=== chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [06:45] Good morning [07:00] Good morning [07:19] jibel: I'm deeply concerned [07:19] pitti, about? === iahmad is now known as iahmad|afk [07:19] jibel: I'm through my morning email, and I'm missing the usual dozen "jenkins failed" mails [07:19] I only got a "fixed" one for python2.7 [07:20] OMGwhatiswrong? [07:22] pitti, do you miss them? [07:22] muchly! no swearing to do this morning [07:23] pitti, I can configure the jobs to send you an email on every run if that makes feel you better? [07:24] oh yes please, more email [07:33] ah, lubuntu is fixed! [08:28] DanChapman, Good morning, lubuntu should be running correctly now [08:29] DanChapman, the problem was that autopilot was running twice because lubuntu now honors etc/xdg/autostart which it didn't few weeks ago [08:29] I don't know if it was a bug before that I naively worked around or if it is a bug now [08:30] anyway, I added lock file to ensure run-autopilot starts only once [08:30] and also retry apt several times if it fails to install additional packages === chihchun is now known as chihchun_afk [09:09] jibel, good morning :-) awesome thank you! well that makes more sense as to why it stopped running, but yes it does sound like it might be a bug === chihchun_afk is now known as chihchun === iahmad|afk is now known as iahmad [11:22] Morning all === _salem is now known as salem_ [13:50] good morning everybody. [13:50] we will have 30 minutes of QA talks: http://youtu.be/Zhu_4xF5aAw [13:50] starting in 10. [13:50] good morning elopio o/ [13:51] hello DanChapman [13:59] pitti, re. secure boot testing, I did an initial test with qemu, a 1GB disk, and automated partitioner failed :/ [13:59] jibel: you mean UEFI? [13:59] jibel: ah yes, I think I remember [14:00] pitti, yes [14:00] jibel: could be that I had to use manual partitioning [14:00] jibel: so, nice test case :) [14:00] jibel: oh wait, 1 GB is not nearly enough -- you need 4 or so [14:00] it allocates half of the drive to the efi partition and there is no room left for the os [14:00] and, starting now http://youtu.be/Zhu_4xF5aAw [14:01] yeah, I'll increase it but it's a base server install, it shouldn't require much space [14:02] jibel: indeed [14:18] mzanetti: so you can't directly access subobjects? what makes the top level widgets special? [14:19] pitti: every file has it's own context [14:19] mzanetti: I don't yet understand why creating these intermediate objects to make properties r/o is helpful [14:19] pitti: while the context gets propagated downwards, it is not propagated upwards [14:19] mzanetti: i. e. if mycat.weight is r/o, why can't I access mycat.internal.weight? [14:19] pitti: you can't do that from the outside of the file [14:20] pitti: for example: [14:20] mzanetti: so you only "see" the top level objects? [14:20] if you use the Cat {} in another file, you can access only the top level stuf [14:20] f [14:20] exactly [14:20] ah, ok; that was missing [14:20] mzanetti: could you make stack.depth an attribute that emits an event when it's updated so that tests can subscribe to that instead of polling ? [14:20] mzanetti: and the findChild() is basically a workaround for this to make them accessible for tests after all [14:20] vila: yes. that's exactly what's happening [14:21] tryCompare() just waits for that signal and compares if the new value is what you want [14:21] pitti: exactly [14:21] mzanetti: \o/ [14:21] mzanetti: thanks; that wasn't clear to me [14:21] pitti: np [14:21] mzanetti: I can't say that the "internal object" property wrapper is the utmost elegance, but good to know that you can if you must [14:22] pitti: yeah. usually I try to pack stuff into existing other items [14:22] pitti: but if you don't have that, you can use some new Item. I suggest QtObject {} in that case because it's the most lightweight [14:22] pitti: as in: it doesn't have anything to paint, Item {} and everything derived from that does [14:32] virtualenv/tox