[01:17] evening all [01:21] yo [01:25] popey, Was thinking of a new test for Jenkins - Tell me if this is within the scope. The check would be to make sure text: "properties", title: "properties" and similar include the i18n.tr() [01:35] balloons, you around? Can you re-approve https://code.launchpad.net/~dpm/ubuntu-filemanager-app/desktop-i18n/+merge/230591 ? [01:35] popey, for MPs that failed autolanding, just re-approve them. [01:36] fginther, jenkins issues? [01:36] fginther: ok [01:37] fginther: what do you mean by re-approve? [01:37] popey, and rebuilding the "generic-mediumtests-utopic" job won't retrigger the landing, for that you need to rebuild the 'parent' job [01:37] popey, just top approve it again [01:38] popey, it was top approved before, but failed due to the jenkins testing issue [01:38] fginther: top-approving triggers jenkins? [01:38] popey, yes [01:38] but wont merge until jenkins passes? [01:38] popey, correct, it will re-run the tests. [01:38] ok, done [01:39] is that the case for all the failed merges? [01:39] just top approve each? [01:39] popey, yes, if they were top-approved once and autolanding failed, they can be top-approved again. [01:39] that seems surprising to me, I've never done that, always followed the link in the merge comment which says " to re-trigger, click here" [01:40] ah okay [01:40] i see now. tired eyes ☻ [01:40] popey, if it's just a MP test (a '-ci' job) then yes, you have to clike the rebuild link for those [01:41] Akiva-Thinkpad, the jenkins issue from the last few days has been resolved, but there is work to rerun all of the MPs that failed as a result [01:42] fginther, fun stuff. To that though, was thinking of a jenkins test, to make sure i18n.tr() is included for text properties. What do you think? [01:47] Akiva-Thinkpad, a style checker is a reasonable thing to have. So long as it can be done in a way that is compatible with the future of click packages. So, if it can be done as part of the cmake process that already builds and test the package, then it should continue to work in the future. [01:47] Akiva-Thinkpad, the stuff we're doing now with jenkins and building deb packages needs to be deprecated [01:47] interesting. Thanks for letting me know. === _salem is now known as salem_ [02:12] blah sorry to anyone who got spammed with me mucking around in LP. Was trying to add a qtorganizer5-eds package to a bug in Ubuntu Calendar App. === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === salem_ is now known as _salem [05:34] Anybody know about the review process? Have an app that's been Pending Review for 2 weeks now. [05:34] Wondering if something's wrong... [05:35] 2 weeks today [06:00] popey: it seems we are having problem with calendar AP or still it is Jenkins issues ? [06:56] morning all [07:00] good morning [07:00] interesting; mx4 stats are up [07:00] perhaps the ubuntu phone platform? [07:01] holy smokes ; 20.7 mp camera [07:12] for the file manager, I'm getting "No executable specified" [07:29] Akiva-Thinkpad: Morning :) [07:29] dholbach: morning :) [07:39] hi mihir [07:40] hey dholbach :) [08:25] Good morning all; happy Newspaper Carrier Day! :-D [08:26] Goodmorning JamesTait [08:27] that can't be real [08:28] I don't make them up - it's on the internet, it must be true. Right? ;) [08:29] I don't mind - it reminded me of http://www.ilovethe80s.com/arcade_fullsize_paperboy1.htm [08:30] JamesTait, game was also made for n64 [08:31] hey do you think its possible to get the label width of an option selector delegate? [08:34] popey, how is the core apps landing going? [08:34] looks like music and reminders got an update [08:35] dholbach: do you know when there will be a new devel version (current on my phone is 203) because I cant unlock my pincode :) [08:37] justCarakas, best to ask in #ubuntu-touch - I don't know, sorry [08:38] dholbach: i spent some time last night pushing things through. [08:38] dholbach: continuing today [08:39] oki [08:48] popey: it seems we are having problem with calendar AP or still it is Jenkins issues ? [08:49] dholbach: https://wiki.ubuntu.com/Touch/Devices/M7 this link provided there says in the lasts post that it has moved to http://forum.xda-developers.com/showthread.php?t=2780273 [08:50] ok? [08:51] I didn't know who to report this to but your name was the last one to change the page [08:51] didn't know if the page needed to be updated than :) [08:51] ah ok - I think I helped the folks set up the port [08:51] some didn't know how to set up a wiki page [08:52] ow oke :) [09:00] mihir: yes. looks like genuine AP failures. [09:01] mihir: I have re-run them and many are now passing, looks like there's some issue with the AP tests. Are you able to take a look at them? [09:02] mihir: i left a comment on this one https://code.launchpad.net/~pkunal-parmar/ubuntu-calendar-app/YearView-curmonth-scroll/+merge/230781 [09:37] popey:yup i saw that comment. [09:37] popey: i'll try to look at APs, [09:41] thanks mihir [09:48] popey, do you know why we dont have a "latest" category in the store ? [09:48] because it hasn't been prioritised I guess [09:48] :/ [09:48] a few people have asked beuno about it [09:49] guess it's on "the list" but not at the top [09:49] yeah ... i'd really like to see it more than "top apps" etc [09:49] likewise [09:50] I recently looked at all the categories but its rather time consuming to do that every time [09:50] yup [09:55] +1 :) [10:02] dholbach: hey, If I forgot to commit something and then accidentally did bzr revert. Would I be able to retrieve the lost contents through some magic command? [10:07] bzoltan: hey, so the new qtcreator-plugin-release works nicely. I like that the console output does not get truncated or lost when running on devices. Also it no longer overwrites an existing app. [10:07] nik90_: Good to hear [10:07] nik90_: I release it right now [10:07] bzoltan: I had one tiny issue regarding the overwriting check. In the emulator, I removed the clock app. But even then I wasn't able to run it on the emulator without changing the name in the manifest file. [10:08] bzoltan: I checked with "click list" and also the /opt folder in the emulator to make sure the clock app was properly uninstalled. [10:08] bzoltan: but as I said this is too tiny of an issue. So yes go ahead with the release [10:08] nik90_, I don't know to be honest :/ [10:08] nik90_: hmmm.. let's check it with zbenjamin [10:08] * zbenjamin hides [10:09] dholbach: no worries it was rather stupid of me to do that :P [10:09] zbenjamin: under the table? Hkhm... :D [10:09] nik90_: did you run click list as phablet or as root [10:09] bzoltan: noooo not under the table ;) that is too dangerous [10:09] zbenjamin: I phablet-shell into it, and then ran click list. So I assume as phablet [10:10] nik90_: weird, i use the Click database object to return me all installed apps [10:10] nik90_: probably its still installed for the root user [10:10] zbenjamin: I adb shell and then tried click list and yes it is there [10:10] zbenjamin: strange. I did a gui uninstall [10:11] nik90_: its a default app, probably you can not completely remove it? [10:11] zbenjamin: I suppose yes [10:11] nik90_: the debug output is much better now? [10:11] zbenjamin: oh so much better :D [10:11] good to hear :) [10:13] zbenjamin: I was thinking of requesting that you append "+testing" to the manifest name in case you notice an app already installed with the name, but then I realised I was just being lazy :P [10:14] and since that would add unnecessary complexity to this [10:14] nik90_: ;) [10:14] zbenjamin: see this is what happens when you give away too many features :) [10:15] nik90_: i can remove some if you don'T like it ;) , lets see. .... [10:15] zbenjamin: no no .. I am going to shut up now :D [10:15] nik90_: