| === _salem is now known as salem_ | ||
| === salem_ is now known as _salem | ||
| === asac` is now known as asac | ||
| mzanetti | Saviq: morning | 07:21 |
|---|---|---|
| mzanetti | Saviq: jenkins finally approved this: https://code.launchpad.net/~unity-team/unity8/launcher-backend/+merge/173189 | 07:21 |
| mzanetti | so if you're still happy with it I think we can try to land it | 07:22 |
| mzanetti | MacSlow: hey | 07:43 |
| mzanetti | MacSlow: all the notification tests fail for various reasons here | 07:43 |
| mzanetti | MacSlow: but not the one you said :D | 07:43 |
| MacSlow | mzanetti, did you kill notify-osd? | 07:44 |
| mzanetti | MacSlow: I started a new dbus session | 07:45 |
| mzanetti | MacSlow: the notification shows up just fine | 07:45 |
| mzanetti | MacSlow: but then everything starts going wrong | 07:45 |
| MacSlow | mzanetti, I have never seen an issue with the opacity | 07:46 |
| MacSlow | mzanetti, I'm not starting a new dbus-session... just getting rid of notify-osd | 07:47 |
| mzanetti | MacSlow: no... I don't want to reboot my machine every time I run your tests :D | 07:48 |
| mzanetti | MacSlow: the d-bus session is just fine... the notifications show up as expected | 07:48 |
| mzanetti | btw... Frank Zappa looks way better than the name would suggest | 07:48 |
| MacSlow | mzanetti, :) | 07:48 |
| MacSlow | mzanetti, I had to use the stock assets :) | 07:49 |
| mzanetti | :D | 07:49 |
| MacSlow | mzanetti, so you just started the notification ap-test with "autopilot run unity8.tests.testnotifications"? | 07:50 |
| mzanetti | MacSlow: autopilot run unity8.tests.testnotifications.TestNotifications.test_sd_incoming_call | 07:50 |
| mzanetti | but yeah | 07:50 |
| MacSlow | mzanetti, ah ok... just the single one... I see. | 07:51 |
| mzanetti | MacSlow: the shell comes up, it unlocks the greeter and the incoming notification shows up. | 07:51 |
| mzanetti | MacSlow: then it times out because it waits for the opacity to become 1 while it stays at 0.99.. | 07:51 |
| * mzanetti believes that might be related to the layer.enabled: true thingie | 07:52 | |
| * mzanetti tries | 07:52 | |
| MacSlow | mzanetti, I get some warnings/failures too... but not any issue with opacity (http://pastebin.ubuntu.com/5860909) | 07:53 |
| MacSlow | mzanetti, I'll try on my laptop | 07:54 |
| mzanetti | MacSlow: hmm.... Indeed adding print()s to onOpacityChanged shows it never goes to 1. | 07:54 |
| Saviq | mzanetti, re: launcher-backend yeah, I rebuilt it to get the actually fixed shell for testing with packages on device | 07:54 |
| mzanetti | MacSlow: and also, I can see the header from the dash behind shining through | 07:54 |
| Saviq | mzanetti, will test now and skim through again | 07:54 |
| mzanetti | Saviq: cool, thanks | 07:55 |
| mzanetti | MacSlow: I changed the check to be Eventually(GreaterThan(0.999999)) | 07:57 |
| mzanetti | seems to work | 07:57 |
| mzanetti | MacSlow: so now I can reproduce your fail | 07:57 |
| mzanetti | MacSlow: this one? | 07:58 |
| mzanetti | self.assertThat (self.action_send_message_triggered, Equals(True)) | 07:58 |
| MacSlow | mzanetti, your issue with the opacity-check is odd still... it also works on veebers machine like that | 07:58 |
| mzanetti | MismatchError: True != False | 07:58 |
| MacSlow | mzanetti, yeah... that's the issue with the callback not touching the bool flag | 07:58 |
| mzanetti | MacSlow: yeah... its odd... no idea why it doesn't really go to 1 | 07:58 |
| mzanetti | Saviq: any idea perhaps?^ | 07:59 |
| mzanetti | MacSlow: actually... the heightBehavior.enabled=true would never be called in my case then | 08:00 |
| veebers | MacSlow, mzanetti: I'm not sure it's an autopilot issue. I tried a really simple example in a script and couldn't get my callback called | 08:00 |
| MacSlow | mzanetti, doh... still need to pull updates on my laptop... still seeing the "black button"-problem there | 08:00 |
| MacSlow | veebers, you think it's a general python-issue then? | 08:00 |
| sil2100 | veebers: hi! Big thanks for fixing the autopilot issue ;) | 08:00 |
| Saviq | MacSlow you're using the Ubuntu easings etc.? | 08:01 |
| MacSlow | veebers, mzanetti: should we try a newer python? atm 2.7 is used iirc | 08:01 |
| veebers | MacSlow: or I'm not constructing the callback et. al correctly :-P | 08:01 |
| veebers | MacSlow: shouldn't matter | 08:01 |
| MacSlow | Saviq, for the anims... yes. | 08:01 |
| sil2100 | I guess you guys are talking about the latest failures with opacity checks? | 08:01 |
| Saviq | MacSlow, mzanetti I suspect the same issue we found in the SDK yesterday | 08:01 |
| veebers | sil2100: no worries, this was the failing tests one yes? | 08:01 |
| mzanetti | ah... | 08:01 |
| MacSlow | Saviq, like what issue? | 08:01 |
| Saviq | https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1199550 | 08:01 |
| ubot5 | Ubuntu bug 1199550 in Ubuntu UI Toolkit "Panel.animating doesn't return to "false"" [Undecided,Fix committed] | 08:01 |
| mzanetti | Saviq: good point | 08:01 |
| Saviq | MacSlow the bezier easing added in rev 600 of the toolkit http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/trunk/revision/600 | 08:02 |
| Saviq | MacSlow, seems to not reach 1.0 | 08:02 |
| Saviq | MacSlow, some rounding error | 08:02 |
| MacSlow | veebers, well there's one "difference" between the autopilot tests and the plain python-regression-tests... | 08:02 |
| Saviq | MacSlow, ideally you should look to the animation's running prop, not compare reals | 08:03 |
| Saviq | MacSlow, mzanetti although I would consider that a bug that an animation that's supposed to reach 1.0 doesn't because of some interpolation error | 08:03 |
| Saviq | Kaleo wasn't convinced yesterday, but I'll have a chat with him about this later today | 08:04 |
| MacSlow | veebers, mzanetti: since we had to do this "from gi.repository import GLib, Notify" I had to change the parameters passed to add_action() ... needs two additional arguments now otherwise it fails | 08:04 |
| mzanetti | Saviq: yeah... but the panel.animating never becoming false might be caused by the same... some thing never reaching the final state because of the same | 08:04 |
| sil2100 | Saviq: to me that's an error, especially that we sometimes assume it reaches 1.0 | 08:04 |
| Saviq | sil2100, exactly, you expect it to | 08:04 |
| Saviq | mzanetti, it definitely is (fixed in toolkit trunk) | 08:05 |
| sil2100 | (like with opacity, suddenly causing failures in many autopilot tests ;p) | 08:05 |
| Saviq | http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/trunk/revision/601 | 08:05 |
| MacSlow | Saviq, ok. | 08:05 |
| Saviq | s/fixed/workarounded/ | 08:05 |
