/srv/irclogs.ubuntu.com/2013/11/06/#ubuntu-unity.txt

veebersSaviq: Is the script runtests.sh in unity8 expected to work at this moment? Perhaps I'm not doing something right (and thus can update any docs)02:02
Saviqveebers, http://s-jenkins:8080/job/unity-phablet-qmluitests-trusty/ says it does...02:02
veebersok thanks, I'll check that out02:04
Saviqveebers, just saw that mediumtests disable the edges demo - we should probably add that to the unity8 suite for test scripts to use02:06
veebersSaviq: ack, in the helpers section yeah?02:07
Saviqveebers, yeah02:07
veeberscool, can look into that02:07
=== duflu_ is now known as duflu
veebersSaviq: are you still around?03:08
sam113101HELLO03:12
sam113101I'm having trubble with unity03:12
Mirvwe'd now want ap1.4/xpathselect1.4 support back in, can someone check https://code.launchpad.net/~timo-jyrinki/unity/get_xpathselect_1.4_back/+merge/194069 ?06:27
tsdgeosoh lol08:05
tsdgeosthe unity7 "panel" is crashing like crazy08:05
tsdgeoslet me dist-upgrade again to see if it fixes itself08:05
tsdgeoslooks like not08:08
seb128tsdgeos, can you get a bt?08:11
tsdgeosi should08:12
tsdgeosneed to find out to which process to attach08:12
tsdgeossomehow it seems i disabled appport and don't know how to bring it back :D08:12
seb128tsdgeos, gdb -p $(pidof unity-panel-service)?08:12
tsdgeossure, that should work :D08:12
tsdgeosseb128: http://paste.ubuntu.com/6369213/08:13
tsdgeoslol08:14
tsdgeosi have two network indicators too08:14
seb128tsdgeos, hum, ido, fun ... "dpkg -l - grep libido"?08:14
tsdgeoswe're adopting the kde ideas? :D (KDE had the same problem with duplicated network stuff lately)08:14
seb128tsdgeos, that's normal if you installed e.g ubuntu-system-settings (that pulls in the new indicator, we still use nm-applet on the desktop)08:14
tsdgeosii  libido3-0.1-0:amd64                                   13.10.0+14.04.20131105.1-0ubuntu1             amd64        Shared library providing extra gtk menu items for display in08:15
seb128tsdgeos, can you install libido3-0.1-0-dbgsym and get a new bt?08:15
tsdgeossure08:15
seb128larsu, it seems like you have an invalid unref in your recent ido changes08:16
=== iahmad_ is now known as iahmad
=== iahmad is now known as iahmad|afk
tsdgeosseb128: larsu: http://paste.ubuntu.com/6369241/08:24
tsdgeosi can valgrindize it if you guys want too08:24
seb128tsdgeos, I guess that would be useful, though it's likely due to http://bazaar.launchpad.net/~indicator-applet-developers/ido/trunk.14.04/revision/15508:29
tsdgeosseb128: do you remember the magic stuff i have to pass to valgrind when debugging glib programs so its useful?08:32
seb128tsdgeos, export G_SLICE=always-malloc G_DEBUG=gc-friendly08:33
seb128tsdgeos, https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1248446 btw for tracking it08:35
ubot5Ubuntu bug 1248446 in ido (Ubuntu) "unity-panel-service crashed with SIGSEGV in g_object_unref()" [High,Confirmed]08:35
tsdgeosseb128: hmmm, i'm having problems starting the panel-service under valgrind since killing it restarts it automatically and then the valgrind one says the bus is already taken, any clue how to make it not restart or to make the new one take over?08:38
seb128tsdgeos, stop unity-panel-service08:39
seb128"stop" as the upstart command08:39
seb128e.g type that08:39
tsdgeosthat makes sense08:39
tsdgeosi'm not much into this upstart mindset :D08:40
seb128upstart job management is handy ;-)08:40
tsdgeosSaviq: look what i found yesterday http://picpaste.com/pics/IMG_00000616-b7AlTUm9.1383728804.jpg09:07
Cimilol09:10
larsuseb128: oops. Already spotted it in that diff (g_file_icon_get_file() is transfer-none)09:20
seb128larsu, e.g g_object_unref (file); is wrong?09:20
larsuyes09:20
larsuI'll fix it right away09:20
=== iahmad|afk is now known as iahmad
Saviqtsdgeos, aaawww thank you :D09:26
tsdgeoswonder what was the guys idea to name it like that09:26
tsdgeosthat or you have a second business here in barcelona and you never told us09:26
