" [18:57] geonames-# [18:58] mihir, the cleanup step will delete the test after we are done [18:58] mihir, did I duplicate it? [18:59] balloons: i see my mistake, i called add_event , instead of edit in edit_event , is that correct? [18:59] popey: http://api.geonames.org/search?q=london&maxRows=2&username=krnekhelesh&style=full&featureClass=P&style=full&featureClass=P [19:00] mihir, something about that method is still a bit off I think yes [19:01] balloons: i am trying understand my mistake and why did it scuccessfully ran on machine [19:02] mihir, yes I broke it, but I'm hoping to make a bit more sense [19:02] ;-) [19:02] balloons: Thanks for breaking , it made me to become perfect :D [19:02] popey: for clock we use only adminName and adminName2 [19:02] popey: never used adminName3, not sure why weather app needs it [19:03] popey nik90: http://bazaar.launchpad.net/~ubuntu-weather-dev/ubuntu-weather-app/trunk/view/head:/components/CitiesList.js [19:03] that's the data used for the defined list of cities [19:04] I'm trying to figure out with mhall119 why my crappy mod isn't showing admin2 [19:05] m-b-o: ok, but isn't adminName and adminName2 sufficient enough to distinguish cities with same name? [19:05] m-b-o: In the list you provides, I am not seeing a benefit of adminName3 [19:05] balloons: pushed :) [19:06] elopio: late reply, sure, I will take it a look later today [19:06] nik90: yes, the list is a bad example [19:06] nik90: I can't actually rememebr why we used admin3 over admin2 [19:07] m-b-o: ah ok [19:07] boiko: thanks. And I have a branch that introduces a qml suite. It works on my machine, but segfaults on Jenkins. [19:07] whenever you have some time please ping me to see if you can make sense of the error log. [19:07] balloons: have good night :) [19:08] elopio: ok, mind if we check that tomorrow? [19:08] nik90: I remember, admin3 was county, admin2 the next bigger area. It makes at least in Germany more sense [19:08] mihir, night.. I pushed some more changes, but it still fails locally, heh [19:08] boiko: yes, no problem with that one. It's not urgent. Whenever you can make some space on your agenda. [19:08] it's the same as jenkins [19:08] balloons: Hmmm [19:09] balloons: ahh, you just ran normally and it broke , or some tweaks to break like jenkins? [19:10] mihir, I restructed it to try and fix things. [19:10] lol. I think _edit_event makes more sense now [19:11] well.. actually no [19:11] heh.. but mihir you should get some sleep [19:11] balloons: hmmm let me re-run on my VM [19:14] more code cutting done, rev 39 [19:14] 397 [19:18] mzanetti: hey I just finished my test at https://code.launchpad.net/~nik90/ubuntu-clock-app/add-first-qmltest/+merge/230702. Can you check if I did it right? It passes for me locally. [19:18] balloons: okay trying to run on my machine [19:18] mzanetti: make test doesn't work. I think the location is incorrect [19:22] Hi, I have been searching for a library for X11 to Hexidecimal (or even RGB) colors in C++ anyone have any suggestions? [19:23] balloons: i see the error, it is not going into edit mode now : [19:24] mihir, yes ofc.. it's not finding the event.. yet it can find it a second later to delete [19:24] but balloons it won't , the action should be after event Details , it should go to Edit page , right? [19:24] nik90: I think you're gonna owe popey a beer or two, after looking at this geonames code [19:25] haha [19:25] mhall119: :D [19:25] popey: I will mail you the beer [19:26] balloons: on event Details page, instead of going to Delete action , it should go and click edit this event right? [19:26] mihir, pushed rev 398 now. [19:26] mihir, it's running the cleanup step [19:26] which uses the exact same method to discover the event [19:27] however, when it goes to click the event for edit_event, it fails to find the event name, as the error says [19:28] mihir, def get_event(self, event_name, visible=True): is where the error comes from [19:29] Hi, I have been searching for a library *TO CONVERT* X11 to Hexidecimal (or even RGB) colors in C++ anyone have any suggestions? === mardy_ is now known as mardy [19:31] balloons: hmm but again it is failing ,did it pass on your machine? [19:31] mhall119, popey: Hope this is not late :/ but once we get AGPS working, it will return the user's lat and long. I will need to query the same server to query the cityname using its lat and long. Do you know if http://geoname-lookup.ubuntu.com/ supports that? [19:32] with geoname I do http://api.geonames.org/timezoneJSON?lat=11.58&lng=43.14&username=krnekhelesh [19:32] balloons: why are we cleaning before editing ? [19:32] It seems that this should exist, as there is /etc/X11/rgb.txt installed... just not sure where to begin looking [19:32] mihir, I think I see the issue [19:32] popey: fyi http://imgur.com/i8pLiiB [19:32] mihir, the cleanup step runs last. it matters not when you add it [19:33] mhall119, popey: actually