| mzanetti | Saviq: yeah... that doesn't seem like a proper fix | 08:06 |
| Saviq | mzanetti, well, it fixes the issue at hand, but not all the others | 08:07 |
| mzanetti | Saviq: it fixes one of the symptoms, but not the cause | 08:07 |
| Saviq | mzanetti, yup, what I meant | 08:09 |
| mzanetti | now... what's a bit more of a problem is that I replaced the UbuntuNumberAnimation with a NumberAnimation and I still see the issue | 08:10 |
| veebers | MacSlow: ugh, I can get it working if I look at these examples: http://hashbang.fr/tutoriel-notify.html | 08:11 |
| veebers | MacSlow: it's using Gtk.main() at the end | 08:11 |
| FernandoMiguel | morning | 08:11 |
| FernandoMiguel | anyone from the am working in unity 7, that can help me debug a bug on 13.10 ? thanks | 08:11 |
| MacSlow | veebers, hm... in the regression-tests I'm using a dedicated Gtk-mainloop too | 08:11 |
| MacSlow | veebers, but I did assume that I don't need that in autopilot-tests | 08:12 |
| MacSlow | veebers, http://bazaar.launchpad.net/~unity-api-team/unity-notifications/trunk/view/head:/examples/interactive-notification.py#L65 | 08:14 |
| MacSlow | veebers, http://bazaar.launchpad.net/~unity-api-team/unity-notifications/trunk/view/head:/examples/sd-example-incoming-call.py#L102 | 08:14 |
| MacSlow | veebers, just for reference | 08:14 |
| Saviq | MacSlow, mzanetti can you check the output from http://pastebin.ubuntu.com/5860941/ | 08:15 |
| veebers | MacSlow: hmm thanks. I might have to ask thomi re: that one tomorrow | 08:15 |
| MacSlow | veebers, ok... at least we're a bit closer now to sort it out | 08:15 |
| veebers | MacSlow: aye | 08:15 |
| mzanetti | Saviq: yeah... thats the cause | 08:16 |
| mzanetti | Saviq: stays at 0.999999999634 | 08:16 |
| MacSlow | Saviq, "true 0 | 08:16 |
| MacSlow | false 1 | 08:16 |
| MacSlow | " | 08:16 |
| mzanetti | MacSlow: you need to upgrade to the latest ubuntu-ui-toolkit to reproduce the bug | 08:17 |
| mzanetti | MacSlow: well, don't... | 08:17 |
| MacSlow | mzanetti, :) | 08:17 |
| mzanetti | Saviq: here's the exact output: | 08:17 |
| Saviq | mzanetti, this could even cause some unnecessary blending | 08:18 |
| mzanetti | true 0 | 08:18 |
| mzanetti | false 0.9999999999361473 | 08:18 |
| mzanetti | Saviq: yes, it does. I can see the background of the notifications shining through | 08:18 |
| Saviq | mzanetti, that's expected, actually ;D | 08:18 |
| mzanetti | Saviq: huh? | 08:18 |
| Saviq | mzanetti, they're see-through - that's per-design | 08:18 |
| MacSlow | mzanetti, certainly... at least in the "black" area | 08:19 |
| mzanetti | Saviq: If I change all the UbuntuNumberAnimations to just NumberAnimations its not the case any more | 08:19 |
| mzanetti | well, I'm talking about the icon mostly | 08:19 |
| MacSlow | mzanetti, only buttons, text and icons are meant to be opaque | 08:19 |
| Saviq | mzanetti, I doubt it's visible (I doubt 8 bits can represent the difference) | 08:19 |
| Saviq | when rendered into ARGB | 08:19 |
| mzanetti | right... I just checked more carefully again.. its still there | 08:20 |
| mzanetti | but tests are passing again | 08:20 |
| Saviq | mzanetti, if you can see through the avatar, for example, that's a visual bug | 08:20 |
| mzanetti | Saviq: I do | 08:20 |
| mzanetti | but only with the black icon... not with the Frank Zappa lady :D | 08:21 |
| MacSlow | mzanetti, Saviq: maybe another newly introduced issue with the SDK? I don't have that here on my desktop-machine... let me look at the version of the installed SDK | 08:21 |
| Saviq | mzanetti, what should I be looking at? | 08:21 |
| mzanetti | MacSlow: actually... why does the same test, the incoming call one most of the times have the picture of that woman, but sometimes some grey abstract thingie? | 08:22 |
| MacSlow | mzanetti, ? | 08:22 |
| MacSlow | mzanetti, never seen that issue | 08:22 |
| * mzanetti tries to reproduce | 08:23 | |
| mzanetti | yeah... just happened again | 08:23 |
| mzanetti | running this "autopilot run unity8.tests.testnotifications.TestNotifications.test_sd_incoming_call" | 08:23 |
| mzanetti | most of the times I get that lady in the icon. Sometimes only some grey icons with slightly orange stripes in it. thats when the background shines through... | 08:24 |
| mzanetti | Saviq: ^ | 08:24 |
| Saviq | mzanetti, sil2100, MacSlow bug #1199662 | 08:24 |
| ubot5 | bug 1199662 in Ubuntu UI Toolkit "New easing causes animations not to reach their target values" [Undecided,New] https://launchpad.net/bugs/1199662 | 08:24 |
| MacSlow | Saviq, mzanetti: for what it's worth... I've 0.1.46+13.10.20130709-0ubuntu1 of qtdeclarative5-ubuntu-ui-toolkit-plugin installed here on my desktop | 08:24 |
| Saviq | mzanetti, which branch btw? | 08:24 |
| mzanetti | Saviq: ~macslow/unity8/notification-autopilot-tests/ | 08:24 |
| jamesh | Saviq: I'm looking at putting together a scope to expose recent apps on the phone (something that is currently handled via custom UI). Do you know if the API it uses to get the list of running applications is available outside of the shell process? | 08:26 |
| Saviq | tsdgeos, http://pastebin.ubuntu.com/5860963/ | 08:26 |
| tsdgeos | Saviq: yeah | 08:27 |
| tsdgeos | no clue where that warning comes from tbh | 08:27 |
| tsdgeos | had a look at the define | 08:27 |
| tsdgeos | and it has just one param :/ | 08:27 |
| Saviq | tsdgeos, k | 08:27 |
| Saviq | jamesh, it is / could be, but I don't believe that's what you should be looking at | 08:27 |
| Saviq | jamesh, I believe we should use zeitgeist there | 08:27 |
| tsdgeos | Saviq: i'll have a second look | 08:28 |
| Saviq | jamesh, while a specialized renederer in the shell would pick up the snapshots | 08:28 |
| Saviq | jamesh, running apps are going to be killed arbitrarily to save resources | 08:28 |
| Saviq | jamesh, so your "Recent apps" would start disappearing randomly | 08:29 |
| tsdgeos | but yeah, all QSKIP in Qt code are like that | 08:29 |
| jamesh | Saviq: I remember you saying that, but when I was talking with thostr_ he said that the design still calls for it to show running apps :( | 08:29 |
| tsdgeos | ./auto/corelib/io/qfile/tst_qfile.cpp:1900: QSKIP("Not enough space to run test"); | 08:29 |
| Saviq | jamesh, that's because they don't have a clue what a running app is ;) | 08:29 |
| Saviq | jamesh, i.e. there can't be a notion of a running app if we're killing and restarting them all the time | 08:30 |
| Saviq | tsdgeos, k, throw a #pragma at it maybe at some point | 08:30 |
| Saviq | jamesh, but we talked about this yesterday | 08:30 |
| Saviq | jamesh, in the dash sync | 08:30 |
| Saviq | jamesh, and I think the agreement is that we should use zeitgeist | 08:30 |
| Saviq | jamesh, granted, you need to be able to "close" such a recent app | 08:31 |