nic-doffaySaviq, I think this is ready when you have a moment: https://code.launchpad.net/~nicolas-doffay/unity8/search-history-persist09:38
Saviqnic-doffay, k09:38
nic-doffaySaviq, nm just saw your top comment.09:43
nic-doffaySaviq, where should I add this PageHeader test?09:54
Saviqnic-doffay, to the PageHeader tests09:54
nic-doffaySaviq, I meant searchHistory sorry.09:54
Saviqnic-doffay, ah no, DashContent or somewheer09:54
Saviqnic-doffay, you need to test that it's common between scopes09:54
Saviqnic-doffay, so you need one of the suites where there's multiple scopes09:54
nic-doffaySaviq, yeah.09:54
nic-doffaySaviq, taking a look at DashContent test now.09:55
Cimican anyone tell me why system-settings-wizard does run but shows no window ? lp:~unity-team/ubuntu-system-settings/wizard-cmake09:56
tsdgeosSaviq: it's ok i keep triggering manual autolandings, right?10:06
Saviqtsdgeos, not now10:06
tsdgeosSaviq: wops10:06
* tsdgeos hides10:06
Cimihah10:06
Saviqtsdgeos, we're waiting for the autopilot 1.4 transition to settle10:07
tsdgeosok10:07
Saviqtsdgeos, you're fine, just don't merge more10:07
tsdgeossure10:08
Cimiif you are bored, I am stuccoed with lp:~unity-team/ubuntu-system-settings/wizard-cmake10:11
* Cimi just saying10:11
MirvSaviq: I was just about to say but I see you already spanked tsdgeos :)10:18
MirvSaviq: I see three failing tests in unity8 with AP1.410:18
SaviqMirv, on mako?10:18
tsdgeosCimi: what's the problem?10:18
MirvSaviq: on mako, yes, http://pastebin.ubuntu.com/6369664/10:18
SaviqMirv, I wasn't really paying attention, thought the ap guys will make sure they pass...10:18
MirvSaviq: so did I. let me grab a log.10:19
Cimitsdgeos, I run system-settings-wizard10:19
SaviqMirv, oh well, they did pass http://10.97.0.26:8080/job/generic-mediumtests-trusty-touch/480/10:19
Cimitsdgeos, the window does not appear10:19
SaviqCimi, works fine here10:21
CimiSaviq, wizard?10:21
Saviqah wait, wizard10:21
tsdgeosMirv: Saviq: then how did my thing get merged?10:21
SaviqCimi, qrc:/qml/Pages/WelcomePage.qml:20 module "Ubuntu.SystemSettings.LanguagePlugin" is not installed ?10:21
Saviqtsdgeos, yeah, that as well10:21
CimiSaviq, I did make install10:21
CimiSaviq, then it finds the plugin10:21
CimiSaviq, but still doesn't load the app10:22
SaviqCimi, well, I won't...10:22
SaviqCimi, view.show()10:22
SaviqCimi, you never *show* the window, so it doesn't show10:22
CimiI'm very embarassed10:23
CimiI commented our viewFullscreen because doesn't work here10:23
MirvSaviq: the attachment in bug #124847710:23
ubot5bug 1248477 in Unity 8 "Three failing tests with unity8 + AP1.4" [Critical,New] https://launchpad.net/bugs/124847710:23
Cimibut I did not replace it with show10:23
CimiSaviq, sorry...10:24
Cimispent so much time on the cmake without realising the issue was in the code :)10:24
SaviqMirv, you got a .crash file?10:26
* Saviq adds daily-build10:27
MirvSaviq: no, the only crash I have is upstart-app-lauch10:29
SaviqMirv, interesting...10:29
SaviqMirv, how about ~/.cache/upstart/unity8.log ?10:29
SaviqMirv, I'm running them locally now from daily-build, let's see what I get10:31
MirvSaviq: http://people.canonical.com/~tjyrinki/u8/10:32
SaviqMirv, to reduce tsdgeos's offence, he's merged a fix of one of the failing tests of unity8 on 5.2 ;)10:33
SaviqMirv, ok, nothing obvious, testing locally - can you reliably fail those tests?10:34
MirvSaviq: sure, that's great :) and yes those tests fail every time for me.10:36
SaviqMirv, interesting10:36
SaviqMirv, 22 tests OK here10:41
SaviqMirv, I just installed unity8-autopilot from daily-build (and its dependencies)10:41
Saviqso it pulled in unity8 and friends, and libautopilot-qt 1.410:42
MirvSaviq: ok, weird. can you dpkg -l | grep autopilot to check also autopilot-touch is uptodate?10:42
SaviqMirv, nope, it's not10:52
* Saviq installs10:53
=== _salem is now known as salem_
MirvSaviq: ok I think it was again this /home/phablet/autopilot, or seems probable at least10:58
SaviqMirv, right...10:59
SaviqMirv, 22 OK with autopilot-touch, too11:04
Saviqfalse alarm, it seems11:05
MirvSaviq: yep, mark the bug as invalid11:05
