[00:01] alesage: veebers: https://code.launchpad.net/~canonical-platform-qa/ubuntu-sanity-tests/apt_update/+merge/249430 [00:02] I'm going to the gym. bbl. [03:38] veebers: I made comments on the branches. [03:39] as before, maybe you can reply before leaving today so we can take over if needed, but following your preference. [03:40] elopio: ack, just finishing off a review now then will review your comments [03:40] elopio: and by branches you mean my one? Or are there others that you would like me to review too? [03:40] oh, right, one is actually mine :) [03:40] https://code.launchpad.net/~canonical-platform-qa/ubuntu-sanity-tests/apt_update/+merge/249430 [03:41] https://code.launchpad.net/~canonical-platform-qa/ubuntu-sanity-tests/selenium-ppa/+merge/249416 [03:41] elopio: awesome cheers, stacked and ready [03:54] elopio: you still around? [03:54] veebers: yes. [03:55] elopio: sweet, just quickly with no lag, is there a reason why we can make the update_apt arg to run_sanity_tests the last arg and give it a default False value? [03:56] i.e. run_sanity_tests(conf_name, tests_to_run, update_apt=False) [03:57] veebers: I made it the argument of the center because I think one would usually call ubuntu-sanity-tests -c test.conf -u test_suite.test_module.test1 [03:57] and I didn't add a default value for it because Argparse takes care of that. [03:57] veebers: but if you prefer to change the order and put a default value, I'll do it. No harm in that. [03:58] elopio: hmm, right but you've had to update all the tests to consider the new argument [03:58] (which is why I initially ask) [03:58] veebers: right. I could remove that part of the diff if it had a default value. [03:58] I'll do it. [03:59] elopio: right, to me it's an indication to me that it's changed something outside of it's own influence [04:00] elopio: regarding whoudl we even do this change, yeah I see the reason (something outside our influence caused it and it's not done under cover, the tester has to specificy it as a workaround) [04:01] elopio: re: the ppa branch, I'm not sure I follow with moving it to a file? [04:03] veebers: ok, going back a little, changing the order and using a default argument I removed one of the test changes. [04:03] coolio, yeah sorry to jump around on you while you're changing things :_) [04:03] the other are still needed, because we are testing the list of arguments received from the command line. And False will be there even if we don't pass -i [04:03] sorry, -u [04:04] I pushed it. [04:09] sweet [04:48] elopio: rats, i was really close to suggesting a sweet cleaner way of mocking Popen, but came across this bug :-\ https://bugs.launchpad.net/python-fixtures/+bug/1358085 [04:48] Launchpad bug 1358085 in Python Fixtures "FakePopen does not work in Python 3.3+" [High,Fix released] [04:48] This would have used fixtures.FakePopen instead of a mocked solution === chihchun_afk is now known as chihchun [05:40] veebers: sorry, there was a black out. [05:47] elopio: 0_0 no worries, hope all is well (I'm not actually around any more) [05:48] veebers: yes, no worries. Some quiet time without electricity is always nice. Enjoy your long weekend. [05:49] heh ^_^. will do, thanks === spineau is now known as spineau_afk [08:50] Hi guys [08:51] I’m looking for a way to calculate some stats [08:51] I need to research how often packages are updates for LTS [08:51] Ideally with distribution by day [08:51] Just wondering if ther eis a way to get such data somwhere === spineau_afk is now known as spineau === spineau is now known as spineau_afk === vrruiz_ is now known as rvr === spineau_afk is now known as spineau === chihchun is now known as chihchun_afk === elfy is now known as hobgoblin === [BNC]elfy is now known as hobgoblin [13:22] and here too gbb :D [13:23] 8(): [13:26] balloons: gbb is a new tester, has been doing some for us, and it was her who discovered that testcase tracker bug last night [13:34] elfy: About test cases: I'm wasn't clear what the difference between 'entire disk' and 'autosize' is. (I'm guessing autosize expects to co-exist with another OS?) Could there be a little extra blurb on top of the test case to explain this? I think perhaps my test case bug about autosize may have been wrong, since I think I did an entire disk test there if I understand this correctly. [13:35] thanks - this is