[00:57] balloons: you gone awol again? [04:15] Good morning [04:15] qengho: oh, how not? I had assumed it would mostly compile for GTK2 as well, perhaps it needs some #ifdefery? [04:15] qengho: the gtk2 plugin should go into a new binary package, though [06:54] good morning === Noskcaj10 is now known as Noskcaj === Ursinha-afk is now known as Ursinha === iahmad is now known as iahmad|afk === LordOfTime is now known as LordOfTime|EC2 === e11bits_ is now known as e11bits [08:41] Good morning all :-) [08:58] good morning all === iahmad|afk is now known as iahmad === psivaa_ is now known as psivaa === JoseeAntonioR is now known as jose === forestpi1kie is now known as forestpiskie === forestpiskie is now known as Guest70885 === Guest70885 is now known as hobgoblin === hobgoblin is now known as forestpiskie === brendand_ is now known as brendand === pleia2_ is now known as pleia2 === netcurli_ is now known as netcurli [15:38] balloons: hi. is everything ok with http://iso.qa.ubuntu.com/? I'm getting "The resource you tried to access doesn't exist. This can happen because of an invalid link or because of a bug." [15:38] slickymaster, ohh.. heh, I would say not [15:39] balloons: I was doing the ISO tests for the Alpha 2 and a moment ago, after submitting a result a got that message [15:39] stgraber, ^^ it appears the qa.u.c domain is down [15:40] balloons: applying security updates [15:40] should be back in a couple of minutes [15:40] stgraber, :-) [15:40] stgraber: balloons thanks, the both of you [16:40] elopio, hmm, having trouble finding ubuntu-ui-toolkit to install.. weird [16:40] ahh it's been renamed :-) [16:41] DanChapman_, dkessel hello [16:41] ubuntu-ui-toolkit-autopilot ? [16:41] balloons, good evening [16:42] balloons: renamed to what? [16:42] elopio, yea.. it wasn't finding it at first.. [16:42] i was trying to install a meta package like ubuntu-ui-toolkit and that wasn't working === om26er is now known as om26er_afk [16:44] balloons, howdy. How u doing? [16:45] balloons: do you know about pushing an app from the qtcreator to the phone? [16:45] I press ctrl+f2, but get ssh: connect to host 127.0.0.1 port 2222: Connection refused [16:45] oooh nice new ubuntu edge price options =) [16:46] elopio, yes from inside qtcreator right? [16:47] balloons: yes. [16:47] dkessel, yes, new options, but act fast if you want a lower tier :-) [16:48] elopio, that should work, I haven't tried it in awhile.. hmm.. does it show your devices under devices? [16:49] ahh, there's a big button there saying enable developer mode [16:49] maybe that's what I was missing. [16:50] elopio, so I'm looking at the final results for the emulator.. and I saw the pep8 stuff and renames.. we ready for primetime use? [16:50] balloons: not yet. The rename hasn't landed. [16:50] elopio, yes, but once the rename has landed [16:51] balloons: yes. And you can start using it now, just rename to ubuntuuitoolkit tomorrow. [16:52] elopio, well I'm going to announce some fanfare, so I'll prep everything and probably just announce tomorrow once it's complete [16:52] I'll probably convert an app today to demo it [16:54] DanChapman_, doing well.. As you can see elopio's hard work on making a proper emulator for qt has landed ;-) [16:54] balloons: sounds good. Let me know if you find any bugs. [16:54] elopio, I'm hoping it'll fix some of the bugs I've introduced :-) [16:54] hehe [16:55] balloons: well, at least, it's now easy to make tests for the bugs we find. [16:56] this is true.. I introduced a weird bug, but it might be on the autopilot side.. it's just hard to know.. but now, we can easily tell [16:58] balloons: and I've just noticed I am assuming there will be just two tabs visible. But if the label of the tab is small, more than two can appear on the header. [16:59] that will bite soon. [16:59] yes, there are apps with 3. it's not uncommon [16:59] why is that? === om26er_afk is now known as om26er [17:01] because I'm naive :) I assumed you would be able to switch just from the current tab to the next, without skips. [17:02] elopio, ahh.. I believe my tab switching logic just scrolls through things.. and you can stop at a tab index or tab label [17:02] so you can in theory create a spin of death where you go through things endlessly because there is no match haha [17:02] yes, I copied that funny feature :) [17:03] we could in theory stop once it's gone round once and