| jamesh | okay. | 08:31 |
| Saviq | jamesh, and that would actually kill the app, if it's running, but otherwise it should just disappear from the recent ones, that's it | 08:31 |
| jamesh | so I guess we still need an API to identify the apps that are currently running? | 08:31 |
| nic-doffay | Saviq, since the OptionSelector has so many variations for content, should the delegate be set externally be the user? | 08:31 |
| nic-doffay | *by | 08:31 |
| Saviq | jamesh, the shell does, the scope doesn't | 08:31 |
| Saviq | jamesh, but please confirm with thostr that's where we're goinf | 08:32 |
| Saviq | going | 08:32 |
| jamesh | okay | 08:32 |
| Saviq | jamesh, both launching and closing apps should be events in zeitgeist, I think | 08:32 |
| Saviq | jamesh, and you should just look at that and compile a list that were recently launched but not yet closed | 08:33 |
| Saviq | jamesh, ordering might be tricky, though - we might need to put focus events in, too :/ | 08:33 |
| Saviq | jamesh, as in "app focused" event to order by | 08:34 |
| Saviq | now it does feel like zg is maybe not the best place | 08:34 |
| Saviq | jamesh, greyback we might need to build a volatile stack of "apps launched, but not closed" | 08:35 |
| jamesh | okay | 08:35 |
| Saviq | jamesh, greyback is working on unity8 integration with Mir where all of this will happen | 08:35 |
| greyback | Saviq: ok. I'd expected as much | 08:36 |
| Saviq | greyback, that's for "running apps", that aren't necessarily running anymore (lifecycle) | 08:36 |
| Saviq | greyback, do you think it'd make sense for jamesh to build a scope that would talk to the current observer on Surfaceflinger | 08:37 |
| Saviq | greyback, or would the new API be different enough that he should target the new api straight away? | 08:37 |
| greyback | Saviq: yep. I'd assumed App Manager will have list of all apps running or suspended or killed-with-resume-data | 08:38 |
| greyback | Saviq: thinking... it doesn't really make sense to do the SF work, if it'll be thrown away within a few weeks. | 08:41 |
| Saviq | greyback, I agree | 08:41 |
| Saviq | greyback, will app manager know the difference between app-with-resume-data and app-with-no-resume-data? | 08:42 |
| greyback | Saviq: that I have to learn. | 08:42 |
| Saviq | jamesh, greyback can you please sync up on those topics? | 08:42 |
| greyback | Saviq: it will be the one sending the kill signal, so I expect so | 08:42 |
| greyback | jamesh: hey, let's have a meeting on this, so we can swap info | 08:43 |
| jamesh | greyback: sure. I currently don't have any code for this, so there isn't any legacy concerns if there is a new API in development | 08:45 |
| greyback | jamesh: indeed. | 08:46 |
| greyback | jamesh: meeting scheduled, does hte time suit you? | 08:49 |
| jamesh | greyback: looks good. | 08:52 |
| greyback | jamesh: cool | 08:52 |
| mzanetti | Saviq: quick'n'easy one: https://code.launchpad.net/~mzanetti/unity8/fix-1199622/+merge/173883 | 08:53 |
| Saviq | mzanetti, obviously the "Critical" there was tongue-in-cheek ;) | 08:53 |
| Saviq | mzanetti, I'd never have noticed (nor did our designers, so...) | 08:53 |
| mzanetti | yeah... | 08:53 |
| mzanetti | Saviq: but funnily Cimi noticed the same for the separator line, but not for the icon | 08:54 |
| Saviq | mzanetti, yeah remember that :) | 08:54 |
| Cimi | mzanetti, oh michael | 08:55 |
| Cimi | mzanetti, off topic | 08:55 |
| mzanetti | hehe | 08:55 |
| Cimi | mzanetti, a friend told me that the ubuntu icon is flipped | 08:55 |
| mzanetti | Cimi: haha | 08:55 |
| Cimi | mzanetti, is he right? don't have the phone running now | 08:55 |
| mzanetti | just fixed that | 08:55 |
| Cimi | ok | 08:55 |
| Cimi | good | 08:55 |
| mzanetti | Cimi: https://plus.google.com/u/0/photos/104438618743851678614/albums/5898788268517651553/5898788275502369234?cfem=1&pid=5898788275502369234&oid=104438618743851678614 | 08:55 |
| sil2100 | andyrock: hello! | 09:17 |
| andyrock | sil2100, hey | 09:18 |
| sil2100 | andyrock: do you know what's the status with compiz on saucy? Can we re-enable it for daily release? | 09:19 |
| andyrock | sil2100, nope... there is a bad regression | 09:19 |
| sil2100 | Oh? | 09:19 |
| andyrock | sil2100, let me find the bug number | 09:19 |
| andyrock | sil2100, https://bugs.launchpad.net/compiz/+bug/1198000 | 09:20 |
| ubot5 | Ubuntu bug 1198000 in compiz (Ubuntu) "Regression: r3751 causes all unmaxed windows to shrink both horizontally and vertically on each subsequent opening, probably by exactly the size of the decoration" [Critical,Triaged] | 09:20 |
| andyrock | we can revert that revision | 09:20 |
| andyrock | but than we have another regression with qml5 apps | 09:20 |
| andyrock | *qt5 | 09:20 |
| sil2100 | oh shit | 09:21 |
| sil2100 | Ok, so there's a deadlock right now, hm - I see no one is assigned here | 09:21 |
| andyrock | i'm trying to fix it, but decor.cpp is not that simple :D | 09:22 |
| sil2100 | hohoho | 09:23 |
| sil2100 | Ooooh no it isn't ;) I had my share of that file | 09:23 |
| mhr3 | andyrock, sil2100, also i can't build trunk compiz because of some gmock issues | 09:23 |
| sil2100 | andyrock: ^ ? Did anything change regarding that? | 09:25 |
| andyrock | sil2100, mhr3 nope, maybe there is some cmake issue with xorg-gtest? | 09:25 |
| mhr3 | well, i saw gmock fixes proposed for nux and unity.. so definitely something wrong there | 09:26 |
| mhr3 | sil2100, btw is unity still failing on the dbus issues? | 09:28 |
| andyrock | mhr3, yeah but I still can build compiz trunk here | 09:28 |
| mhr3 | andyrock, from clean tree? on s? | 09:28 |
| andyrock | mhr3, let me clean it | 09:28 |
| andyrock | yeah on S sure | 09:29 |
| sil2100 | mhr3: not sure, didn't see the DBus hang this week, but there are a lot of tests failing anyway | 09:29 |
| mhr3 | sil2100, well that is the dbus hang | 09:30 |
| sil2100 | It is? But it's only around 21 failures per platform and all seem to be preview tests | 09:30 |
| mhr3 | sil2100, all those other super-hangs were caused by the recordmydesktop issue | 09:30 |
| sil2100 | MismatchError: After 10.0 seconds test on DashView.preview_displaying failed: True != dbus.Boolean(False, variant_level=1) | 09:30 |
| sil2100 | These are because of DBus? | 09:31 |
| sil2100 | http://10.97.0.1:8080/job/autopilot-saucy-daily_release/387/testReport/ <- that's why I wanted someone to take a look at yesterdays failures, since to me those looked like normal failures | 09:32 |
| sil2100 | But I might be mistaken | 09:32 |
| mhr3 | sil2100, i see preview failures, hud failures, search timeouts, bamf failures, all dbus stuff | 09:33 |