can you give me the mac adress from one of your devices? i will hardcode missing features extra for you ;) [10:16] hehe [10:21] zsombi: ping [10:22] nik90_: wazzup? :) [10:22] zsombi: I need your expertise on https://code.launchpad.net/~nik90/ubuntu-clock-app/conditionally-load-alarmmodel/+merge/232412 [10:22] zsombi: I am having an issue with that branch [10:22] zsombi: I am loading the AlarmModel{} using a Loader which is set to asynchronous: true [10:23] zsombi: However when the clock app opens the loader start loading the alarm model, if you start pulling the bottom edge, the whole UI freezes [10:23] and I have to restart the clock app to fix it [10:23] nik90_: Loader loads Items, AlarmMode is a QObject [10:23] AlarmModel [10:23] nik90_: if you want to load the Model dynamically, you should create the instance diferently [10:24] nik90_: not with Loader [10:24] zsombi: oh [10:24] zsombi: well the loader is helping me delay the loading of the alarm model until after the start up animation is complete though [10:25] which has resulted in a more smoother startup [10:25] nik90_: dude, AlarmModel is NOT an Item, so Loader will be a bit messed [10:26] zsombi: Is there any other way you recommend I can load it both dynamically and in a separate thread so that my UI won't be affected? [10:26] nik90_: of course there is :) [10:27] nik90_: Qt.createComponent(from_a_file), then with teh resulted component call create() [10:27] pls do enlighten me. I can do further research on it [10:27] ah [10:27] cool [10:27] nik90_: or, if you have the AlarmModel inside a Component{} then you call teh create of that! [10:28] zsombi: and woudl that creation happen in a separate thread? [10:28] zsombi: since I noticed that whenever the alarm model loads, the seconds hand stops briefly until the alarm model is loaded fully [10:28] nik90_: depends on how you call it. if the file is local, most likely not [10:29] nik90_: yes... becaus ethe AlarmModel fills its list from the fetched one... also, might initiate a fetch from the backend [10:30] hmm [10:30] nik90_: I was planning to get rid of that behaviour. My plan was AlarmModel to use the backend directly, and then the data should be only under the backend's hands [10:31] nik90_: which means that only the first AlarmModel used will shoot the backend to fetch [10:31] nik90_: which in your case it is the same... [10:31] yes [10:31] nik90_: but it coudl be that after all EDS does fill teh data synchronously... [10:32] nik90_: Alarm services functions are ALL async [10:32] may be I should use an animator for the clock hand animations to get that to happen in another thread [10:32] nik90_: but if teh backend blocks, I cannot really do much :( [10:32] so that they dont lock up [10:32] nik90_: that'd be a solution as well... [10:33] nik90_: but remember, Loader is not for objects, only for QQuickItem derivates [10:33] yes will do thnx [10:34] btw I filled in more info to https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1364949 to help out [10:34] Ubuntu bug 1364949 in Ubuntu UI Toolkit "Editing alarm sounds doesn't actually change them" [Undecided,Confirmed] [10:34] zsombi: let me know if you need any more debugging from the clock app [10:34] nik90_: ok, thx!!! [10:35] nik90_: but good that you reminded me!!! I'll jump on it right away! [10:35] :D [10:38] nik90_: ok, so: the Alarm shows that the sound has been changed, therefore seems the only thing missing is the update of teh Audible reminder [10:38] rpadovani: could you take a look at the AP tests for calculator? They seem to be failing on all merges.. [10:38] zsombi: ok [10:38] brb..in a hangout atm [10:39] popey, I started to test them locally, but they don't fail ( the one I tried so far) [10:39] rpadovani: hm [10:39] rpadovani: see https://code.launchpad.net/~gang65/ubuntu-calculator-app/ubuntu-calculator-app-slowdownfix/+merge/232631 [10:41] nik90_: and yes, the guitly is that one, so easy fix :) [10:43] popey, about https://code.launchpad.net/~rpadovani/ubuntu-calculator-app/1357983/+merge/232630 [10:43] popey, http://paste.ubuntu.com/8232075/ [10:43] popey, now I test also the one you linked [10:48] rpadovani: Hmm that passes here === om26er_ is now known as om26er [10:49] rpadovani: lemme ask qa === MacSlow is now known as MacSlow|lunch [11:08] rpadovani: are you able to run those tests on mako? [11:08] popey, i try [11:11] rpadovani: thanks! [11:20] hello! o/ [11:21] can i get help about ubuntu sdk for ubuntu touch here? [11:25] anyone? [11:25] coderus: hi, perhaps I can help you [11:25] oh, thanks god :D [11:25] basically: i'm running ubuntu 14.04 x64 on my laptop [11:26] installed ubuntu-sdk and ubuntu-emulator from usubntu-sdk ppa [11:26] from wizard installed all kits [11:26] creating emulator from wizard failed, created it manually [11:26] emulator running fine [11:26] but i can't build project :( [11:26] error: http://paste.ubuntu.com/8232604 [11:27] any help and/or ideas appreciated [11:29] coderus: Hmm, I have no clue about make errors unfortunately [11:29] I know that the current dev version of the sdk needs Qt 5.3, which is only available in 14.10 [11:29] perhaps it's related to that [11:30] popey: Hi, can you help? ^^ [11:30] but it seems something wrong at very beginning: E: 15binfmt: update-binfmts: unable to open /var/lib/schroot/mount/click-ubuntu-sdk-14.10-i386-7241155e-9b7a-4167-b339-21bbd7867975/bin/sh: No such file or directory [11:31] Also in "manage kits" all ubuntusdk kits are marked with red exclamation mark [11:31] probably you right, because in Qt version field it shows: Qy (ubuntu-sdk) (invalid) [11:32] any ppa i can upgrade Qt to 5.3 for test? [11:32] * coderus dont want to upgrade to unicorn now, needs stability [11:34] coderus I know, that you can somehow upgrade via a ppa to 5.3, but that can give you problems when it the upgrade to 5.3 will happen [11:35] you'll perhaps stick to the emulator... [11:35] nik90_: kalikiana: https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/alarm-sound-update-fix/+merge/233340 [11:37] m-b-o: hm, but with qt installer i can install it in other path, isn't it? [11:37] zsombi: cool thnx. Will test it [11:39] m-b-o: that's not true...I have been coding in trusty and testing clock on emulator for a while now [11:39] nik90 coderus ah, ok [11:39] zsombi: test case? 0--:-) [11:39] coderus: did you create the 14.10 schroot kits in qtcreator? [11:40] kalikiana: I knew you will ask about :) [11:40] nik90_: yes i created UbuntuSDK for i386 (GCC ubuntu-sdk-14.10-utopic) [11:40] and emulator from utopic-proposed [11:40] coderus: and was that kit chosen for the emulator? [11:40] kalikiana: the problem is that in order to have it, I should add functionality to the adapter that wouldn't be used anywhere else but in tests... [11:41] nik90_: http://i.imgur.com/OJAznge.png [11:41] kalikiana: so the Q is: do we want something like that? The test would be to see if the QOrganizerTodoItem has the audible reminder set with the proper sound file... [11:42] coderus: hmm why does it say Qt version invalid there? [11:42] kalikiana: I cannot fetch the QOrganizerItem in the tst_alarms, as it is implementation specific, and there should not be anything that is impl. specific in the tests... [11:42] coderus: are you having issues only with running in emulator or in the desktop as well? [11:43] zsombi: why isn't the memory backend enough for that? [11:43] coderus: also did you create that kit yourself or did qtcreator automatically create it for you? [11:43] nik90_: Desktop settings: http://i.imgur.com/uQSxXfg.png [11:43] kalikiana: it's not the about the backend, it's teh whole organizer itself. tst_alarms has no knowledge about the backend at all [11:44] nik90_: desktop build working ok [11:44] zsombi: yes. but the code is agnostic. I don't