raise an error [17:03] balloons: but well, make sure to mention on your post that it's a first version in heavy development :) [17:03] elopio, ofc ofc.. but bugs and such can be reported, and it can be updated sanely, etc [17:04] yes, and I think we also can factor out some other common things. [17:04] like a base test case class, things that are the same on all the apps. [17:06] and the final feature that ties everything together is what zoltan and jppiiroi1en are doing on the qt creator. It now has a template for the autopilot tests :D === Ursinha is now known as Ursinha-afk [17:16] balloons, thats cool will check it out. I have basically re-written the ubiquity tests the last few days :-) it should sort the fails on jenkins. just finishing off custom partition then will move on to dual-boot === DanChapman_ is now known as DanChapman [17:21] elopio, the autopilot template is in? wahoo? [17:22] DanChapman, I've been watching the results.. a bit mixed as you say, so I hope the changes solve that [17:27] balloons, it helps now i have the test runner jibel created. makes it easier to find the bugs. If your not too busy would you be able to give the tests a once over for me?? Could do with some fresh eyes on it :-) [17:28] DanChapman, I have a moment now sure :-) [17:28] balloons, cheers. I havn't pushed custom part up yet though https://code.launchpad.net/~dpniel/ubiquity/autopilot [17:29] big branch :-) [17:32] still going . . . [17:32] lol :-) [17:33] the numbers keep going up [17:33] yeah it ends some where near 120000 i think [17:33] I should have updated the branch I had already pulled.. lol.. a couple more mins I think I 'll have it [17:34] Ive done that a few times myself :-) [17:36] and done [17:37] lol [17:37] k, so what specifically should I look at? [17:38] ohh.. you should note in your readme kubuntu is special -- it has it's own installer and your stuff won't work for that [17:38] DOes it say kubuntu in it? [17:40] right the biggest problem i keep coming across is grub install fail or the installer crashes. which usually occurs during the progress bar. Can you look at the track_install_progress_bar func. Do you think thats a reasonable way of handling it? [17:40] it doesn't.. it just says all other flavors.. which is true, except for kubuntu ;-_) [17:40] lol i will update it. [17:40] so I'm curious that the ubiquity_did_install_complete test and this assert doesn't timeout normally: self.assertThat(self.complete_dialog.visible, Eventually(Equals(1))) [17:41] k, looking at the progress func [17:42] balloons are you wondering about the sleep? [17:43] ok so your throwing a message, but then have an assert about the dialog.. I'm confused at first glance what your trying to do [17:43] balloons, you're :P [17:43] brilliant.. knome is going to fix me yet [17:44] why just today knome I noticed myself and corrected it! [17:44] nice! [17:44] the education works ;) [17:44] Thats more for me at the moment as when using the test runner locally you get ALL the output from the env and not just autopilot output so i stuck the message so i could spot that the func had picked up the dialog. I will remove it in the end :-) [17:44] DanChapman, ahh I see you sit in ubiquity_progress_bar_test before hitting ubiquity_did_install_complete [17:45] and it has a 300 second timeout (might not even be long enough, heh) but yea, that was my expectation of a long timeout [17:46] ok, so I will say why watch the progress bar? Why not just wait for the dialog to appear? Are we concerned about it hanging and we don't know it or ? [17:47] I would just wait for the install complete dialog to appear. And if we want to "watch" in the interim, grab the progress and assert that the progress reported changes withing X amount of time [17:47] can we also reliably detect if something goes awry -- aka a dialog window other than the install complete appears/ [17:50] DanChapman, ^^ [17:50] I did think of just looping waiting on the confirm dialog. But following the progress kinda shows that its updating ok. [17:51] The 300 timeout is a bit extreme as it just waits for the progress to go back to 0.0 for the install part. [17:51] oh it's just the copying files piece I see.. not the install [17:53] I've had no probs with the progress bar and tracking it. Its trying to catch an error dialog while stuck in the loop. As the loop never exits if it doesn't catch