| FernandoMiguel | https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1199696 | 09:34 |
| ubot5 | Ubuntu bug 1199696 in unity (Ubuntu) "SIGSEGV, unity-panel-service crashes and eats a core" [Undecided,New] | 09:34 |
| FernandoMiguel | if anyone from unity7 can look it up | 09:34 |
| mhr3 | sil2100, is there any way to re-run the testing with bustle-pcap running? | 09:35 |
| mhr3 | sil2100, but probably just for one run, cause 50minute bustle-pcap session will produce a file that will be a few hundred mb | 09:35 |
| === dandrader is now known as dandrader|afk | ||
| sil2100 | mhr3: hm, I guess we could do that, but not sure how to do it only for one run, although we might modify the configuration of otto temporarily and then resume it back to the way it was | 10:22 |
| sil2100 | mhr3: good to know it's all caused by dbus issues | 10:23 |
| sil2100 | mhr3: since really, the logs don't say anything about that | 10:23 |
| sil2100 | At least not explicitly | 10:23 |
| mhr3 | sil2100, guess we can just connect to the lxc once the testing starts and run it manually | 10:23 |
| === dandrader|afk is now known as dandrader | ||
| mhr3 | sil2100, although i see that libunity on powerpc failed | 10:24 |
| sil2100 | Yes, and one scope, looking into that | 10:24 |
| sil2100 | (for i386) | 10:24 |
| === hikiko is now known as hikiko|lunch | ||
| sil2100 | Need to re-try the jobs | 10:25 |
| mhr3 | sil2100, ping me once the ap testing will be about to start i'll try to do the bustle capture | 10:25 |
| mhr3 | jibel, if i want a log file to survive and be exported as a result in jenkins is it enough to just put it some special dir? | 10:27 |
| sil2100 | mhr3: ok - what about the libunity failure for powerpc? I saw some unit test failing I think... although I also think we're not really caring about the powerpc status | 10:27 |
| sil2100 | mhr3: I think it's sufficient if it land in some non /tmp directory | 10:27 |
| sil2100 | We can fetch if from the diff then | 10:27 |
| mhr3 | sil2100, looks like some odd race, just re-run | 10:27 |
| sil2100 | k | 10:29 |
| tsdgeos | Saviq: https://bugreports.qt-project.org/browse/QTBUG-32338#comment-209093 and https://codereview.qt-project.org/#change,60809 | 10:33 |
| jibel | mhr3, put it in /var/local/autopilot/ | 10:34 |
| jibel | and it will be attached as artifact under results/autopilot/ | 10:34 |
| mhr3 | jibel, great, thx | 10:34 |
| tsdgeos | Saviq: nic-doffay: any reason this is not top-approved? https://code.launchpad.net/~saviq/unity8/improve-openeffect/+merge/173752 | 10:40 |
| sil2100 | mhr3: re-ran the stack, will poke you once the check job gets pushed to otto | 10:42 |
| mhr3 | sil2100, cool thx | 10:42 |
| sil2100 | mhr3: ok, autopilot job started | 10:43 |
| sil2100 | mhr3: http://10.97.0.1:8080/job/autopilot-saucy-daily_release/396/ | 10:43 |
| nic-doffay | Saviq, figured you might want someone else to take a look. But I'll do it now. | 10:51 |
| === 6JTAA2M02 is now known as tvoss | ||
| === MacSlow is now known as MacSlow|lunch | ||
| === hikiko|lunch is now known as hikiko| | ||
| === hikiko| is now known as hikiko | ||
| Saviq | nic-doffay, this is a simple enough change, and easy to test (did you get to test on a Nexus 10, where the slowness is most visible?) | 11:17 |
| Saviq | nic-doffay, but it's mostly about assuring that you see no regressions | 11:18 |
| Saviq | tsdgeos, awesome | 11:19 |
| Saviq | tsdgeos, src/quick/items/qquickgridview.cpp only has whitespace change, is that allowed in Qt patches? | 11:21 |
| tsdgeos | Saviq: hmmm, hmmm | 11:21 |
| * tsdgeos looks at the code | 11:21 | |
| Saviq | tsdgeos, scratch that | 11:22 |
| tsdgeos | ok | 11:22 |
| Saviq | tsdgeos, I just can't read the gerrit diff | 11:22 |
| tsdgeos | i moved the code outside the if | 11:22 |
| Saviq | tsdgeos, yeah | 11:22 |
| Saviq | tsdgeos, too bleak colors there :) | 11:22 |
| tsdgeos | he he | 11:23 |
| nic-doffay | Saviq, I didn't test on a Nexus 10 which is part of the reason I held off. | 11:27 |
| === alan_g is now known as alan_g|afk | ||
| Saviq | nic-doffay, you don't have one, right? | 11:27 |
| nic-doffay | Saviq, negative. | 11:27 |
| Saviq | tsdgeos, mzanetti you guys have a n10? | 11:28 |
| mzanetti | Saviq: I do | 11:28 |
| tsdgeos | i do | 11:28 |
| Saviq | mzanetti, could you look at https://code.launchpad.net/~saviq/unity8/improve-openeffect/+merge/173752 quickly | 11:28 |
| mzanetti | Saviq: sure | 11:28 |
| Saviq | tsdgeos, if you want to, try that↑, too | 11:28 |
| Saviq | it's a simple fix for nasty UX :) | 11:28 |
| tsdgeos | well, it'll take ages | 11:28 |
| tsdgeos | have to charge, flash, try :D | 11:29 |
| mzanetti | same here ^^ | 11:29 |
| Saviq | lol | 11:29 |
| mzanetti | at least I already have the flipped image on it... so maybe a reflash is not required | 11:30 |
| Saviq | guys, it only takes like 10 minutes to download & flash here ;P | 11:30 |
| Saviq | and anyway most of the download is common between devices, so :P | 11:31 |
| mzanetti | Saviq: yeah... but charging it enough to survive a flash takes a bit longer | 11:31 |
| Saviq | mzanetti, :P | 11:31 |
| mzanetti | Saviq: nah... 10 mins should be ok | 11:31 |
| sil2100 | mhr3: any luck? | 11:32 |
| mhr3 | sil2100, did it finish yet? | 11:33 |
| sil2100 | mhr3: still running - did you start bustle-pcap? | 11:34 |
| mhr3 | sil2100, yea, it's half a gig already :) | 11:35 |
| sil2100 | ohshit | 11:35 |
| sil2100 | jibel: hope there's enough space for that? | 11:35 |
| sil2100 | ^ | 11:35 |
| mhr3 | that's why didn't want to do it on every run :) | 11:35 |
| === alan_g|afk is now known as alan_g | ||
| jibel | sil2100, enough space for what where ? :) | 11:40 |
| sil2100 | jibel: additional logs from bustle-pcap - it weights over 0.5GB already! ;) | 11:42 |
| sil2100 | (just for one run) | 11:42 |
| Saviq | tsdgeos, some of the filters are taken from unity-2d, so © Kaleo 2011 might be right :) | 11:42 |
| Saviq | ®©™ | 11:42 |
| tsdgeos | ahhh | 11:42 |
| * Saviq loves compose :D | 11:42 | |
| jibel | sil2100, it will be fine on the test host but will maybe causes some problem with the publication to jenkins.qa.u.c | 11:43 |
| sil2100 | jibel: can we not publish that one build? | 11:44 |
| Saviq | tsdgeos, I just love how small the Qt patch is :) | 11:44 |
| tsdgeos | yep :-) | 11:44 |
| jibel | sil2100, if it's running it's too late | 11:44 |
| Saviq | tsdgeos, it might actually get in upstream :D | 11:44 |
| sil2100 | booo... | 11:45 |
| Saviq | tsdgeos, will setting null / undefined to those props call reset*? | 11:45 |
| tsdgeos | that is a good question | 11:46 |
| tsdgeos | i was wondering what uses the RESET property function | 11:46 |
| tsdgeos | tbh not sure | 11:46 |