see anything that would be backend-specific [11:45] kalikiana: QOrganizer is backend specific... the whole QOrganizer stuff is implementation specific, we can anytime switch to some other backend [11:45] kalikiana: let's say we decide some day to use timed, and that's not based on QOrganizer at all [11:45] coderus: hmm ...mine looks like http://pbrd.co/1rvKHgP [11:45] zsombi: what I mean is, the code is not checking which backend it is… so where does the implementation matter? [11:46] kalikiana: it woudl if I'd try to see whether teh sound is saved in teh QOrganizerTodo as reminder or not [11:47] nik90_: probably i just need to instakk right Qt manually? [11:47] kalikiana: in which case I'd need to bring QOrganizer to tst_alarm [11:47] coderus: my qtversions looks like http://pbrd.co/1rvLr5y [11:47] coderus: do you have the SDK ppa installed? [11:48] nik90_: also available qt versions looks like this: http://i.imgur.com/h1nnqZa.png [11:48] kalikiana: one solution woudl be to have a function in alarms_adapter, which would check these kind of things, but that would be used only for testing purposes... [11:49] nik90_: Qt Versions tab: http://i.imgur.com/vCD5IF4.png [11:49] coderus: not sure, but that looks wrong [11:49] yes [11:49] ubuntu-sdk ppa enabled [11:49] kalikiana: and we would have it as useless code otherwise... [11:49] coderus: may be try starting fresh by deleting the config folder in that screenshot [11:49] nik90_: deb http://ppa.launchpad.net/ubuntu-sdk-team/ppa/ubuntu trusty main [11:50] zsombi: why can''t the test case use the normal public API that apps can use? there's no need to know about the backed, it's already abstracted [11:50] coderus: the .config/ubuntu-sdk [11:50] kalikiana: because there's no API in Alarms which checks whether I have a detail like reminder set or not === _salem is now known as salem_ [11:51] kalikiana: changing the sound must be implemented with organizer so that it is saved in description AND reminder is updated [11:51] kalikiana: other backends may need less/more [11:51] nik90_: after removing folder and restarting ubuntu-sdk: http://paste.ubuntu.com/8232776 [11:52] zsombi: except the clock is able to get that information [11:53] you can edit an alarm and see the sound you picked [11:53] coderus: I think you pasted the wrong link :P [11:53] kalikiana: yes, and even till now that was good. The problem was that the clock had shown that the sound is updated, but teh alarm was still playing the original sound === tsimpson_ is now known as tsimpson [11:53] kalikiana: so the test case that woudl be based on that wouldn't fail even without this change [11:53] nik90_: ubuntu-sdk folder: http://paste.ubuntu.com/8232779/ [11:54] nik90_: yep, pastebin plugin just lagged [11:54] coderus: try deleting that folder. That should reset the ubuntu-sdk config file === ubot5` is now known as ubot5 [11:54] coderus: may be it might remove those invalid entries [11:54] kalikiana: that's what I'm trying to say: it's not enough to use public API, to properly test this we need to go under the hood [11:55] zsombi: hmmmm so the solution would actually be a test backend so we know it saved the correct value [11:55] kalikiana: yes... so the easiest - and perhaps teh clearest - way woudl be to provide a func for that [11:55] nik90_: ls output was right after deleting its folder, and this is screenshot: http://i.imgur.com/O4gumKs.png [11:55] kalikiana: the Adaptation layer could provide a verifier func [11:56] kalikiana: that we use for testing purposes only [11:56] coderus: then I am out of ideas :/ [11:56] coderus: but I have been developing on trusty with the help of the emulator (for normal qml projects and cmake projects_ [11:57] coderus: qmake is not supported [11:57] zsombi: btw can you join #security? I'm gathering people to talk to about XDG_ folders etc [11:58] zsombi: I see now why it needs new functions; to avoid the whole backend thing. so I guess that makes sense === om26er is now known as om26er|afk [12:02] popey: dailymotion submitted a webapp, but our "automatic" checker doesn't like the camera permission...do I need to take that out or can it be manually reviewed through? === MacSlow|lunch is now known as MacSlow [12:06] nik90_: ok i removed all kits and tryig to install it again [12:16] zoopster: I actually reviewed and rejected it [12:16] zoopster: yes, it needs updating/fixing and re-submitting with a new version [12:30] * nik90_ waves good bye to the old add world city button [12:31] popey, calculator tests on phone don't fail [12:32] popey, and I'm not able to reproduce the bugs the tests raise in any condition [12:34] popey: it was made by dbarth - the rejection is because of the camera permission? [12:44] popey, understood. https://code.launchpad.net/~gang65/ubuntu-calculator-app/ubuntu-calculator-app-slowdownfix/+merge/232631/comments/569346 === charles_ is now known as charles [13:01] rpadovani: thanks [13:05] mhall119: missing file http://developer.ubuntu.com/api/qml/sdk-1.0/QtLocation.location-maps-cpp/qgeoserviceprovider.html coming from http://developer.ubuntu.com/api/qml/sdk-1.0/QtLocation.location-maps-cpp/ [13:18] What should I do if my app is stuck in Pending Review (2 weeks yesterday)? [13:28] kalikiana: hmmm, it looks like it's referencing C++ docs :( [13:31] mhall119: well, those are C++ docs. so that's not as such surprising :-D [13:31] why are there C++ docs in my QML space? [13:32] * mhall119 grumbles at Qt [13:33] hehe [13:35] brendand: sry, but do you still have the link which shows how to translate items in a listmodel? [13:36] brendand: like an idiot I forgot to bookmark it [13:36] or DanChapman ^^ [13:38] nik90_, it was a pastebin, so i don't have the link - but: http://paste.ubuntu.com/8233508/ [13:38] that's a new one [13:38] brendand: sry for the trouble, thnx [13:39] nik90_: ping [13:39] mihir_: pong [13:40] nik90_: to merge this in other branch , i have to set tag as Merged , instead of Approve , https://code.launchpad.net/~ubuntu-calendar-dev/ubuntu-calendar-app/revamp-reminder-deisgn/+merge/232725 ? [13:40] mihir_: you need to manually merge it :) [13:40] mihir_: jenkins doesnt know about that branch since you are not targeting the trunk branch [13:40] Ahhh, you meant by checking the difference :-s ? [13:41] mihir_: no go to our common branch [13:41] and then from there type "bzr merge yourbranchname" [13:41] nik90_: okay will do that [13:41] that should merge your branch into our common branch [13:41] after that commit it and say "merged mihir's branch or something" === om26er|afk is now known as om26er [14:18] nik90_: ping [14:21] i enabled ppa:canonical-qt5-edgers/qt5-proper and did dist-upgrade [14:21] now i have 5.3.0 in usubtu-sdk compilers, but... [14:21] it tagged as unknown compiler: http://i.imgur.com/YBZA5Gk.png [14:22] and this error in kits appear: http://i.imgur.com/GpdO8f9.png [14:22] full kits info: http://i.imgur.com/iGuzJJi.png [14:22] coderus: let me check what compiler I use [14:23] can you list your ppa with ubuntu-sdk qt 5.2.1 ? [14:23] coderus: I have Ubuntu GCC Compiler chosen [14:24] yes i saw your screenshot: http://pasteboard.co/2N56VroP.png [14:24] coderus: the only sdk related PPA i have is http://ppa.launchpad.net/ubuntu-sdk-team/ppa/ubuntu [14:24] can you list your ppas? [14:24] hm [14:24] coderus: and that doesn't give me Qt 5.3 [14:26] what versions qtcreator-plugin-ubuntu-common you have? [14:26] mine is 3.1.1+14.10.20140903.3-0ubuntu [14:26] mine too [14:26] strange [14:27] can you describe the issue you have in the ubuntu touch mailing list [14:27] that might give it more exposure and people can take their time to come up with a solution [14:27] show me your Compiler tab with i386 selected please [14:27] sure [14:28] nik90_: sure, please give me link to info about