for more general holistic QA stuff :) [13:35] ok, bear with me, this is all new and thus confusing %-) [13:35] you do guess right [13:35] yea :) [13:35] xubuntu-devel is xubuntu only :) [13:36] and it's actually only been in the last couple of years that testing seems to have had much talk in there [13:36] * gbb admits she isn't quite sure of the difference. Other than that that xubuntu is of course superior ;-P [13:37] I'm not sure that there is any info anywhere specifically about the different types of install - that's more a general how to install type thing [13:37] gbb: of course ;) [13:37] https://help.ubuntu.com/community/GraphicalInstall [13:37] for example [13:39] I suppose that a line could be added to image testcases making it more obvious [13:40] gbb: balloons is the Canoncial Community Manager who deals with us QA types - he's really quite friendly, make the case for this to him :) [13:40] he's in Pacific time - will be about a bit later on [13:40] hello. [13:40] hi elopio [13:41] A little precis (one line) on top of the test cases that explains what we're trying to do would be useful in general tho. Just to set the scene. [13:41] hi elfy. [13:42] gbb: yea I understand what you're saying - balloons is the one to make the case too really, I'd likely be the one who did the donkey work on it :) [13:45] gbb: looking at them, the resize test does mention that you'll need another installed system and sufficient disk space for 2 systems [13:46] back in a bit [14:11] elopio: so you are unable to click on 'buy now' while running locally? [14:18] dobey: yes. [14:19] when you remove the skips and make autopilot, do you get all tests passing? [14:19] elopio: i don't have a utopic vm set up to do that in. so i've not been able to run the tests locally [14:20] elopio: afaik though, alecu ran them and they passed for him, as he wrote those tests [14:20] dobey: why utopic? I ran on vivid, is that ok? [14:20] elopio: well i don't have a vivid vm either [14:21] you are living in the past :) [14:21] elopio: i'm running lts as my host. [14:21] dobey: if I understand correctly, after clicking buy the window should be closed. [14:21] elopio: even if i were running vivid as my host, i'd still need a vm. i don't want tests fiddling with my actual home director/dbus/etc [14:22] that's not happening. But the button is gettting clicked. [14:22] elopio: yes, it should "buy" the thing [14:22] well the button isn't receiving the click event [14:22] otherwise the logs would show more info [14:23] ie, it would have a debug message of something like "Purchase requested: blah blah" [14:25] dobey: where are the logs stored when I make autopilot? [14:25] elopio: do they not go to stdout/stderr? [14:27] only autopilot's log. [14:27] ah, wait, I think you are right. [14:28] 2015-02-12 08:24:46,468 - DEBUG - Emit signal certificateFound [14:28] that seems to come from the app. [14:28] yeah [14:28] and yes, no events logged after the click. [14:28] and then nothing after that [14:29] which means something is probably stealing the click and it's not getting through to the app === josepht_ is now known as josepht === pgraner-afk is now known as pgraner [15:00] dobey: putting some more logging here. The one that doesn't get executed is the onBuy in payui.qml. launcPurchase in CheckoutPage.qml is executed. [15:05] hmm [15:08] dobey: If I'm getting it right, accountView.currentItem is null [15:08] elopio: about https://code.launchpad.net/~paulliu/unity8/notification_helper/+merge/249211 [15:08] elopio: I wondering why we need a get_data() function. [15:08] so accountView.currentItem.email fails and the pageCheckout.buy is not executed. [15:08] elopio: The Notification will disappear and the data might get cleared. [15:09] elopio: Shound't we backup those data? [15:09] funny silent error, I dont' like you javascript. [15:09] paulliu: hi. [15:09] paulliu: I see. [15:10] paulliu: so, I think that in that case, it would be better to make wait_for_notification [15:10] return the data. not the object. [15:10] elopio: ok. got it. [15:12] def wait_for_notification(self, timeout): [15:12] notify_list = self.main_window.select_single('Notifications', objectName='notificationList'); dialot = notify_list.wait_select_single('Notification', visible=True); return{ 'summary': dialog.summary,'body': dialog.body, [15:12] 