slides: http://goldb.org/talks/2013/canonical-tox [14:32] thanks cgoldberg [14:35] mzanetti: so, when should we use javascript? Is there something that we can't or shouldn't do with the declarative lang? [14:36] elopio: rule of thumb: if you find yourself typing the word function there's a re really good chance you're on the wrong path [14:36] elopio: functions are useful for triggering stuff at components. for example the Cat.feed() [14:36] elopio: but even in there, try to keep the javascript as short as possible [14:36] elopio: try to just set one property and calculate the rest with property bindings [14:37] mzanetti: got it. [14:37] an interesting follow-up session would be best practices for combine c++ with qml [14:37] mzanetti: do you know somebody that can give that presentation? [14:37] elopio: tsdgeos, saviq, or myself I guess [14:38] mzanetti: ok, I'll ping them to get new people participating, and fall back to you :) [14:38] heh [14:38] mzanetti: last thing, can you make your slides public please? [14:38] ah, sure [14:39] elopio: done [14:39] thank you! [14:41] mzanetti: I'm sorry, what would be the link? [14:41] elopio: https://docs.google.com/a/canonical.com/presentation/d/1O6lLjxgCT-kXwFAszvIN2bwG8ROIVpzm3dBvYZhiY9Y/edit#slide=id.g17ad694ca_038 [14:51] elopio: your dialer-app pep8 fix still fails; apparently there's some longer-time uninstallability of Qt5 in trusty-proposed :/ [14:51] pitti: yes, I've just read the email. :( [14:52] pitti: thanks for pushing it again. Should I talk with CI about the error? or have you done that already? [14:52] elopio: well, it's not a CI problem [14:53] /var/local/autopilot/setup.log: dialer-app : Depends: qtdeclarative5-ubuntu-telephony0.1 but it is not going to be installed or [14:53] /var/local/autopilot/setup.log: qtdeclarative5-ubuntu-telephony-plugin but it is not installable [14:53] elopio: that's a problem in the packaging of either dialer-app, or the telephony qt stuff [14:54] well, I was hoping CI to find the culprint and somebody to fix it. Maybe it's too much to ask from them :D [14:54] * pitti checks http://people.canonical.com/~ubuntu-archive/proposed-migration/update_output.txt [14:55] ah, so evolution-data-server and folks seem to break it [14:55] *culprit [14:57] elopio: see #u-devel [14:57] thanks pitti. [15:04] cgoldberg: nice talk. note that the venvs that tox creates can be activated and deactivated too. very handy [15:04] you can create new environments [15:04] pitti: ^^ [15:05] barry, thanks! right.. once created you can go to .tox/ , and see your envs and activate/whatever [15:05] i do that all the time :) .tox/py33/bin/activate etc... [15:05] * barry should really package up detox [15:06] sitepackages=True [15:06] cgoldberg: ^^ [15:06] detox is super cool. i'm surprised it's not just a switch on tox, instead of another package [15:06] agreed [15:07] re: sitepackages vs pypi. it's actually a bit problematic for debuntu. i'm in conversation w/upstream about things that would make tox nicer for distros [15:07] great talks, thanks! === chihchun is now known as chihchun_afk [16:43] barry: +1 in packaging detox :) [16:47] barry: would you be interested in doing a small talk about python 2 - 3 incompatibilities and how to solve or avoid them? [16:47] elopio: sure [16:48] barry: :) at what time do you EOD? [16:49] elopio: well, probably not *today* :) long usa holiday weekend coming up. but generally eod at 2300utc if my math/dst is correct :) [16:49] that's perfect. [16:50] barry: I'll look for the other speaker and get back to you to see a date. Maybe in two or three weeks, at a time close to your eod. [16:50] elopio: sounds good! [16:50] thank you [17:27] jibel, does xterm need to be open during the test? an assertion error keeps popping up ('The following apps were started during the test and not closed: %r', [, ]) which is strange considering xterm is open long before ubiquity launches [17:28] that error also suggests that xterm is top of the