joining it and i.e. [14:28] coderus: https://imgur.com/jKr3kOC [14:29] x86 please [14:29] but you have ABI values unset too [14:29] how does it detect if can build with selected compiler or not then? [14:29] coderus: https://launchpad.net/~ubuntu-phone has the mailing list option at the bottom [14:30] Policy: You must be a team member to subscribe to the team mailing list. [14:30] how? [14:33] ah, sorry, dumb me :D [14:33] joined list [14:34] :) [14:34] it is a open team === Kaleo_ is now known as Kaleo [14:36] coderus: x86 is not that different https://imgur.com/0ABq4fr [14:36] nik90_, pour vous mon ami: http://www.theorangenotebook.com/2014/09/autopilot-test-runners.html [14:37] balloons: merci beaucoup mon ami [14:37] de rien [14:38] I pushed it forward in the list to get it published [14:38] cool [14:38] balloons: oh btw I talked to pitti [14:39] nik90_, awesome, I was going to bug him today about his thoughts [14:39] we technically could/should open a bug to bring it to trusty [14:39] balloons: at the end we came to the conclusion that people with trusty can get it quick from "git clone git://git.debian.org/git/autopkgtest/autopkgtest.git" [14:39] balloons: and then run it from that folder without any compilation since it has a nice little script [14:39] odd, he didn't like the idea of stealing the deb? [14:40] balloons: no he did tell me to get the deb [14:40] balloons: but for further updates it would be better to track the git repo since updates require backports as well [14:40] yea, you would have to regrab the deb [14:40] yup [14:40] nik90_: just sent email to ubuntu-phone list [14:41] coderus: cool [14:51] nik90_: i manually added compiler and selected abi optios as following: http://i.imgur.com/T4FLBSx.png [14:51] nik90_: now my kits are happy: http://i.imgur.com/IphSkLY.png [14:52] but :D [14:52] nik90_: new error: http://i.imgur.com/ntzgo0K.png === fgallina is now known as fgallina-lunch === boiko_ is now known as boiko [14:58] ping renatu: this is the branch we discussed last week: [14:58] https://code.launchpad.net/~canonical-platform-qa/address-book-app/autopilot-get_contacts/+merge/233104 [14:58] it's ready for your review. [14:58] i'm tired fixing it, why it cant jsut work out of the box like Sailfish SDK? [14:58] i'm waiting for mailing lists answer and give up then :D [14:59] elopio, could you do the MR against the staging branch? [14:59] renatu: ah, sorry, I forgot you have staging. One second. === caribou_ is now known as caribou === daker_ is now known as daker [15:03] renatu: done. I'll ping you when jenkins confirms the tests also pass there. [15:03] thanks [15:10] balloons: when did they add autopilot-sandbox-run ? that's so coool.. I remember asking about this at Malta :D [15:11] nik90_, it's been around for a bit.. [15:11] hmm well I never knew about it [15:11] there's some other goodies I imagine you'll like too. I need to finish the post about autopilot features [15:11] I am still in the process of reading the first one. [15:11] nik90_, yep, I want to highlight some of this stuff no one knows about [15:12] I wrote the runners pretty quick, so ping if something is confusing [15:12] *runners post [15:13] ok === fgallina-lunch is now known as fgallina [15:46] ping jhodapp: can you please review this one? [15:46] https://code.launchpad.net/~canonical-platform-qa/mediaplayer-app/fix1359040-skip_seekbar/+merge/231508 [15:47] elopio, sure [15:47] thanks. [15:47] np [15:57] balloons: could you please upload http://s-jenkins.ubuntu-ci:8080/job/terminal-app-click/lastSuccessfulBuild/artifact/out/com.ubuntu.terminal_0.5.137_armhf.click to the store? [15:57] thank you! [15:58] popey, https://myapps.developer.ubuntu.com/dev/click-apps/157/changerequest/ [15:59] BOOM! [16:02] popey, hop on the call matey [16:03] I'm trying to build a QML app that displays the meals served at my local university canteen. The source is a (buggy) HMTL site. Is it possible that QML is currently not able to parse HTML? I tried everything, from XMLHttpRequest to XMListModel, nothing works and lots of standard JavaScript functionality seems to be missing, e.g. XMLHttpRequest.responseXML is always "null" and XMLListModel cannot cope with the buggy HTML input. [16:05] mr2515, no that will never work [16:05] mr2515, html is not really designed to be parsed like that - but there is stuff like beautiful-soup which can do it [16:05] mr2515, not sure qml has anything available for that [16:07] balloons: hmm, i approved terminal and it's not showing up as an update for me... [16:07] * balloons looks [16:07] brendand: Great, so I probably have to resort to a C++ QML plugin for HTML parsing, which is currently a pain to build for Ubuntu Touch. [16:07] I see an update popey.. to 13 [16:07] oh i see it on my nexus 7 [16:07] sweet [16:08] 0.5.137 [16:08] mr2515, or look at what's available in javascript [16:08] \o/ [16:08] mr2515, qml supports js easily [16:09] balloons: want to get as many of these outstanding merges in as possible and then push to the store later tonight ready for the friday image build at 3am.. will you be able to upload some more later? [16:09] popey, yep indeed. Keep approving them and pushing them in [16:14] ya [16:20] mr2515: I am not sure when the last time you checked, but these days a qml + c++ plugin app is so easy to build and run on all devices. [16:20] mr2515: I do that everyday now with the clock app [16:21] mr2515: the only thing you need is figure out how to use cmake [16:45] popey: weather pagestack MP ready to land :) [16:54] m-b-o: SWEET! [17:06] balloons, ping [17:07] ahayzen, pong [17:07] balloons, i'm on the latest image trying to run autopilot.... [17:07] balloons, and i get .. ERROR proxies:410 - Introspect error on :1.145:/com/canonical/Autopilot/Introspection: dbus.exceptions.DBusException: org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken. [17:07] 18:05:07.938 DEBUG proxies:413 - Executing introspect queue due to error [17:08] balloons, is this due to me having something out of date? or doing something wrong? [17:08] balloons, the actual app starts just it can't find anything to click on? [17:08] ahayzen, that error typically means you need to run the click hook [17:08] if not, something else might be wrong [17:08] ...click hook ? [17:09] phablet-click-test-setup ? [17:12] balloons: ping [17:12] balloons: are you still making to changes to your MR ? [17:13] mihir, yes [17:14] ahayzen, phablet-config autopilot --dbus-probe enable [17:14] balloons, ....just tried that as well...maybe a restart is in order :) [17:15] balloons: okay no issues...it is failing because of addressbook issues ? [17:16] mihir, yea.. trying to make sure it will still run properly using the packaged version of addressbook [17:16] balloons: okay ,thanks for looking at this , let me know if you need my help to hack it together [17:29] balloons, ok a restart fixed it :) [17:29] renatu: https://code.launchpad.net/~canonical-platform-qa/address-book-app/autopilot-get_contacts/+merge/233373 it's good to go. [17:29] elopio, thanks let me check it [17:30] thanks. [17:31] elopio, just one thing that I am worried about [17:31] objectName: 'contactDelegate' + index [17:31] is this really necessary? [17:31] renatu: it makes it easier to find the elements, and more stable in case of UI changes. [17:31] I am afraid of create new bindings for delegates [17:31] we have so many bindings already [17:32] but I could use only the y position to get the index. That's a little harder to understand and maintain. [17:32] renatu: what's the problem with the bindings? [17:32] this consume extra creation time and memory [17:32] is a small portion BUT [17:33] if we can avoid that on delegates [17:33] will be nice [17:36] renatu: the problem would be that each time we have to select