| tsdgeos | that's why i added the Q_INVOKABLE to reset* | 11:46 |
| tsdgeos | let me see if undefined indeed calls the reset* and then don't need the Q_INVOKABLEs | 11:47 |
| === dandrader is now known as dandrader|afk | ||
| === MacSlow|lunch is now known as MacSlow | ||
| Saviq | tsdgeos, btw, should we allow only setting one of them? I know it doesn't make much sense... and printing debugs will just spew errors out on normal usage due to lack of sync | 12:07 |
| * Saviq wants tuples ;) | 12:07 | |
| mzanetti | Saviq: improves things a little yes. | 12:12 |
| mzanetti | Saviq: there is still a small hickup when it loads the image for the first time | 12:12 |
| mzanetti | Saviq: but once that happened, its perfectly smooth now | 12:12 |
| Saviq | mzanetti, yeah, the hiccup... hickup? hikkup? stupid speller | 12:13 |
| Saviq | mzanetti, it's the Image / UbuntuShape | 12:13 |
| Saviq | mzanetti, we have the same in the dash views now when whole categories are created at a time | 12:13 |
| Saviq | mzanetti, we should try and isolate that to either and make sure we fix it if it's the shape | 12:14 |
| pete-woods | can anyone point me in the direction of the media player app's QML plugin for the MPRIS API? | 12:15 |
| Saviq | I knew it was "hiccup", both en_US and en_GB, stupid aspell or whichever | 12:15 |
| pete-woods | I'm told it uses the C API generated from http://bazaar.launchpad.net/~unity-team/libunity/trunk/view/head:/src/unity-sound-menu-mpris.vala | 12:15 |
| pete-woods | but I just can't see it in the mediaplayer-app | 12:15 |
| Saviq | pete-woods, jhodapp and renato would be the best PoCs | 12:15 |
| pete-woods | Saviq: thanks | 12:15 |
| Saviq | pete-woods, they might not be here yet | 12:15 |
| pete-woods | Saviq: okay, seems everyone I need today is in the wrong timezone :) | 12:16 |
| === greyback is now known as greyback|lunch | ||
| * mzanetti wonders if its actually a good idea that we install app-specific plugins (not intended for usage in other apps) int QT_QML_DIR | 12:20 | |
| didrocks | mzanetti: not sure what QT_QML_DIR is set to, but it should be some kind of libexec/<app_name>/(plugin?) as a private library dir | 12:21 |
| mzanetti | didrocks: QT_QML_DIR is /usr/lib/<arch>/qt5/qml/ | 12:22 |
| mzanetti | didrocks: basically where Qt stuff itself is too | 12:22 |
| didrocks | yeah, should s/qt5/<appname>/ IMHO | 12:22 |
| mzanetti | didrocks: yeah. I agree | 12:22 |
| didrocks | if it's not going to be reused by multiple apps | 12:22 |
| Saviq | mzanetti, didrocks +1, what we did for the shell | 12:24 |
| Saviq | mzanetti, which plugins are you thinking? | 12:24 |
| mzanetti | Saviq: PhoneApp, Browser, CameraApp... | 12:25 |
| mzanetti | Saviq: bascially everything in there except Qt stuff itself and Ubuntu.Components | 12:25 |
| mzanetti | and a few other exceptions | 12:25 |
| mzanetti | lide Dee probably | 12:25 |
| mzanetti | like | 12:25 |
| mzanetti | ShareApp might be an exception too... depending on whats in there | 12:26 |
| Saviq | mzanetti, then yeah, +1 | 12:29 |
| Saviq | mzanetti, they should be installed in private under $LIBDIR/<app_name>/qml | 12:29 |
| * mzanetti just realized that the terminal-app uses KDE's konsole :) | 12:31 | |
| mzanetti | tsdgeos: ^ | 12:31 |
| === pete-woods is now known as pete-woods-away | ||
| pete-woods-away | the terminal-app could do with exiting when you type exit! :p | 12:34 |
| sil2100 | mhr3: the AP tests finished | 12:34 |
| mhr3 | sil2100, yep, downloading the log | 12:35 |
| sil2100 | mhr3: 511 MB ;) | 12:35 |
| === pete-woods-away is now known as pete-woods | ||
| kgunn | Saviq: ping | 12:42 |
| Saviq | kgunn, pong | 12:43 |
| === dandrader|afk is now known as dandrader | ||
| tsdgeos | mzanetti: yep | 12:56 |
| Saviq | mzanetti, add =2 to pkg_check_modules(LAUNCHER_API REQUIRED unity-shell-launcher) maybe? | 13:01 |
| mzanetti | Saviq: is that really a question? | 13:01 |
| Saviq | mzanetti, you decide :D | 13:02 |
| mzanetti | dude | 13:02 |
| mzanetti | :SD | 13:02 |
| Saviq | mzanetti, it'll fail if the version is different, so yeah, probably makes sense :) | 13:02 |
| mzanetti | yes. I agree | 13:02 |
| * mzanetti fixes | 13:02 | |
| === greyback|lunch is now known as greyback | ||
| mzanetti | Saviq: pushed | 13:08 |
| Saviq | mzanetti, thanks | 13:08 |
| mhr3 | sil2100, ehm, so great, the bustle tools can't really handle huge captures | 13:20 |
| mhr3 | $ bustle --count dbus.bustle | 13:20 |
| mhr3 | Stack space overflow: current size 8388608 bytes. | 13:20 |
| mhr3 | oh and yes, it was using 6gb of ram | 13:21 |
| Saviq | lol | 13:23 |
| sil2100 | eeh | 13:25 |
| sil2100 | So we're again with nothing? Or can you | 13:25 |
| sil2100 | read something from what we got? | 13:25 |
| sil2100 | (or we got nothing) | 13:25 |
| === zoktar_ is now known as zoktar | ||
| === broder_ is now known as broder | ||
| Saviq | mzanetti, hmm =2 didn't work | 13:32 |
| mzanetti | so... who is going to take notes... Saviq or Cimi? | 13:32 |
| dednick | Saviq: jenkins approve!! | 13:33 |
| mzanetti | Saviq: hmm. I did a test run here... printed "found launcher (version 2)" | 13:33 |
| Saviq | dednick, nice | 13:33 |
| Saviq | mzanetti, yeah, change it to 1 or 3 | 13:33 |
| Saviq | mzanetti, says that it found (version 1) etc. | 13:33 |
| mzanetti | huh? stupid pkgconfig | 13:33 |
| Saviq | Cimi, standup? | 13:34 |
| === pete-woods is now known as pete-woods-lunch | ||
| Saviq | mzanetti, ok, so - if you can get pkgconfig to fail if !=2, that's the last thing I have there | 13:46 |
| Saviq | BOOM | 13:57 |
| tsdgeos | Saviq: do we really not need any ppa anymore? | 14:00 |
| Saviq | tsdgeos, not for build dependencies, for runtime... potentially, but those should start popping up in distro | 14:00 |
| Saviq | tsdgeos, we might need to add ppa:ubuntu-unity/next later, though | 14:01 |
| tsdgeos | :D | 14:01 |
| Saviq | tsdgeos, for stuff that's not yet distro-ready | 14:01 |
| tsdgeos | so waht do we do with https://code.launchpad.net/~saviq/unity8/clean-build-scripts-coding/+merge/173847 ? | 14:01 |
| tsdgeos | want me to get a VM and see if i can build and run with that code? | 14:01 |
| tsdgeos | or? | 14:01 |
| Saviq | tsdgeos, yeah please | 14:01 |
| Saviq | tsdgeos, but I can tell you that you can | 14:01 |
| Saviq | tsdgeos, indicators are not gonna be there, but they're not really there today either | 14:02 |
| tsdgeos | sure, but we're supposed to review the changes, no? | 14:02 |
| tsdgeos | :D | 14:02 |
| Saviq | tsdgeos, :F | 14:02 |
| tsdgeos | Saviq: we only support saucy, right? | 14:03 |
| Saviq | tsdgeos, +1 | 14:04 |
| Saviq | dednick, hmm I just tried "autopilot run unity8.indicators_client"... I noticed that was wrong, but shouldn't we keep all our AP tests under unity8? | 14:05 |