'iconSource': dialog.iconSource } [15:12] paulliu: makes sense? [15:12] oh, I missed your previous ping. okay then :) [15:13] elopio: yeah. But I'll change it with a timeout? not using wait_select_single. [15:14] elopio: or we don't need a timeout. [15:14] paulliu: ah, right, with wait_select_single the timeout is not needed. Well, there's one but the default is alright. [15:14] I prefer wait_select_single. [15:31] gbb, elfy hello :-) [15:37] hi balloons [15:44] elopio: oh right, ok, that makes sense. hrmm [15:50] elopio: ok, well i think i have an idea how to fix that, but i think it would best be done in another branch, so i think we should leave those skipped for now and i'll file a bug about that [15:54] i wonder why qml isn't spitting an error to the console for that though [15:54] thanks for finding that elopio [15:54] elopio: i filed a bug for that: https://bugs.launchpad.net/pay-ui/+bug/1421277 [15:54] Launchpad bug 1421277 in pay-ui "Autopilot tests fail due to missing account model" [High,Triaged] [16:07] balloons: do we need both of these pages? or would you be happy to see an amalgamation of them -https://wiki.ubuntu.com/Testing/TestCaseFormat and https://wiki.ubuntu.com/QATeam/ContributingTestcases/ManualStyleGuide [16:08] elfy, it's a little confusing, but we do dual purpose on the https://wiki.ubuntu.com/Testing/TestCaseFormat page [16:09] yes I know - but does that matter? [16:09] we talk about what testcases look like, but also mention what they look like for contributing [16:09] the manual style guide as it where, is an attempt to get folks to get the naming, wording and format correct [16:09] dobey: ok to me. [16:09] I would say you could change that page a bit easier than the other [16:10] dobey: my only question remaining for your branch is the return '' [16:10] effectively that one is just a rehash of info from the testcaseformat page [16:10] I can't see a logical reason not to add the 'extra' from manualstyleguide to the other one [16:11] but - whatever, just wanted to ask the question before doing anything :) [16:15] elopio: i'lll have to look at that one again. i really need to get lunch though. so will respond to that after [16:16] and i have been getting pinged like crazy this morning, so didn't get to concentrate on that [16:25] dobey: that's alright. You have already done all the parts I was worried about :) Thanks a lot for that. [16:25] I'll copy some of your stuff to the webapps. [16:33] fgimenez: like [16:34] + from ubuntuuitoolkit import fixture_setup [16:34] + self.useFixture(fixture_setup.InitctlEnvironmentVariable( [16:34] + global_=True, [16:34] + UBUNTU_WEBVIEW_DEVTOOLS_HOST=ip, [16:34] + UBUNTU_WEBVIEW_DEVTOOLS_PORT=str(port))) [16:34] the global might not be needed. [16:38] elopio: ok, instead of calling initctl with subprocess, right? [16:38] elopio: initctl set-env [16:38] fgimenez: it will do the same, but takes care of the cleanup. [16:38] elopio: ok thanks! === roadmr is now known as roadmr_afk === roadmr_afk is now known as roadmr [18:28] elopio: ok, just replied to your comments on the MP. finally. got pinged a bit upon return from lunch, too [18:34] dobey: +1 === roadmr is now known as roadmr_afk [20:05] balloons, hello [20:05] Letozaf_, hello [20:06] balloons, I found the solution to solve this mp's problem: https://code.launchpad.net/~nikwen/ubuntu-filemanager-app/compressed-tars/+merge/249250 [20:06] balloons, but I do not have permission to push the changes [20:07] Letozaf_, awesome. You can ask him to merge your changes into his branch [20:07] you can even create an mp for it, just change the target from trunk to his branch [20:08] balloons, ok fine I will do that thanks [20:08] yw, thanks for finding the issue :-) [20:09] balloons, my pleasure :) [20:19] dobey: do you know what we would have to access the internet fromt the qemu test bed? [20:19] is there an easy command to share the host network? [20:21] elopio: i'm pretty sure it works already, given that adt-run installs stuff with apt [20:23] ok, let me give it another try. I might be doing something else wrong here. === roadmr_afk is now known as roadmr [21:33] balloons, this is weired, I wanted to install ubuntu-filemanager-app on my desktop, but even if I got core apps PPA cannot find it to install [21:34] balloons, I've searched ubuntu-filemanager-app filemanager-app but nothing