a contact, we would have to get them all, making the tests slower and a little repetitive. And a future problem would be when selecting an item form a big list of contacts, because they won't be all on the QML tree at the same time so there's no easy way to get them all. [17:36] renatu: but if you prefer that, I will make the change. [17:38] elopio, maybe this is is not a big problem, let me ask Kaleo [17:38] Kaleo, what do you think about using something like: objectName: 'contactDelegate' + index on listview delegate for help on tests [17:39] I am concerned that will use a very small amount of memory and impact on object creation time, for something that will used only on tests [17:42] renatu, http://developer.ubuntu.com/api/devel/ubuntu-14.10/python/autopilot/faq/faq.html#q-what-is-the-impact-on-memory-of-adding-objectnames-to-qml-items [17:42] renatu: it is a concern I share [17:42] renatu: though I do not have measurements [17:43] balloons, my concern is not adding a object name but creating a binding [17:43] Kaleo, the link notes "With 10000 different objectNames 1336 kB of memory are consumed which is around 127 Bytes per Item." I didn't do the work to arrive at that, but there you are [17:44] balloons: it is not memory that is the concern [17:44] balloons: but delegate creation time [17:45] balloons, objectName: 'contactDelegate' + index this will cause QML to create a binding to monitor "index" value [17:45] to make sure that everytime that index change the objectName get updated [17:48] renatu, Kaleo ahh.. this dynamic objectname has a runtime effect [17:48] more than a onetime binding at creation [17:53] renatu, Kaleo so I suppose we should measure this and get an answer and some perspective [17:56] popey: ping [17:57] mihir: yo [17:57] m-b-o: did you want me to bump this weeks meeting to tomorrow or are you okay this week? [17:57] greetins [17:57] popey: i have merged the all branch to new event redesign, so you can have look and give your inputs [17:57] mihir: will do! thanks. [17:57] popey: also, fixed ap..hope it passes all [17:58] are linux app development questions acceptable here? [17:59] popey: i'm alright eith it. waiting to get started! Is it ok for you? [17:59] elopio, so are you going to make the change off the cuff? I think we should take a moment to figure out the impact before abandoning the dynamic namin [17:59] ok m-b-o [17:59] renatu, Kaleo, balloons: interesting. [18:00] balloons: I don't know what to do. The change for small lists is easy. But making it work for lists that need swiping it's not. [18:00] we are using this pattern in many places, so it would be nice to know if we should stop doing it. [18:01] elopio, I do not think this is a problem for small amount of objects for example controls in a page [18:01] but this can have big impact in list delegates [18:01] elopio, exactly. So I think we should pounce on thomi and veebers when they show up in a couple hours and dedicate a few cycles to understanding this. Then we can update the FAQ as well so we have a best practice [18:01] for example in my phone I have more than 1000 contacts [18:01] how can we measure the impact? [18:02] every time that I remove a contact I will have 1000 new signals fired because of the index change [18:02] elopio, there is a profiler inside qtcreator [18:02] renatu: but how many signals do you fire without this line? [18:03] and we can get data back on how many bindings we made, etc [18:03] if it's only one, then this is really bad. But if you are already sending 5 per contact, it's not so much. [18:03] elopio, probably I am firing a tons of signals :D [18:04] I guess renatu's point is simply that any impact merely for the sake of tests is suspect in shall we say, end-user code [18:04] but is nice to avoid necessary signals [18:04] *unnecessary [18:04] balloons, yes this my concern. Thanks ;) [18:05] renatu, actually, you might be better suited to doing the work to discover the impact.. You are probably more familiar with the profiling tools than me [18:05] I agree there. But if we ever want to automate a test with 1000 contacts, ugh, autopilot will take a lot of time swiping back and forth. [18:05] also the problem that more than what's visible on the screen is loaded on the QML tree, so that's not a simple piece of code. But it's doable, of course. [18:09] balloons, elopio, yes I can try measure that but I do not have time these days [18:10] and I do not want to block elopio MR [18:10] renatu: I'm going to change my MR, with a comment that it will just work for small lists that can be loaded in memory. [18:10] and make a note for after RTM to check the impact of obectName: index on delegates. [18:10] mhall119: hey, are you still have issues with repeating alarms? [18:11] elopio, I guess I'm the outsider here, but I would rather see the impact now [18:11] balloons: I guess you won't have time for that next week, right? [18:11] I could dedicate cycles to it tomorrow [18:12] balloons: awesome then :) [18:14] renatu, balloons: the change is a 5-liner. So I'll make it, and if you find that the impact is not important, we can revert it to support big lists on the future. [18:15] nik90_: i found the issue while it loads the new event page directly [18:15] mihir: oh did you? What was causing it? === om26er_ is now known as om26er [18:19] renatu, how did you get 1k contacts? got an easy way to do so? [18:19] balloons, we have a test account [18:20] nik90_: but i don't know why it is happening , because it is even there in trunk but if i remove it works great.. [18:20] mihir: remove what? [18:20] balloons, let me find the vcard that we use for create these contacts [18:20] ty [18:21] nik90_: this line , http://bazaar.launchpad.net/~ubuntu-calendar-dev/ubuntu-calendar-app/trunk/view/head:/calendar.qml#L226 [18:22] renatu, balloons, Kaleo: http://bazaar.launchpad.net/~canonical-platform-qa/address-book-app/autopilot-get_contacts/revision/305 [18:23] nik90_: fixed the AP..now it should be ready to merge :) [18:23] mihir: wow really..that was fast [18:23] mihir: does jenkins also pass it? [18:24] nik90_: haven't proposed yet.. [18:24] elopio I am ok with that [18:24] nik90_: still running APs.. [18:24] mihir: ah ok [18:24] nik90_: it was trival change , but it did work :) [18:24] mihir: how? I thought we changed so much of the ui? [18:24] nik90_: i am still afraid of removing that line , as of now I have commented it [18:24] renatu: I am ok too for now. If we ever want to automate the performance of actually opening a contact out of 1000, this falls short. [18:24] nik90_: yup reason, I then realised that we don't have test cases for Recurrcence & Reminder [18:25] mihir: ah :D [18:25] Is there another mechanism to get an asynchronous notification of child process termination in linux besides handling SIGCHLD? [18:26] nik90_: I have changed whatever field we changed from NewEventEntryField to textField / TextArea [18:26] balloons: your results will be really interesting. Because if that 1000 test is slow because of the testability features we added to automate it, we will be running in circles. [18:28] renatu: I've kicked jenkins to get new results. [18:30] balloons: oh, and also it would be nice to measure the impact of a list with boundaries. For example, the list of months that won't have more than 12 items. [18:34] elopio, can you set the max? [18:35] balloons: I don't know if you can put a boundary on the model [18:35] mihir: cool [18:35] but the boundary here is on the domain. [18:35] mihir: I will try my best to review the code in that MR. [18:35] mihir: but lets try merging asap [18:35] nik90_: i am doing it all test cases pass on my system [18:36] what I would like to see is if we can use the simple way at least for lists that are small, but not so small as