window stack aswell :-/ [17:41] Ursinha, i see that the phone bug people are running into the Bug Control Only features that they need to set bugs triaged... xD [17:41] oop i might've mishighlighted IDK.. my logs went away :/ [17:41] if i mishighlighted ignore :) [18:32] guten abend [18:49] DanChapman, no you can remove it, it was just a convenient to have a reliable terminal which always works in a VM and is rarely affected by rendering or performance issues [19:18] testing pidgin [19:19] ...and that is a pass on the manual test :) [19:22] balloons: is there going to be a ubuntu cadence schedule or are you just doing the exploratory thing [19:22] evening :-) [19:22] and hi ;) [19:22] just exploratory testing. We've got works items to spruce up the page a bit and to communicate it more [19:23] ok - thanks - just needed to know if you had a schedule at all - we'll roll our own :) [19:24] so yea, nothing beyond the milestones.. and for ubuntu there are no milestones before beta [19:24] actually be doing a bit of both 'cadence' and exploratory [19:24] yep [19:24] instead as we spoke about in the call for testing event, I wanted to push letting us collectively schedule testing events [19:24] so between that and exploratory testing, aka dogfooding, which many already do each cycle, I think we're covered [19:25] I'd like to just talk about being effective while dogfooding, etc [19:25] well to be honest I'm afraid I want to get on talking to our team - we've a lot of new stuff planned to land which needs to have some sort of schedule QA wise [19:26] not got time to wait for others to decide what they're going to collectively do ;) [19:30] balloons, transmission-gtk launches fine for me, what autopilot and transmission versions do you have? [19:31] Noskcaj, let me check [19:31] Autopilot Source Version: 1.4.0 Autopilot Package Version: [19:31] 1.4+14.04.20131125-0ubuntu1 [19:32] i'm pretty sure that's what i have too [19:32] autopilot launch -i Gtk transmission-gtk Gtk-Message: Failed to load module "autopilot" [19:46] jibel, lovely thanks mate, I'll just comment it out in the MP i have waiting atm. Easy enough to re-enable if needed that way [19:46] Noskcaj, good evening. I have attached a screenshot to bug 1252438. can you have a look if you see the search box on it? [19:46] bug 1252438 in Ubuntu Manual Tests "lubuntu software center testcase - no search box" [Undecided,New] https://launchpad.net/bugs/1252438 [19:48] Noskcaj, are you on saucy or trusty? also, what output do you get in running transmission? [19:49] btw: is software center crashing for you guys, too? [19:49] on trusty? [19:50] dkessel, crash on start or ? [19:50] yup... window shows up, then window becomes unresponsive... [19:50] balloons, trusty, http://paste.ubuntu.com/6486135/ [19:50] Noskcaj, really? that's wild [19:51] dkessel, yes, same [19:51] balloons, ok [19:51] dkessel, lsc has the search bar for me, but not for you [19:51] Noskcaj, strange... [19:52] Noskcaj, could it be... longer translations for some of the stuff on the top? so there is no more space for the search box? [19:52] dkessel, That would make sense, have you tried fullscreen? [19:53] Noskcaj, dang.... there it is... [19:53] lol! [19:53] still... many users wouldnt try that [19:54] no, i'll add a mention to try fullscreen and push that now [19:54] k thanks [19:58] Is 1619 in the testcase bzr branch? [19:59] don't know... I found it on the package testing tracker [20:00] It's not, i've just fixed this on the tracker though [20:00] * dkessel browses code [20:00] yeah... missing [20:05] anybody idling around who has had some contact with me: I plan to apply for ubuntu membership. If you want to help me, leave a nice testimonial on my wiki page: http://wiki.ubuntu.com/dkessel ;) [20:14] dkessel, Can i suggest you add you launchpad page to the contact info [20:14] https://wiki.ubuntu.com/phillw is still what i model my wiki page off [20:15] Noskcaj, good idea. done [20:16] dkessel, i have seen