| Saviq | dednick, they don't necessarily need to run with `autopilot run unity8` | 14:06 |
| Saviq | mzanetti_, thoughts ↑? | 14:06 |
| === pete-woods-lunch is now known as pete-woods | ||
| dednick | Saviq: dont understand. you want them to run with unity8 or not? | 14:09 |
| Saviq | dednick, not necessarily | 14:09 |
| Saviq | dednick, don't know, really | 14:09 |
| Saviq | dednick, it's cool keeping them separate | 14:10 |
| dednick | Saviq: the ap tests are in the unity8-autpilot package, but they are separated into a different suite | 14:10 |
| Saviq | dednick, but referenced by "unity8.indicators_client" nevertheless | 14:10 |
| Saviq | dednick, it's probably just a case of moving them under the unity8 dir | 14:10 |
| Saviq | dednick, I think we shouldn't install them directly on PYTHONPATH | 14:10 |
| dednick | Saviq: i c. | 14:10 |
| dednick | Saviq: although in that case they would run with unity8 i'm guessing | 14:11 |
| Saviq | dednick, not if they're not in unity8.tests | 14:11 |
| dednick | Saviq: ah. right | 14:11 |
| Saviq | dednick, you'd have to import them in there, I think | 14:11 |
| * Saviq checks | 14:11 | |
| Saviq | dednick, yeah, imports need unity8. but that's about it | 14:12 |
| Saviq | dednick, nope, it recursively looks for tests | 14:13 |
| Saviq | dednick, I'm still better with that than polluting PYTHONPATH, to be honest | 14:13 |
| Saviq | mzanetti_, ↑ need your opinion here | 14:14 |
| === zoktar_ is now known as zoktar | ||
| === Cimi_ is now known as Cimi | ||
| pete-woods | is there a process for getting your API HTML docs onto developer.ubuntu.com? | 14:21 |
| seb128 | dpm, ^ seems like a question for you ;-) | 14:24 |
| seb128 | pete-woods, dpm can probably help there | 14:24 |
| pete-woods | seb128: thanks :) | 14:25 |
| seb128 | yw | 14:25 |
| === popey_ is now known as popey | ||
| dpm | pete-woods, essentially get all the HTML docs in a -doc package and pinging me or mhall119 to get them imported into d.u.c | 14:27 |
| pete-woods | dpm: I already have them in a -doc package - it's libusermetrics-doc | 14:27 |
| dpm | pete-woods, for the nitty gritty details there's http://bazaar.launchpad.net/~api-doc-publisher-hackers/api-doc-publisher/trunk/view/head:/README#L37 | 14:27 |
| dpm | pete-woods, ah, cool. What kind of API docs are they? I.e. which generator do you use? qdoc? doxygen? And for what programming language is the API for? | 14:28 |
| pete-woods | dpm: it's doxygen, the language is Qt/C++, there's also a pure C API | 14:29 |
| dpm | pete-woods, let me see if I can get the docs published now. I assume they are for saucy only? | 14:30 |
| pete-woods | dpm: yeah, the package is saucy only atm | 14:30 |
| pete-woods | dpm: and thanks! | 14:32 |
| mhall119 | pete-woods: why doxygen and not qdoc? | 14:32 |
| dpm | pete-woods, there you go: http://developer.ubuntu.com/api/devel/ubuntu-13.10/cplusplus/usermetrics/ | 14:33 |
| pete-woods | mhall119: er, no good reason? | 14:33 |
| pete-woods | dpm: awesome! | 14:34 |
| === mzanetti_ is now known as mzanetti_sa | ||
| === mzanetti_sa is now known as mzanetti | ||
| mzanetti | Saviq: dednick_: yeah... make them separate please | 14:39 |
| * mzanetti was cooking | 14:39 | |
| mzanetti | we can then add multiple test suites to jenkins | 14:39 |
| dednick_ | Saviq: unity8/shell & unity8/indicators_client ? | 14:47 |
| mzanetti | dednick_: tests/autopilot/unit8 (the existing ones) and tests/autopilot/indicators_client (the new ones) | 14:49 |
| mzanetti | dednick_: can be both in the same autopilot package tho | 14:49 |
| dednick_ | mzanetti: yeah, that's what it is at the moment, but Saviq's issue is the pollution of the PYTHON_PATH folder. | 14:50 |
| mzanetti | ah | 14:50 |
| dednick_ | although it's a sep app, so i dont see an issue really. | 14:50 |
| mzanetti | me neither... also, I believe (not entirely sure) that it neds to be top level for autopilot being able to process it | 14:50 |
| mzanetti | but that would need to be proved... as I said. not sure | 14:51 |
| dednick_ | mzanetti: apparently not | 14:51 |
| dednick_ | just need to import the folder in __init__ | 14:51 |
| dednick_ | i think Saviq may be cooking as well. | 14:53 |
| nic-doffay | mzanetti, need your thoughts on something super quick. | 14:53 |
| mzanetti | nic-doffay: I'll b back in 20 mins | 14:54 |
| nic-doffay | mzanetti, cool no rush. | 14:54 |
| nic-doffay | mzanetti, it's basically "how much of a hack do you think think this is" | 14:54 |
| dednick_ | if you have to ask that... | 14:56 |
| nic-doffay | dednick_, good point. | 14:59 |
| nic-doffay | dednick_, I guess I wanted to check how guilty I am. :P | 14:59 |
| === charles_ is now known as Guest68669 | ||
| === kgunn is now known as Guest40271 | ||
| === me4oslav_ is now known as me4oslav | ||
| nic-doffay | Saviq, my expandable component currently inherits from ListItems.Base. I'm trying to wrap it around the ListView in my OptionSelector so that the expandable can handle the expansion. | 15:14 |
| nic-doffay | It's obscuring all the items in the list view though :P | 15:14 |
| nic-doffay | Any idea why? Even when I set it to opacity: 0 it doesn't make a difference. | 15:15 |
| === dandrader is now known as dandrader|afk | ||
| Saviq | dednick_, mzanetti, well indicators_client is not really an app, is it... | 15:17 |
| Saviq | dednick_, should we just go unity8.shell and unity8.indicators_client, then? | 15:17 |
| Saviq | nic-doffay, I'd say Expandable should inherit from Empty at most | 15:18 |
| Saviq | nic-doffay, or is Base even higher up the stack? | 15:18 |
| * Saviq doesn't remember | 15:19 | |
| nic-doffay | Saviq, empty is. | 15:19 |
| nic-doffay | Saviq, done. However I still need to disable the MouseArea support for the expandable component because it's swallowing the clicks that should be sent to the ListView delegates. | 15:19 |
| nic-doffay | Any ideas on how to achieve this? | 15:19 |
| mzanetti | Saviq: dednick_: if autopilot can handle it, its fine with me | 15:20 |
| mzanetti | nic-doffay: would be here now if you still need me | 15:20 |
| Saviq | nic-doffay, Empty doesn't have a MouseArea | 15:21 |
| Saviq | dednick_, ah, see! you read my mind :) | 15:22 |
| dednick_ | Saviq: :) | 15:22 |
| Saviq | dednick_, I'd go for "autopilot-shell" for the target name, though >< | 15:23 |
| dednick_ | Saviq: yup. when i make the changes to the folder structure i'll add a name to the add_autopilot_test macro | 15:24 |
| Saviq | dednick_, ah now I get what you said... was thinking you already moved the folders around | 15:27 |
| Saviq | dednick_, you probably need an empty __init__.py there | 15:28 |