Mirvor I can11:05
* Mirv did11:05
=== alan_g is now known as alan_g|afk
dandradermy "date & time" indicator disappeared (talking about ubuntu desktop trusty here). Has it happened for any of you guys? If so, is there an easy way to get it back?11:25
seb128dandrader, is indicator-datetime-service running?11:26
tsdgeosdandrader: run /usr/lib/x86_64-linux-gnu/indicator-datetime-service11:26
tsdgeosdandrader: happens randomly here11:26
pabloff9I'm also having this11:26
seb128tsdgeos, dandrader: does it happen at login? (that bug is supposed to have been fixed recently)11:27
pabloff9not on my PC now to test it, though11:27
seb128pabloff9, ^11:27
tsdgeosseb128: it happens randomly at login yes11:27
tsdgeosit may have been fixe11:27
dandraderseb128, no11:27
seb128tsdgeos, that was bug #123971011:27
ubot5bug 1239710 in indicator-session (Ubuntu) "indicator-datetime or -session missing ~10% of the time" [High,Fix committed] https://launchpad.net/bugs/123971011:27
dandraderseb128, yes (for the second question)11:27
seb128tsdgeos, I didn't see the issue since I've the updates11:27
tsdgeosseb128: maybe, don't remember the last time it happened11:28
dandraderseb128, I just turned on my computer today and I noticed it was missing the date-time indicator11:28
seb128dandrader, dpkg -l | grep indicator-datetime11:28
dandraderseb128,  13.10.0+13.10.20131023.2-0ubuntu111:29
seb128dandrader, did you just install upgrades?11:29
dandraderseb128, yes11:30
dandraderseb128, so rebooting will solve it then?11:30
seb128dandrader, ok, so maybe you just got that update11:30
=== alan_g|afk is now known as alan_g
seb128dandrader, well, restarting unity-panel-service should fix it11:30
seb128dandrader, e.g kill it or use "restart"11:30
tsdgeosSaviq: i honestly don't think i can fix the singleton issue myself in 5.211:30
dandraderseb128, thanks. it did fix it! :D11:31
Saviqtsdgeos, meaning "our side"?11:31
tsdgeosunless i put hundred hours to understand the V4 code which I doubt we want to do11:31
dandraderseb128,  "restart unity-panel-service", I mean11:31
tsdgeosSaviq: well, it just broke, i can remove the test to fix it :D11:31
seb128yeah11:31
Saviqtsdgeos, we might need that temporarily11:31
Saviqtsdgeos, managed to talk to Qt guys about it?11:32
tsdgeosSaviq: i'll repport a new bug with regression in the topic11:32
tsdgeosnope11:32
tsdgeosthey're all awol11:32
Saviqtsdgeos, slackers11:32
* tsdgeos checks when DevDays SF happens11:32
tsdgeostoday :D11:32
Saviqright, makes sense ;)11:32
tsdgeosso yeah11:32
tsdgeosi think they have a good excuse11:33
Saviqtsdgeos, so yeah, skip the failing test with a TODO and a QTBUG link11:33
Saviqtsdgeos, so as not to block the switch to 5.2 (assuming we'll actually work!?)11:33
tsdgeosSaviq: he he, well once it builds on the ppa we can test it easierly11:33
Saviqtsdgeos, indeed11:33
dandraderSaviq, https://code.launchpad.net/~dandrader/unity8/drag-crash-1228336/+merge/192183 is all green and happy. Can we have it in?11:34
Saviqdandrader, we're waiting for the autopilot 1.4 transition to conclude11:34
dandraderah, ok11:35
Saviqdandrader, then we can merge again11:35
Saviqtsdgeos, do you know if QtC 3.0.0-beta is supposed to be buildable with 5.0?11:42
Saviqtsdgeos, README mentions that it requires 5.2 or 4.8 (with restrictions)... not sure it means that 5.0/5.1 should work with restrictions or not...11:42
tsdgeosnot really uptodate with QtC but i remember reading something along the lines that would say "no"11:42
Saviqtsdgeos, 's what I thought11:43
SaviqMirv, I've a branch that (almost - docs seem to be broken) builds QtCreator 3.0, want a look?11:45
SaviqMirv, on that note, how do you do "syncing with upstream"? I usually just move the debian and .bzr folders over and commit the changes...11:45
Saviqbut I'm sure there's a better, more established way...11:46
MirvSaviq: ping bzoltan about that, please11:49
MirvSaviq: I import the new upstream tarball with bzr import-upstream (I think, or was it merge-upstream..)11:50
MirvSaviq: then commit that and make packaging changes as needed11:50