it. [17:53] DanChapman, I'm missing where self.install_progress.fraction is coming from [17:53] yea you have your while progress < complete: loop that won't end right? [17:55] install_progress selects the progress bar in ubiquity_progress_bar_test [17:56] ahh.. there it is, ty :-) [17:57] ok far enough.. so we loop while waiting for complete. Each iteration we update the progress, and then check for errors. [17:58] so let's dig into those error checks.. you have checks for visible windows of failures and then breaks to leave the loop [17:58] and your finding that the loop isn't ending on failures because of the if statement check for the error is never becoming true, or it's not breaking properly? [17:59] if the "visible" property isn't working out so well, can we just check from existence of the object, or do they exist from the beginning? [18:01] its catching the error dialogs 'ok'. Just dnt know if there is a better way to do it. All the objects are there from the start apart from the grub install failed dialog which is created once the error occurs. hence the it being in a try [18:05] i did think to try just getting the visible window count like in the FF test and if window count increased then it would fail but i couldn't figure out how to do that. and it would have a very meaningful reason for failing as we wouldn't know what it failed on [18:06] *wouldn't [18:09] from an autopilot point of view that is. I know the install log would give details, but it would be good to see that the correct dialog appears for the correct error i suppose. [18:09] Heh, testing the error windows too :-) [18:10] yea, so since the object exists, visible works well enough. I just don't like to rely on it, but perhaps that's not well placed [18:30] balloons, just run the default test a few times and got a grub install failed and it caught it ok. So do you think leave it as it is and see how it goes. In theory it shouldn't happen often if at all anyway :-) [18:31] DanChapman, I think the idea is sane enough, and I like monitoring it.. I would add a timeout check though I think if nothing happens for X amount of time, aka the progress bar doesn't move [18:32] we don't want to loop forever without us knowing about it.. and yea, that should work [18:35] balloons, i like it. Will add a timeout check [18:38] Will sort that tomorrow. I'm calling it a day. Catch you all tomorrow [18:40] DanChapman, catch you on the flipside === wxl_ is now known as wxl === Ursinha-afk is now known as Ursinha [19:18] good evening Letozaf_ [19:18] balloons, Hi [19:19] did you see the rss reader stuff landed -- it's a bit different than before :-_ [19:20] balloons, no didn't see it, I must take a look :) read you did something [19:24] balloons, by the way did you see the sudoku tests ? [19:25] Letozaf_, what's up with them? [19:25] balloons, nothing :) I hope [19:25] lol [19:26] balloons, where can I see the rss reader stuff ? I was looking in my branch and the rssreader one... [19:27] balloons, what happened to it ? [19:36] Letozaf_, what do you mean what happened? [19:36] balloons, you said it's different ... [19:37] Letozaf_, ohh yes, the main trunk has new code in it [19:37] including the theme :-) [19:38] balloons, oh I just did a bzr branch, but there is nothing in there [19:38] balloons, :( [19:38] bzr branch lp:ubuntu-rss-reader? [19:39] balloons, maybe I did something wrong: bzr branch lp:ubuntu-rssreader-app [19:39] balloons, yep [19:39] balloons, oh wait I did bzr branch lp:ubuntu-rssreader-app [19:39] yea, 22 revisions [19:39] you had it right.. I mispelled it ;-) [19:40] balloons, but the test is only 27 lines long in it [19:40] Letozaf_, anyways, once you branch you can see the add and edit feed stuff are a little different [19:41] balloons, ok I will re-branch it [19:42] balloons, I get 20 revisions ? [19:42] balloons, bzr branch lp:ubuntu-rssreader-app [19:42] balloons, that's weired [19:42] Letozaf_, should be 22 revs [19:42] hmm [19:42] balloons, did it twice, cancelled the directory and branched from scrap [19:43] Letozaf_, lol, you've got it then, sorry [19:43] i have post branch commits in mine [19:43] whoops! [19:43] balloons, :D [19:43] I checked http://bazaar.launchpad.net/~ubuntu-rssreader-dev/ubuntu-rssreader-app/trunk/revision/20 [19:44] balloons, I'm curios to see what you guys did to it :P [19:44] looks much nicer [19:48] balloons, I was looking for the next