http://api.geonames.org/findNearbyJSON?lat=47.3&lng=9&username=krnekhelesh&style=full [19:33] balloons: cleanup delete the created event , have I understood correctly? [19:33] nik90: it doesnt [19:34] m-b-o: ooh! [19:34] mihir, we aren't passing the original test_event name to edit_event [19:34] that's the issue [19:36] * balloons testing [19:37] works :-) [19:38] you made change in _edit_event right [19:38] ? [19:39] I made lots of changes, heh [19:39] balloons: ahahah :| [19:39] but no, the final was in test_edit_event_must_change_it_from_day_view [19:39] did you push ? [19:40] not yet [19:42] mihir, pushed 399, but the asserts need fixed now [19:43] balloons: asserts should be either on Details page or Day view [19:44] balloons: i'll try to play with it tomorrow. [19:45] mihir, rev 400 closer.. hehe [19:45] balloons: thanks a lot for your help :) [19:46] no one knows? [19:47] israel_, #ubuntu-devel might be a better place for your question [19:48] nik90: do you really need the MainView there? [19:49] mzanetti: if I dont include mainView, then the header buttons don't appear [19:49] the whole header is missing [19:51] oh, I see [19:51] ballons: thanks! The title of that one said "not for app development' But I will try there! [19:51] mzanetti: I looked at the SDK unit tests, and noticed that they also include MainView to test the header sections. [19:51] mzanetti: but yeah it complicates things :) [19:51] nik90: well, in that case you might want to just use MainView, instead of the wrapping Item {} [19:51] just an idea, didn't try it [19:51] but both is fine I guess [19:51] the test is ok [19:52] awesome. I will try without the item [19:53] mzanetti: any thoughts why make test doesn't work? [19:56] not really atm === salem_ is now known as _salem [19:58] nik90: http://paste.ubuntu.com/8038996 [19:59] mzanetti: yup I am not sure why it doesn't the test [19:59] mzanetti: it has the same working folder as the autopilot test [19:59] but its working, no? [19:59] bbiab [20:01] mzanetti: ah..I expected some verbose output about what passed. But yeah it works [20:05] nik90: m-b-o mhall119 http://146.185.188.87/test/?query=Bromley [20:05] pls test [20:06] has admin1 and admin2 [20:06] popey: admin3 is used in weather [20:07] will figure out how to add admin2, it's not in the database dump [20:08] nik90: ctest -V [20:08] popey: admin2 or admin3? [20:08] 3 [20:08] popey: ok. it's very usefull, at least in germany [20:09] I'll figure it out ☻ [20:09] m-b-o: example city pls? I want to check [20:09] nik90: Wolverhampton [20:10] davmor2: cities with duplicate names :) Yours returns only 1 result :D [20:10] no other Wolverhampton in the world [20:10] nik90: Birmingham there is one in the uk and one in the states [20:10] nik90: Washington [20:11] davmor2: awesome thnx [20:11] nik90: leek, leeds, new castle [20:11] davmor2: good lord there are so many Birminghams in uk and us [20:11] perfect example to test === _salem is now known as salem_ [20:12] nik90: no worries chap :) [20:14] nik90: shrewsbury, stratford will be another good example too I think [20:14] mzanetti: where do I add the cTest -V ? [20:14] nik90: run that instead of make test if you want more verbose [20:14] lowercase t btw, ctest -V [20:16] mzanetti: I got http://paste.ubuntu.com/8039105/. Is there a way to replace basic with AlarmLabelPage (test suite name for instance) [20:17] m-b-o: do you have an example German city? [20:18] Ludwigshafen [20:18] nik90 popey: ^ [20:18] nik90: strange... this is what I get: http://paste.ubuntu.com/8039113 [20:18] ta [20:20] it's odd, in all the dumps that geonames provide, the lookup for admin3 isn't there... [20:20] http://download.geonames.org/export/dump/ [20:20] so while I can find out that Ludwigshafen has admin code 3 of 07338, I can't look that up [20:20] nik90 popey: most time you can distinguish enough with admin1 and admin2, but the name used in admin2 is very common. admin3 is on the car plates [20:20] sorry: admin2 not verycommon [20:21] I don't doubt admin3 is useful, just trying to figure out how to look it up ☻ [20:23] mzanetti: strange after running cmake again, it shows what you pasted :) [20:28] found it! [20:36] m-b-o: so would you expect in that json response to say admin3: 07338, or admin3: Rhein-Pfalz-Kreis ? [20:37] popey: the last one, the name [20:39] hmmm [20:40] popey: there's a premium dump too. so my guess is, admin3 is there.... [20:41] oh? [20:41] premium. as in not cc licensed? [20:41] I can't see any easy way to pick the admin3name out. hmm [20:45] popey: yes, in the postal codes dump is a field for it, but no data [20:53] balloons: can you have a look at http://91.189.93.70:8080/job/generic-mediumtests-utopic-weather-app/46/? please? [20:53] balloons: seems there's an issue with the header emulator and small icons [20:55] popey: m-b-o: nik90: what data is it that you actually *need* from this? [20:55] mhall119: https://bugs.launchpad.net/ubuntu-weather-app/+bug/1326918/comments/5 [20:55] Ubuntu bug 1326918 in Ubuntu Clock App reboot "API key for geonames.org" [High,In progress] [20:56] popey: but what is the data in admin3 that we need? [20:56] mhall119 popey: https://bugs.launchpad.net/ubuntu-weather-app/+bug/1326918/comments/4 [20:57] mhall119: it's a regional name, like in the UK example we had GB (country), ENG (admin1), GLA (admin2), the Admin3 might be "Bromley" [20:59] mhall119 popey: if admin3 it's not possible, it's not that big problem. [21:01] popey: do you have an example of what's in the admin3 field of geonames for Bromley? [21:01] well, bromley was an example [21:04] mhall119 popey: admin3 gets only displayed, when the location name and admin1 is not unique [21:04] in general admin3 is a numeric value [21:04] adminName3, sorry [21:06] adminName1 an adminName3 gives you a distinction around up to 50km, adminName1 and adminName2 up to 200km. At least here [21:07] but adminName3 is not very often used [21:08] mhall119, popey: In http://146.185.188.87/test/?query=Birmingham do I really need to show adminName2 ? [21:08] m-b-o, what's the issue you think? [21:08] Shouldn't Name + adminName1 + Country suffice? [21:09] balloons: the click for opening the Sheet happened not on the icon [21:09] balloons: but that's just a guess [21:16] nik90: it appears that would give enough data to pick the correct one [21:16] popey: I still can't figure out what sphinx is being used for [21:17] heh [21:17] mhall119: yup that's what I think. I am in a bit of dilemma as to how to present this data properly. [21:17] its the initial lookup of city name to city id i think [21:18] popey: yeah....but that's all in the database, so I don't know *why* sphinx would be needed [21:18] ideally I think name \n AdminName, Country should do [21:18] nik90: that should work for > 90% of the cases at least [21:18] you need to try some names which have lots of dupes [21:18] like Newport [21:19] and make sure you can differentiate between them [21:19] I tried Springfield which we have a lot of in the USA, and nik90's proposed display would be fine [21:21] ah, nope, Jefferson breaks it, because there's 2 on Georgia [21:21] adminName1 is "England" [21:22] mhall119, popey: Do we have so much accuracy though? I mean we need this for the world clocks to choose timezones essentially. What does it matter if there are 2 jeffersons in Georgia? They are bound to have the same timezone right? [21:22] mhall119, popey: Choosing either one without knowing which one is still alright since the timezone will be correct [21:23] i think there are some edge cases, but it's "good enough" [21:23] we have other bigger issues to solve ☻ [21:23] :) [21:23] yeah, I'm happy with "good enough" [21:23] I say for the clock app, I should be good with Name, AdminName1 and Country. [21:23] adding too much will clutter our beautiful clock :D [21:24] nobody likes Georgia anyway [21:24] rofl [21:24] http://en.wikipedia.org/wiki/List_of_popular_place_names [21:24] there's your test cases [21:24] in order [21:25] good lord [21:25] * mhall119 loves the List_of_* articles in Wikipedia [21:26] http://en.wikipedia.org/wiki/List_of_Wikipedia_Lists goodbye everybody, I'll be on wikipedia until sometime in 2021 [21:28] http://uncyclopedia.wikia.com/wiki/List_of_weapons_that_don't_exist,_but_should is better :) [21:32] lol [21:35] Is there a way from an app to know the wifi around and/or the cell tower? I'm thinking about implementing https://mozilla-ichnaea.readthedocs.org/en/latest/api/index.html#service-api [21:41] mhall119 popey nik90: adminName2 would be a nicetohave in weather [21:42] ok, I'll push my changes which are at http://146.185.188.87/test/?query=San%20Jose [21:43] I'll do that in the morning when I am not half asleep ☻ [21:43] this does mean I'll be the most recent person to change that code which probably means it's now mine forever [21:45] popey: thank you [21:45] np [21:47] popey: thnx. I will test again with your server tomorrow morning as well [21:48] kk === salem_ is now known as _salem [22:42] balloons: ping [22:43] Can my click package not contain a '+' symbol? If it does, I can't launch it from the Dash. http://paste.ubuntu.com/8040118/ [22:43] GDBus.Error:com.canonical.URLDispatcher.BadURL [22:46] karni: i have no clicks on my device that have a + in the name [22:46] karni: but it probably isn't good, no [22:46] popey: ^([a-z0-9][a-z0-9+.-]+)_([a-zA-Z0-9+.-]+)(?:_([0-9][a-zA-Z0-9.+:~-]*))?$ [22:46] popey: just got this from Chipaca [22:46] cat fall asleep on the keyboard? [22:47] popey: trying to append +build to the version (semantic versioning), and it's a no go [22:48] popey: treasure found, https://wiki.ubuntu.com/AppStore/Interfaces/ApplicationId [22:49] gotcha [22:53] kalikiana: hi! when's the chance we'll get the config path (with application domain) fix in the image? or is it already there, and I should fix my app somehow?