| dednick_ | Saviq: yeah. just with the imports in it i think | 15:28 |
| nic-doffay | Saviq, it does have an onClicked | 15:32 |
| Saviq | nic-doffay, Empty doesn't? | 15:32 |
| dednick_ | Saviq: ok, done. Jenkins should be able to test it without manual trigger now there is only 1 suite. | 15:33 |
| === dandrader|afk is now known as dandrader | ||
| Saviq | dednick_, indeed, we'll modify it later to have separate unity8.shell unity8.indicators_client maybe | 15:33 |
| Saviq | not sure that's desirable, though | 15:33 |
| Saviq | but doesn't really matter much | 15:33 |
| nic-doffay | Saviq, I'm pretty sure it does :P | 15:34 |
| nic-doffay | It doesn't compain with my onClicked logic. | 15:34 |
| nic-doffay | *complain | 15:35 |
| Saviq | nic-doffay, ah! it's an AbstractButton of course | 15:35 |
| nic-doffay | Saviq, which brings me to the swallowing touches question. | 15:35 |
| Saviq | nic-doffay, but then it shouldn't matter, anything you add _into_ it will be on top, why should it swallow the touches? | 15:35 |
| nic-doffay | Saviq, at the moment the Expandable is a parent of the ListView. | 15:35 |
| Saviq | nic-doffay, sure, Empty > Expandable { ListView } | 15:36 |
| Saviq | nic-doffay, so ListView is on top | 15:36 |
| nic-doffay | Saviq, yep. | 15:36 |
| Saviq | nic-doffay, so it takes over all mouse events | 15:36 |
| nic-doffay | Which I thought would be cool. | 15:36 |
| Saviq | nic-doffay, it is | 15:36 |
| nic-doffay | But it doesn't. | 15:36 |
| nic-doffay | Saviq, there might be some other issue though. | 15:37 |
| nic-doffay | ^ most likely the case. | 15:37 |
| Saviq | nic-doffay, http://pastebin.ubuntu.com/5861973/ works | 15:39 |
| Saviq | nic-doffay, so I see no reason why yours wouldn't | 15:39 |
| tsdgeos | Saviq: should i top approve https://code.launchpad.net/~saviq/unity8/clean-build-scripts-coding/+merge/173847 or want kgunn to have a look? | 15:41 |
| mzanetti | Saviq: found the issue... pushed the fix | 15:44 |
| kgunn | MacSlow: hey, so strehl and i were just chatting | 15:45 |
| MacSlow | kgunn, about the autopilot-issue? | 15:46 |
| Saviq | mzanetti, cheers | 15:46 |
| Saviq | tsdgeos, I assigned kgunn for it really to push it to the website | 15:46 |
| kgunn | MacSlow: sorry about notifications in general | 15:46 |
| kgunn | MacSlow: questoin is for things that want to produce sound or haptics | 15:46 |
| Saviq | tsdgeos, so you can approve | 15:46 |
| tvoss_ | sil2100, could you please try including QGLShaderProgram and getting rid of the manual gl version detection there? | 15:47 |
| MacSlow | kgunn, there's an optional "sound-file" hint... http://www.galago-project.org/specs/notification/0.9/x344.html | 15:48 |
| kgunn | MacSlow: so do we have that ? (sorry i'm on a nexus7 ...w/o modem) | 15:49 |
| MacSlow | kgunn, regarding "haptics" ... since we introduced our own hints... there's nothing stopping us for doing that for haptic-feedback on notifications should we chose to do so. | 15:49 |
| tvoss_ | sil2100, you can just check for the default gles version macro then to define the texture target | 15:49 |
| MacSlow | kgunn, not atm... Design never pushed for this. | 15:49 |
| kgunn | MacSlow: but aren't notif's used for phone calls/sms's etc.... | 15:49 |
| kgunn | vibra would be needed | 15:50 |
| MacSlow | kgunn, but at first glance it doesn't seem to be too much trouble to implement it... if there's a solid spec for it :) | 15:50 |
| kgunn | MacSlow: ok...sounds like we should add it as a work item in the bp | 15:50 |
| MacSlow | kgunn, since sound and haptic/vibration are not visual it could all just happen on the backend without the frontend/renderer needing to know?! | 15:51 |
| * tedg always wanted his phone to vibrate the calling number is Morse code | 15:52 | |
| tedg | in | 15:52 |
| kgunn | MacSlow: this is why i was asking...who would really own | 15:52 |
| kgunn | ? | 15:52 |
| kgunn | MacSlow: strehl was thinking we should own | 15:52 |
| kgunn | MacSlow: is notif spec silent on this topic ? | 15:53 |
| MacSlow | kgunn, that question was never really answered... nobody really owns it... I tried what I can to cover for this in the mean-time... but having a team really own the spec (functional-design, tech-spec) would be great | 15:54 |
| nic-doffay | Saviq, see this is giving me the issues: https://pastebin.canonical.com/94121/ | 15:54 |
| nic-doffay | The Expandable which is an Empty still swallows the clicks and I'm unable to see any of the ListView delegates. | 15:55 |
| MacSlow | kgunn, what spec are you refering to? ours or the freedesktop/galago one? | 15:55 |
| sil2100 | tvoss_: k, thanks! | 15:55 |
| tvoss_ | sil2100, yup :) | 15:55 |
| tvoss_ | sil2100, let me know if that works | 15:55 |
| kgunn | MacSlow: galago one | 15:55 |
| MacSlow | kgunn, there's just this optional "sound-file" hint in http://www.galago-project.org/specs/notification/0.9/x344.html | 15:56 |
| Saviq | nic-doffay, what's your ListView's height, then? | 15:56 |
| nic-doffay | Saviq, 0 for some reason. | 15:56 |
| nic-doffay | But the delegates all print out their correct heights. | 15:56 |
| Saviq | nic-doffay, ListViews don't grow to their delegates by default | 15:57 |
| MacSlow | but speaking of the "incoming phone-call" use-case... wouldn't the phone-app ring/vibrate anyway? | 15:57 |
| Saviq | nic-doffay, height: contentHeight if you want it | 15:57 |
| nic-doffay | Saviq, kk thanks. | 15:57 |
| === boiko_ is now known as boiko | ||
| Saviq | dednick_, you didn't add the __init__.py in tests/autopilot/unity8 | 16:10 |
| Saviq | dednick_, it can actually be empty | 16:10 |
| Saviq | dednick_, autopilot goes recursively through them it seems | 16:11 |
| Cimi | seb128, who do I need to ask if xorg crashes? | 16:15 |
| seb128 | Cimi, #ubuntu-x you can try ping mlankhorst or tjaalton directly but they might be eod already | 16:16 |
| Cimi | seb128, like segfaulting :( | 16:16 |
| Cimi | on vmware | 16:16 |
| Cimi | after last update | 16:16 |
| Cimi | pixman | 16:16 |
| Cimi | or I try mir | 16:17 |
| mlankhorst | vmware doesn't have support for mir | 16:17 |
| Saviq | mzanetti, hmm, I'm not sure I'm liking what you did in CMake | 16:18 |
| mzanetti | Saviq: I feared that | 16:18 |
| Saviq | mzanetti, ;) | 16:18 |
| mzanetti | Saviq: whats your reason? | 16:18 |
| Saviq | mzanetti, my hope is that the plugins are as self-contained as possible (and I know they're not, yet) | 16:19 |
| mzanetti | Saviq: so... the thing is, it I do the check_pkg thingie twice, the second (in mocks) will override the first | 16:19 |
| mzanetti | Saviq: that was the reason why the check didn't work | 16:19 |
| Saviq | mzanetti, yeah, I get it, but if both have =2, it'll work, right? | 16:20 |