to be loaded in memory. [18:36] the list of months is that example. [18:36] * mihir excited for new event re-design [18:37] mihir: me too :D [18:38] popey, ogra_: http://pasteboard.co/2Nw5b5MX.png ....snooze away my frnds...but don't be late for the meetings...not my fault :P === Elleo_ is now known as Elleo [18:40] elopio, so you think there might be a couple ways to do this as a best practice depending on your list size? [18:42] balloons: not sure. If the list is small and the impact is big, then we go with swiping to collect all items always. If the list is big and the impact is small, then we go with the simple objectName index and swipe less. [18:43] if it depends on the list size, then we might chose to combine. I'm not sure yet. [18:43] nik90_: ooooh! snoooooze [18:45] http://news.softpedia.com/news/Ubuntu-Touch-Can-Now-Be-Used-to-Control-AR-Drones-457694.shtml ☻ [18:45] nik90_: popey there You go , https://code.launchpad.net/~ubuntu-calendar-dev/ubuntu-calendar-app/revamp-newevent-page/+merge/233414 [18:45] nik90_: i might have missed something someWhere need second eye to check :) [18:47] popey: coool apps coming to ubuntu touch...awesome times [18:48] mihir: you need to merge trunk and resolve code conflicts [18:48] nik90_: oops i forget that [18:48] mihir: there seems to be a conflict in NewEvent.qml file as expected [18:48] will do thatt now [18:48] mzanetti: your app made the news :D [18:48] i'll delete this branch and re-submit new one [18:49] mihir: why? [18:49] mihir: just merge it and fix the conflicts and then do "bzr resolve" [18:49] nik90_: okay [18:49] mihir, I think my mp is ready [18:49] balloons: okay, i made some change to make it work for new event new design [18:50] mihir, ohh excellent, where is the code? [18:50] I never got to that point lol.. this was all housekeeping :-) [18:50] * mihir get the code change for AP, [18:52] balloons: if you go to end of this merge diff , you'll see AP changes..they're trivial [18:52] balloons: https://code.launchpad.net/~ubuntu-calendar-dev/ubuntu-calendar-app/revamp-newevent-page/+merge/233414 [18:54] mihir, so I don't need to include them in my mp yes/ [18:54] ? [18:54] balloons: yes , otherwise it'll fail in current trunk [18:55] balloons: i need to merge your trunk once it get merged. [18:55] mihir, lol.. So I should take your code changes or do you plan to merge your changes first? [18:56] I can just take them if you wish [18:56] balloons: nope , it is fine you go ahead, this MR needs to be reviewed by nik90_ thoroughly and then by kunal [18:56] kk, I'll swipe them [19:03] nik90_: merged from trunk [19:03] nik90_: let me know when you get time to review this :) [19:03] mihir: cool [19:04] nik90_: if now we both can review , i can fix it before I go to office :( [19:04] mihir: oke..I will do it now [19:04] * mihir hugs nik90_ :D [19:05] * nik90_ hugs mihir back for his AP fixes and following up diligently on the MR. [19:06] nik90_: as of now I have commented the line in calendar.qml , i am not aware of consequence :-s [19:07] mihir: I am not aware either..I suppose only kunal will knwo [19:07] s/knwo/know [19:14] rpadovani: ping [19:15] rpadovani: Don't you think we should increase number limit ? [19:15] mihir, not for now, I talked about it with popey and he said that for now we leave all how is it now, because we'll have soon new design :-) [19:16] rpadovani: agreed [19:19] balloons: is it me or is this taking a very long time http://91.189.93.70:8080/job/ubuntu-weather-app-ci/ [19:19] popey, hmm.. you aren't the only one to suspect something [19:19] fginther, ping [19:21] fginther, it seems the core apps jenkins have lost contact with each other. The 2 running jobs are frozen in time, and no new jobs are spawning [19:21] balloons, looking [19:22] rpadovani: do you need any other review, or should Top Approve thate? [19:22] that* [19:23] mihir, we're waiting for design approval [19:26] beuno: we have another app which isn't showing up in any lists I can see, it's in limbo. https://myapps.developer.ubuntu.com/dev/click-apps/475/feedback/ [19:26] beuno: looks like the user missed some button or something to move it to be submitted so it's sat there with him wondering what we're doing and us not seeing it. [19:28] mhall119: ^ [19:28] balloons: did you merged? [19:31] nik90_: you reviewing , shall i wait ? [19:31] mihir: yees...few mins for dinner [19:32] nik90_: no issues, just need to know.. [19:32] popey, 2 things [19:32] there is a new version [19:32] so there's something going on there [19:33] second, beowulf from my team is now in charge of going through the whole flow [19:33] making it better [19:33] easier to understand [19:33] and a lot of etc [19:33] right, there is a new version, yes, but somehow the user didnt submit it [19:33] matiasb, can you check into that ^ [19:34] * matiasb looks [19:35] beuno, popey: as popey commented there, the user needs to resubmit the app for review (since this is not a new version for a published app) [19:37] ah [19:39] nik90_, all done, 3 top approve and 1 little fix in the code, so you know that I read all the code very carefully :-) [19:42] fginther: balloons lemme know when I can continue using jenkins pls [19:43] popey, balloons, I had to do a restart [19:44] I'll try a few jobs and if it works, will give an all clear [19:47] renatu: jenkins is happy https://code.launchpad.net/~canonical-platform-qa/address-book-app/autopilot-get_contacts/+merge/233373 [19:47] ok [19:49] rpadovani: hehe [19:50] rpadovani: erm I don't see your inline comment [19:51] nik90_, ops, I didn't publish it :P [19:51] :D [19:52] nik90_, I tought that inline comments are published on them own, but you have to do a comment to publish tem [19:52] *them [19:52] yup [19:57] nik90_: ping :D [19:58] mzanetti, are you about? I;m curious if you ever looked into the impact of dynamic objectnames on qml performance [19:58] coderus: hi [19:59] popey, balloons, one of the filesystems ran out of space causing the weird behavior, it's ok to use again. [19:59] nik90_: can you explain a bit about using cmake? [20:00] i upgraded to 14.10, now my qtversion is correct [20:00] coderus: its just an alternate to .pro and .qmake [20:00] template project running on destop kit correctly [20:00] coderus: essentionally a packaging system telling the ubuntu sdk where to install the necessary files on the device or where ever you install it [20:00] but when selecting ubuntusdk it doesnt build [20:00] popey, I also restarted that reminders MP you indicated earlier. the unusual log messages were caused by some failed cleanup that appears to have been related to the bigger issue we saw over the weekend [20:01] coderus: and this is with a template project? [20:01] nik90_: make: *** No targets specified and no makefile found. Stop [20:01] coderus: then you are doing something wrong. [20:01] nik90_: app with qml extension [20:01] coderus: please take a look at http://developer.ubuntu.com/start/ubuntu-sdk/using-device-kits/ [20:02] nik90_: i'm not doing anything yet [20:02] coderus: and http://developer.ubuntu.com/apps/sdk/tutorials/running-apps-from-the-sdk/ [20:03] coderus: if not, let's schedule a hangout tomorrow sometime where I can show you how to do it. [20:03] coderus: I am a tiny bit busy at the moment reviewing code proposal and merging stuff. [20:03] coderus: is that cool? [20:04] nik90_: sure [20:04] just give me as many links as you can [20:04] i'll read them :) [20:05] why if I call pageStack.pop() from a dialog does the app freeze? is it known bug? [20:05] (I've checked in very simple app, with just a dialog and a pagestack) [20:06] gcollura: why would you call pagestack.pop() ? [20:06] gcollura: shouldn't it be pageUtils. something something? [20:06] coderus: trust me the documentation helps [20:06] coderus: you only need it for the first time :) [20:06] coderus: ping me tomorrow whenever you are free [20:07] nik90_, I call pageStack.pop() because after deleting an item I need to go back in the home page [20:08] the weird part is that if I call pageStack.pop() from a button for example, the app still works [20:09] rpadovani: fixed the issues [20:09] gcollura: not sure I haven't tried it. Best would be to create a sample app and attach it to bug report. I will try to validate it tomorrow [20:10] ok I will submit a test app with a report [20:10] thanks for the help [20:12] balloons: yes I did... its not much [20:14] mzanetti, so linking a property to another property like "myObjectName" + index is negligible? [20:14] doesn't it require updating each time things change.. we're consider if there's 1000's of these created it could be crazy [20:15] balloons: ah... hmm... well, the do trigger binding updates obviously [20:15] balloons: but pretty much the same as doing "property string foobar: "myObjectName" + index [20:15] its a normal string property [20:15] ends up in a QString in c++ [20:16] elopio, ^^ :-) [20:16] renatu, too, if you are about ^^ [20:17] mzanetti, so it would be nice if we had stats on this like http://developer.ubuntu.com/api/devel/ubuntu-14.10/python/autopilot/faq/faq.html#q-what-is-the-impact-on-memory-of-adding-objectnames-to-qml-items, so we could list the impact. It sounds like it's not something you are worried about at all [20:18] popey, FYI all music-app autopilot will be failing at the moment (due to ms2 db change) so don't bother trying to rerun....we have an mp in progress to fix it but it needs to extra work [20:21] mihir, care to review the giant diff? It should be ready now, lol. It passes on my device and desktop https://code.launchpad.net/~nskaggs/ubuntu-calendar-app/fix-1359167/+merge/233387 [20:22] * mihir looks at giant diff :D [20:24] ahayzen, btw, I stole your cmake mod to allow *.py to show up in ubuntu sdk.. nice [20:25] balloons, hehe thanks :) just need a mod to make autocomplete work with cmake without having to open a qmlproject :( [20:25] bug 1338774 [20:25] bug 1338774 in qtcreator-plugin-ubuntu "CMake projects do not provide syntax autocompletion" [Undecided,Confirmed] https://launchpad.net/bugs/1338774 [20:25] balloons: unable to locate address_book_service_testability [20:25] balloons: is there any PPA? [20:26] mihir, for you? [20:26] balloons: yup , when i run it says no module address_book_service_testability , and when i do apt-get install it says unable to find [20:27] mihir, you on trusty? [20:27] it's utopic only [20:27] balloons: utopic [20:27] mihir, ohh then, install it [20:27] from which PPA? [20:27] mihir, bah, I suck [20:28] address-book-service-testability [20:28] it's in the archive [20:28] balloons: this is what i get http://paste.ubuntu.com/8253190/ [20:29] sudo apt-get install address-book-service-testability should fix it for you [20:29] it's in the debian/control file [20:29] but I had it as address_book_service_testability :-( [20:30] ahayzen: ok! [20:30] popey, yey upstream changes \o/ [20:31] balloons: that works for me :D [20:31] nice catch mihir thank you [20:32] balloons: also we need to fix the AP set up for calendar..all dependecy doesn't come while we install [20:32] mihir, what do you mean? all the depends are listed in the control file [20:32] balloons: we have missed some , like dummy address service [20:33] address-book-service-dummy? it's listed in there [20:33] balloons: i had to do it manually my time , will re-check [20:33] take a look in debian/control, and at the manifest [20:33] balloons: also, contacts service [20:35] mihir, does the app depend on contacts service? what package? note the depends under calendar-app [20:36] balloons: Jenkins is failing because of dh_install: calendar-app-autopilot missing files (usr/lib/*/dist-packages/address_book_service_testability/*), aborting [20:36] debian/rules:8: recipe for target 'binary' failed [20:36] balloons: sure, i'll do that [20:36] balloons: it seems it is looking for same file package ? [20:38] mihir, where do you see that? none of the runs for jenkins yet are correct [20:39] balloons: http://91.189.93.70:8080/job/ubuntu-calendar-app-utopic-amd64-ci/308/console [20:39] balloons: it is last Jenkins run on your MR 39 seconds ago [20:40] mihir, yea I keep canceling those [20:43] balloons: okay, [20:43] balloons: all ran successfully except one on my machine [20:43] mihir, everything tested ok except the new test I started to add: test_new_event_must_start_with_default_values [20:43] I really should leave that out [20:43] yup [20:43] balloons, i remember you said that one other app was doing url-dispatcher in their autopilot? can you remember which one so i can add some tests to music? [20:43] balloons: lets do that in saperate MR, [20:43] balloons: it is breaking it [20:45] done [20:46] balloons: great , lets merge this and ask other to merge with new AP on their branch [20:48] balloons: it works great now :) [20:49] balloons: can you trigger Jenkins now ? [20:51] sure === salem_ is now known as _salem [20:52] balloons: as of now we don't have test cases for recurrences and reminder, [20:52] balloons: if we write scenario, we have to scroll for a year if we put that occurence ? [20:53] mihir, yes we need to expand the new event test to add recurrence and the other features [20:54] mihir, you could setup a scenario matrix of all the possibilities you want to test, then it would run through all of them [20:54] the test would be the same.. just one test, keying off the different values we feed it from the scenario [20:54] you can see an example of this in clock app or terminal app [20:54] both have them [20:54] balloons: understood, so if say we have recurrence for 100 occurence , it will sasert 100 bubbles? [20:54] just wondering [20:55] mihir, ohh, hehe, I suppose we'll have to think about that won't we? [20:55] but yes, we'll have to check [20:55] I would go with a small number :-) [20:55] balloons: okay , obviously i'll to but was curious to ask :) [20:55] in case of date, and we have all scenaro for month, day, year [20:56] yes.. we'll have something for today, and a random date I guess [20:56] something in the future and past [20:57] balloons: got it [20:57] and all day events and something that has a timeframe [20:57] balloons: again Jenkis failed [20:57] mm I see [20:57] it's been switched to python3 now [20:58] I wonder why it's unpacking in python2.7 [20:59] one more try :-) [21:00] balloons: (Y) [21:03] more fails [21:05] got it [21:05] balloons: some diff errors, seems version issues [21:06] I feel good about this one [21:06] running again [21:08] balloons: was it version issues, or something else, just curious to know [21:08] more py3 stuff [21:08] broke again [21:09] balloons: okay [21:10] I really feel good about this! [21:17] ahayzen, number4, awesome [21:17] balloons, \o/ [21:17] ahayzen, want to migrate to python3 as part of it? [21:17] I discovered why jenkins is still py2 just now.. because the debian and cmake stuff still uses py2 [21:17] balloons, probably....but our landings are broken for the moment...waiting for victor's branch to db resolve patching [21:18] balloons, ah [21:18] balloons,how much do we have to do from our end to change to py3? [21:18] so change everything to python3 in debian control, then in the cmake file for autopilot swap to python3 [21:19] and remember to add ()'s around the print statement in there [21:19] :) [21:19] balloons, shall i do that in a separate mp? [21:19] balloons, and FYI this is the mp we're waiting to land to resolve the issues... https://code.launchpad.net/~vthompson/music-app/fix-1365247/+merge/233303 [21:19] balloons, but victor just needs to make some mods to it [21:20] ahayzen, I would yes. once it's ready we can ask francis to swap music to python3 only.. then you will have to land that mp before you can land anything else [21:20] so make it it's own thing [21:20] balloons, awesome i'll do that then :) [21:21] balloons, i wanted to add url-handler tests and u said there was another app that did some? can you remember which one it was so i can use them for reference? [21:21] ahh right.. reminders [21:22] balloons, cool thanks [21:33] balloons, are these the only changes i need to make? http://bazaar.launchpad.net/~andrew-hayzen/music-app/autopilot-move-py3/revision/614 [21:34] balloons, the tests themselves have been running py3 on device for some time right? so they should be ok [21:34] yep [21:35] ahayzen, looks good [21:35] balloons, ok i'll propose it..but it'll fail until victor's mp lands [21:36] yes.. it'll fail till python3 is switched [21:36] balloons, cool it is here for reference https://code.launchpad.net/~andrew-hayzen/music-app/autopilot-move-py3/+merge/233431 [21:37] balloons, now i'm gonna try and make some tests for url-dispatcher :) .... is there any way of doing autopilot tests for content-hub yet? [21:49] balloons, ah looks like reminders sends url-dispatcher calls.... we actually receive them are there any examples/docs on how to do this? [21:50] ahayzen, ahh.. well, you can fake recieving it [21:50] elopio, ^^ [21:50] he might have an example for you ahayzen [21:50] balloons, i need to know how to make a call and send it to the app [21:50] dialer-app, gallery-app are examples I would give [21:50] balloons, ah cool thanks :) [21:50] they might have what you need [21:51] balloons, i'll see if i can spot anything [21:56] balloons, there not obvious if they do have them (or i am blind) ;) [21:56] that's all I have off the cuff [21:56] *they're [21:57] hah hmmm [22:07] ahayzen, balloons: you can fake a call with something like url-dispatcher {arg} [22:08] url-dispatcher is a binary that comes with url-dispatcher-tools [22:08] and to make sure that your project is calling url-dispatcher properly, you can use the fake url dispatcher fixture. [22:08] that one comes with url-dispatcher-testability [22:09] elopio, ah so do i just call that directly through subprocess? not really much mocking required? ... oh there is a fixture :) [22:09] elopio, are there any examples of this that i can use for reference? [22:09] ahayzen: for calling, that's it. For receiving, it's a little more work. Like this: [22:09] https://code.launchpad.net/~canonical-platform-qa/dialer-app/fake_url-dispatcher/+merge/230403 [22:09] let me see if I can remember where do we call url-dispatcher. I think in unity. [22:10] elopio, well basically all we need to test is another app going album:///artist/album ... and then we play the correct album (and then the same for file:///) [22:11] ahayzen: here: http://bazaar.launchpad.net/~ubuntu-testcase/ubuntu-autopilot-tests/ubuntu-experience-tests/view/head:/ubuntu_experience_tests/tests/application_life_cycle/test_url_dispatcher.py [22:11] ahayzen: what you need to take care is to make sure that the testability envirnoment is set. [22:11] otherwise autopilot won't be able to introspect the app launched by url dispatcher. [22:12] elopio, awesome thanks :) i'll try and get this working [22:54] popey: if you want you can update clock app in the store [22:54] popey: all MPs are in [22:55] yay [22:56] nik90_: 3.1.83? [22:58] popey: yup [22:58] balloons: can you please upload http://s-jenkins.ubuntu-ci:8080/job/clock-app-click/lastSuccessfulBuild/artifact/out/com.ubuntu.clock_3.1.83_armhf.click ? [23:04] popey, https://myapps.developer.ubuntu.com/dev/click-apps/121/changerequest/ [23:06] balloons, victor is asking if you could check over https://code.launchpad.net/~andrew-hayzen/music-app/ap-helper-refactor-003/+merge/232655 as well? also note that 004 tidies things up a bit more (as noted in the comments) [23:08] balloons: thank you. nik90_ approved [23:08] popey: awesome thnx [23:08] Thank you! [23:08] enjoy [23:08] * popey updates his phones [23:14] ooh, thats changed [23:18] popey, do we still have to reboot after updating things from the store? [23:20] no, you should be able to search [23:21] popey, ah yes ....new clock \o/ .... is this going to be fixed? [23:21] i hope so! [23:21] hah [23:21] * popey sets an alarm for the morning [23:21] nik90_, ooo the header comes down as the bottom edge goes up...when did u add that? [23:22] :D [23:22] recently [23:22] ooh thats nice [23:22] nik90_, hehe... ah that explains that conversation you were having about not being able to set the header title though [23:23] nik90_, is there now way of setting it to 'Alarm' so it doesn't appear blank? [23:25] popey, FYI we are going to fail CI for this next image...but the fix for it is already in our trunk ...so whenever the next store build goes through will fix it (but we are trying to land a few other things as well for this store build if we can) [23:25] popey, there is a bug here if the landing team ask bug 1365247 [23:25] bug 1365247 in Ubuntu Music App "Music app AP tests need to step up to new Mediascanner 2 schema version" [High,Fix committed] https://launchpad.net/bugs/1365247 [23:26] balloons: how long you around? [23:26] ahayzen: is this something you're likely to get in tonight or more likely tomorrow? [23:26] popey, the fix has landed :) ... we were just trying to get a few other mps in as well [23:26] cheeky! ☻ [23:27] popey, hehe but the image is already building so we missed it right? [23:27] * popey triggers a build anyway [23:27] no [23:27] build starts at 3am [23:28] === trainguards: IMAGE 222 building (started: 20140904 22:10) === [23:28] ? [23:28] thats a manually triggered one [23:28] another will start via a cron job at 3am [23:28] ah guess we could get into the auto one then :P [23:28] well, we need balloons to be around to upload, i cant [23:29] popey, well the mp we were trying to push for to get into this build we would like balloons to check anyway... and there would be another mp after that...so guess they could go in the next one as it is unlikely they'll both get reviewed tonight [23:31] popey, so in summary push trunk to the store if you can please :) [23:31] 614 hasn't built yet [23:31] hmmm [23:31] i just triggered it [23:31] ah [23:36] just watched https://www.youtube.com/watch?v=CE8UhcyJS0I which is fascinatingly fun [23:38] looks like balloons is afk, sorry, can't upload tonight [23:39] sergiusens: you about? [23:40] popey, no worries if you can't just don't panic when CI fails :) [23:40] popey: yup [23:40] hey [23:40] upload? [23:40] please http://s-jenkins.ubuntu-ci:8080/job/music-app-click/lastSuccessfulBuild/artifact/out/com.ubuntu.music_1.3.614_all.click [23:40] * sergiusens connects to vpn [23:42] i leave myself connected all day, which probably is bad for some reason [23:43] oooh! [23:43] * popey goes to put the bins out [23:47] sergiusens: problems? [23:47] popey: sorry, just multi pinged at the moment [23:47] one sec [23:48] popey: uploaded; btw music is still in halloween mode [23:48] in case you want to change [23:48] sergiusens, hehe there is a bug for that [23:48] sergiusens, but we are nearly at halloween soooooo ? [23:49] ahayzen: yeah; it's like the christmas lights [23:49] haloween mode? [23:49] oh, blimey, has it been that long! [23:49] popey: screenshots [23:49] popey, the screenshots [23:49] wow. time flies [23:50] popey, yah ... we always seem to have massive UI overhauls WIP so we kept deferring the rescreenshoting [23:50] thanks sergiusens [23:50] np [23:50] sergiusens, thanks :) [23:51] 614 approved [23:51] \o/ [23:51] thanks ahayzen [23:51] popey, sergiusens thanks for the pushing to the store :) [23:51] np. [23:51] teamwork \o/ [23:52] \o/ [23:52] now sleep