your nick around this channel - hi! i would suggest you extend that page even further, which i'm sure you've planned to do :) [20:19] knome, ok. I see... yes I'd better do that... :) [20:20] basically any link to actual work you have done is good [20:20] balloons, hi [20:20] Letozaf_, hello :-) [20:21] balloons, I fixed yesterday's test, I need to ask something about music-app, must I ask Victor ? [20:34] Letozaf_, did you push it to see that it would pass? [20:35] balloons, wow, your question makes we wonder something went wrong, I thought I pushed it, let me check [20:36] balloons, I will push again now, something didn't work [20:37] ahh.. did you commit first, hehe [20:37] balloons, I think I did, let me check [20:39] balloons, done, now it's pushed I checked [20:39] balloons, sorry I thought I had it pushed :P [20:48] Letozaf_, still building [20:48] Letozaf_, anyways, feel free to send a mail or post in the MP [20:48] Letozaf_, if you want victor specifically, request him at the top :-) [20:49] balloons, ok thanks [20:49] * Letozaf_ got fingers crossed [20:55] balloons, yay, it passed :D [20:58] good work! [20:59] balloons, thanks! [21:02] Letozaf_, there's a few tweaks you can do to the code in looking at it [21:02] switch_to_tab will return the tab object so you don't have to get it :-) [21:02] and then you can drop the assert as well [21:03] balloons, ok fine I will correct the code [21:03] balloons, I was a bit too used to putting asserts everywhere [21:05] Letozaf_, yes we can go without in many causes now thanks to the changes in ap 1.4 and by using the emulator :-0 [21:05] balloons, cool, just have to get used to it :P [21:06] Letozaf_, also I'm not sure eventually makes sense in these: self.assertThat(albumartist.text, Eventually(Equals(artistName))) [21:06] nothing is being updated as far as I can tell [21:07] I would drop those 2 asserts as well, and if there is a timing thing, add it into the get functions instead [21:07] make sense? [21:08] balloons, thinks so, let me make the changes and push it so you can check [21:08] balloons, well first I will check it still runs [21:08] sure.. I mention this little things to make sure it doesn't cause us trouble once we're running on the phone [21:09] but it's awesome you checked it as well :-) [21:09] balloons, yes thanks, as I told you I was used to put asserts everywhere, I will have to learn to not do it [21:11] balloons, but the second self.assertThat(albumartist.text, Eventually(Equals(artistName))) I put it to check the album sheet opened is the correct one [21:12] balloons, is that wrong, I mean, how do I know I did not open another sheet [21:13] balloons: was there a chance that the custom views of http://reqorts.qa.ubuntu.com/reports/qa/qa-touch.html is likely to gain any traction? [21:14] elfy, yes.. It's an idea [21:14] as with everything, needs someone to do it :-) [21:15] but i'm going to gather a bit more data [21:15] yea - I get loads of ideas :( [21:15] bu then can't do it so forget about it till next time I think of the same thing [21:23] balloons, the test passed before I could run it on my phone :) looks like it's ok, I left the second assertion for checking I had opened the right album sheet, is that ok ? [21:24] Letozaf_, looking [21:25] Letozaf_, looks better. [21:25] Letozaf_, did you want victor to review or ? [21:25] balloons, yes [21:26] so Letozaf_ https://code.launchpad.net/~carla-sella/music-app/show_albums_sheet/+merge/196785/+request-review [21:26] add him :-) [21:27] balloons, done! [21:27] knome, I guess my wiki page looks better now? [21:28] balloons, at least I think it's a good idea that Victor reviews the test, right ? [21:29] Letozaf_, yea.. you should add the bugs you are closing as well [21:29] victor created the bugs :-) [21:30] dkessel, yep, better already [21:30] balloons, ok thanks [21:31] dkessel, I added something to the bottom as well [21:33] ty balloons === salem_ is now known as _salem === thomi_ is now known as thomi