| mzanetti | Saviq: I could copy/paste the same into both | 16:20 |
| mzanetti | so the first one in plugins would be just "documentation" but basically useless | 16:20 |
| Saviq | mzanetti, the mock one will remain with us while Unity.Launcher can get moved to a separate project | 16:20 |
| mzanetti | Saviq: thats true, yes | 16:20 |
| mzanetti | ok... I can move it back into there | 16:21 |
| Cimi | mlankhorst, if pitman segfaults? | 16:21 |
| Cimi | *pixman | 16:21 |
| Saviq | mzanetti, dednick_ it's a race, btw, your branches conflict in debian/control. so whoever gets in first gets to not fix the conflict ;D | 16:22 |
| Saviq | unless one of you merges the other one's branch in and prerequisites it, of course ;D | 16:22 |
| tvoss_ | mzanetti, ping | 16:23 |
| * mzanetti runs | 16:23 | |
| mzanetti | tvoss_: pong | 16:23 |
| mzanetti | tvoss_: the run was not meant for you :D | 16:23 |
| Saviq | rotfl | 16:23 |
| Saviq | greyback_, can you invite me to your chat with jamesh please? | 16:25 |
| mlankhorst | Cimi: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-vmware/+bug/1199403 ? | 16:25 |
| ubot5 | Ubuntu bug 1199403 in xserver-xorg-video-vmware (Ubuntu) "Xorg crash" [Critical,Triaged] | 16:25 |
| Cimi | mlankhorst, might be | 16:26 |
| Cimi | mlankhorst, I have the same issue | 16:26 |
| Cimi | so I guess it's the driver | 16:26 |
| greyback_ | Saviq: we had it at 11am UTC today | 16:26 |
| Saviq | greyback_, lol | 16:27 |
| john-mcaleely | mfisch, ping... | 16:27 |
| mfisch | hey mhall119, john-mcaleely spotted a typo in the scopes tutorial, can you fix it? | 16:27 |
| Saviq | greyback_, thought jamesh was EOD by that time | 16:27 |
| mfisch | mhall119: https://bugs.launchpad.net/ubuntudeveloperportal/+bug/1199839 | 16:27 |
| ubot5 | Ubuntu bug 1199839 in Ubuntu App Developer site "typo in writing-a-unity-scope" [Undecided,New] | 16:27 |
| mfisch | its just a misspelling of "apt" | 16:27 |
| sil2100 | tvoss_: strangely, now it doesn't find glUniformMatrix4fv when trying to build with GLES2 | 16:37 |
| sil2100 | tvoss_: even though it seems to include the right headers | 16:37 |
| sil2100 | tvoss_: no, wait | 16:38 |
| john-mcaleely | mhall119, while I'm at it (spotting bugs), I'll have a look at adding some packaging to that example. | 16:39 |
| Saviq | didrocks, indicators-client is going away, like, tomorrow | 17:02 |
| Saviq | didrocks, i.e. it's getting merged into lp:unity8 | 17:02 |
| didrocks | Saviq: I'm sure you will be happy the autopilot tests will run with nothing :p | 17:02 |
| didrocks | Saviq: thanks for the head's up! | 17:02 |
| didrocks | fginther: I see a lot of B09googletests hooks used in jenkins | 17:10 |
| didrocks | fginther: but most of the time, we already run the tests as part of the package | 17:10 |
| didrocks | so they are run twice | 17:10 |
| didrocks | fginther: we should maybe look at remove those which makes sense (and if we see that one isn't running tests while building the package, activating that?) | 17:11 |
| fginther | didrocks,yes. It's really a project by project check | 17:11 |
| didrocks | fginther: ok, I'll have a look tomorrow at that | 17:12 |
| didrocks | fginther: if nothing in the stack runs the tests, we don't need to have the googletests_cmake and _install? | 17:12 |
| fginther | didrocks, I'm not following. | 17:13 |
| fginther | do you mean project? | 17:13 |
| didrocks | fginther: I mean, if we don't nede the B09googletests hook in any project of the stack | 17:13 |
| didrocks | fginther: we should remove the other *googletests* hooks from the stacks? | 17:13 |
| fginther | didrocks, ah, right | 17:13 |
| didrocks | ok, will work on that tomorrow :) | 17:14 |
| fginther | didrocks, cool! | 17:14 |
| mzanetti | Saviq: ok. moved it back. | 17:14 |
| Saviq | mzanetti, thanks | 17:15 |
| mzanetti | Saviq: I left the add_subdirectory where it is now... next to the other one and the comment | 17:15 |
| mzanetti | fits better there | 17:15 |
| Saviq | mzanetti, k | 17:15 |
| === tvoss_ is now known as tvoss|dinner | ||
| === thomi_ is now known as thomi | ||
| mzanetti | Saviq: still around? | 19:45 |
| Saviq | mzanetti, knida | 19:48 |
| Saviq | kinda, even | 19:48 |
| mzanetti | Saviq: http://paste.ubuntu.com/5862687/ | 19:48 |
| mzanetti | Saviq: run this and watch your CPU | 19:48 |
| mzanetti | noticed unity8 going crazy when I left it with a popover opened | 19:49 |
| Saviq | mzanetti, ah, same issue I believe | 19:49 |
| mzanetti | ah, the animation? | 19:50 |
| Saviq | mzanetti, http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/trunk/revision/584 | 19:50 |
| Saviq | mzanetti, https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1197801 | 19:52 |
| ubot5 | Ubuntu bug 1197801 in Ubuntu UI Toolkit "Unity8 is hogging the CPU when idle" [Critical,Fix committed] | 19:52 |
| mzanetti | Saviq: but I only see this when I open a popover, I believe | 19:54 |
| Saviq | mzanetti, yeah, that's because the popover uses the shape (and is only created when you open it) | 19:54 |
| Saviq | mzanetti, and destroyed when you close it | 19:54 |
| mzanetti | Saviq: oh... is it the UbuntuShape that can clip everything, not just images | 19:54 |
| mzanetti | I see | 19:54 |
| Saviq | mzanetti, well, Shape itself can, there's an issue with updating the textures | 19:55 |
| Saviq | mzanetti, I commented on the bug and pinged Kaleo | 19:56 |
| Saviq | pung? | 19:56 |
| mzanetti | lol | 19:56 |
| === kgunn is now known as Guest55003 | ||
| === hyperair is now known as Guest96566 | ||
| === ashams is now known as Guest64962 | ||
| mterry | tvoss|dinner, poke about meeting | 20:04 |
| Saviq | dednick_, you still around? | 20:08 |
| Saviq | dednick_, there's no need for the imports in unity8/__init__.py IIUC? | 20:09 |
| === jhodapp is now known as jhodapp|afk | ||
| === AlanChicken is now known as AlanBell | ||
| === shiznix_ is now known as Guest83929 | ||
| === rsalveti_ is now known as rsalveti | ||
| === Guest55003 is now known as kgunn | ||
| === kgunn is now known as Guest77008 | ||
| Saviq | veebers, hey, so I didn't get to getting you the data you requested, but the VM is up and I'm leaving it that way | 22:02 |
| veebers | excellent, thanks Saviq | 22:02 |
| veebers | Saviq: if you're still around, I get a connection time out w/ that machine | 22:06 |
| kgunn_ | Saviq: did you want me to try ...https://code.launchpad.net/~saviq/unity8/clean-build-scripts-coding/+merge/173847 | 22:21 |
| kgunn_ | sorry just catching up on the todo's | 22:22 |
| === AlanChicken is now known as AlanBell | ||
| === _salem` is now known as _salem | ||
| dednick_ | Saviq: imports removed | 22:43 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!