app to write test for ... for instance the music-app, but once you create a playlist, how do you put music in it ? [19:48] balloons, :? [19:48] Letozaf_, vthompson is one of the music app authors, and I believe dkessel wrote one of the tests for it? [19:49] anyways, there are some tests in there, so it shouldn't be too hard to hack on it [19:49] on your question however, I'm not sure off the top of my head [19:49] we can always ask the app authors :-) [19:49] balloons, I will take a look thanks [19:50] balloons, oh yes sure [19:51] balloons, but do you think I can work on it or is dkessel doing it ? [19:51] balloons, you know that as soon as I pick something someone else has done it already :P [19:55] Letozaf_, i am not working on any more music app tests at the moment [19:57] Letozaf_, indeed dkessel wrote just the initial test.. no one will steal your thunder on this one.. I would contact victor though. .would you like his email? [19:58] dkessel, thanks for the info [19:58] good night [19:58] balloons, ok in case I need to ask him somethin [19:58] dkessel, good night [19:58] night dkessel ! [20:14] morning [20:17] balloons, Do you know if anyone is going to be able to look after the testdrive hackfest between 1600 and 2000 [20:22] Noskcaj, no takers on the thread right? [20:22] I'm still concerned about having it run for more than a few hours [20:22] ok. Maybe i but it back to only 6 hours? [20:25] I would rather see an amazing couple hours as opposed to having 6 hours with 4 hours of downtime.. But yea, it's up to you. I wouldn't run it for longer than you can be around. Getting commitment from others to show up is excellent too, but make sure your around. The idea is to kickstart new folks on development as well as close bugs right? [20:25] BTW, you should send a message to ubuntu-devel as well once everything is in place [20:25] I saw your target work, so plan on running it for as long as it would take to finish that off, with some buffer [20:26] assuming of course you think you can get through the whole list :-) [20:26] remember you can always have another event ;-) [20:26] Noskcaj, I'm not remembering but how long is it going to run for atm? [20:28] balloons, 11 hours, so i'll cut that back so howard and i can run it ourselves [20:28] Noskcaj, yea, that sounds like the way to go [20:31] good stuff mate [20:32] elopio, so I'm having trouble using your emulator.. perhaps you can shed some light on my python newbie blunders [20:32] balloons: of course. Tell me what you are doing. [20:34] elopio, well I'm having trouble creating a new instance of an emulator class.. For instance, toolbar [20:35] ie, I want to do something like emulators.Toolbar.click_button( "newgamebutton"). I've imported I hope sanely; from UbuntuUiToolkit import emulators [20:37] balloons: first, you need to launch the application with the emulator. [20:38] * balloons facepalms [20:38] with the base emulator. Let me search for an example. [20:38] elopio, I get it I think [20:39] self.app = self.launch_test_application( [20:39] "/usr/lib/" + arch + "/qt5/bin/qmlscene", [20:39] "-I", get_module_include_path(), [20:39] qml_path, [20:39] emulator_base=emulators.UbuntuUIToolkitEmulatorBase) [20:40] elopio, yes, so more or less the __init__.py template for core apps needs to be updated to reflect this as well [20:40] as atm, it's handling all the launch code [20:41] let me see what I can do ;-) [20:41] balloons: yes. I think it would be better to have launch helpers on ubuntu-ui-toolkit. [20:41] so we have just one place to update. [20:42] hmmm hmm [20:42] also that part of get_input_device, and get_module_include_path. I see them everywhere. [20:42] ohh yea, it's horride [20:43] ok, so perhaps you can just help directly with this.. so I'm picking on dropping letters to convert over.. it doesn't have any tests, just the old layout. I was going to convert and add a test or two as the example [20:44] balloons: ok, and once you have the application launched with the emulator loaded, it's like: [20:44] self.app.select_single(emulators.MainView) [20:45] for the main view I take it? ;-) and then select and have fun from there? [20:46] balloons: yes. with the main view you can make open_toolbar, for example. [20:46] balloons: I'll propose a base test case, that can be inherited by all the apps. === Ursinha is now known as Ursinha-afk === qengho_ is now known as qengho