nic-doffaySaviq, unsure why this is failing now, I can't see anything in the logs. https://code.launchpad.net/~nicolas-doffay/unity8/filter-selector/+merge/19114511:50
SaviqMirv, right, didn't know that :)11:50
MirvSaviq: the Ubuntu plugin needs to be ported over to 3.0 before the QtC 3.0 itself can be considered. but push your branch somewhere anyhow.11:50
SaviqMirv, yeah11:51
Saviqnic-doffay, that's temporary11:51
Saviqnic-doffay, autopilot 1.4 isn't yet published11:52
Saviqnic-doffay, restarted the job so that it picks up the daily-build ppa to get 1.411:53
nic-doffaySaviq, cool.11:53
tsdgeosSaviq: https://code.launchpad.net/~aacid/unity8/singleton_52/+merge/19412312:05
=== MacSlow is now known as MacSlow|lunch
Saviqtsdgeos, thanks12:10
dandraderSaviq, do we still support SurfaceFlinger. It so, until when?12:11
dandraders/It/If12:11
Saviqdandrader, "support" is an overstatement... but yeah, it's still there, hopefully not for long12:12
Saviqdandrader, i.e. if we regress on sf that's ~fine12:12
dandraderok12:13
Saviqdandrader, once we get Mir on the Nexus10, I'll be pushing for dropping sf altogether12:13
Saviqdandrader, tsdgeos, fyi, once we get smoke results from the next image, and they're green for unity8, is when we can start merging again12:16
tsdgeos\o/12:16
Saviqtsdgeos, looking at https://code.launchpad.net/~aacid/unity8/singleton_52/+merge/194123... unity-notifications will fail as well, as they rely on it :/12:16
* Saviq pushes unity-notifications to qt5-beta212:16
SaviqMirv, ok ↑?12:16
tsdgeosSaviq: do they12:17
tsdgeosthat's bad12:17
Saviqtsdgeos, erm wait12:17
Saviqtsdgeos, I meant unity-api12:17
SaviqMirv, ↑12:17
Saviqand the tests in there12:17
Saviqat least12:17
Saviqright, unity-api doesn't run the tests in autopkgtest... so they built just fine12:18
tsdgeosok, i'll try to compile+test  unity-api againsst my 5.212:18
Saviqtsdgeos, yes please, qml tests, that is12:18
tsdgeosSaviq: can you quick review this? i realized i had it in my qtubuntu from the roaming fixes i did on release day but was unpushed https://code.launchpad.net/~aacid/qtubuntu/const_ref/+merge/19412612:23
=== iahmad__ is now known as iahmad
MirvSaviq: ok, or do a recipe?12:51
dandraderouch! I'm unable to build unity8 packages on the device: "c++: internal compiler error: Killed"12:52
tsdgeoslol12:55
tsdgeosdandrader: do you have enough disk space?12:55
dandradertsdgeos, think so. "/dev/mmcblk0p12                15G  3.6G   11G  25% /home"12:58
dandraderstanding for "Filesystem                    Size  Used Avail Use% Mounted on"12:58
=== alan_g is now known as alan_g|lunch
Saviqdandrader, `powerd-cli active` as root13:07
Saviqdandrader, when it gets into low-powered mode, it will interrupt system calls under high load13:07
dandraderSaviq, ah, interesting.13:07
Saviqtsdgeos, "const QRectF & relativeGeometry" please fix spacing?13:07
Saviqdandrader, connecting to the PC (not charger) helps as well13:08
=== MacSlow|lunch is now known as MacSlow
mhr3Saviq, so you were ok with the approach of taking unity and removing all the non scope bits to get the "scope test app"?13:26
mhr3one thing i kinda like about it is that we can keep merging things that get into unity13:27
Saviqmhr3, yeah, if you need to refactor some things to e.g. load Dash.qml directly, that might be a good solution as well13:27
Saviqmhr3, which would keep us even closer13:27
mhr3Saviq, meh, this works, although i do wonder what to do about tests13:28
mhr3i guess we don't really care about the ui tests and would be only interested in unit tests13:28
mhr3as in pure c++ unit tests13:28
Saviqmhr3, yeah, UI tests would happen on our side13:29
mhr3Saviq, but ultimately this setup isn't going to help much when merging back into unity, we'll loose all the history... problem?13:29
Saviqmhr3, or you could just work on lp:unity8 directly...13:30
Saviqmhr3, or... extract the Unity plugin first13:30
Saviqmhr3, and only work on lp:unity8 for the UI bits13:31
Saviqmhr3, and keep the unity plugin separate13:31
mhr3yes, lots of options, i'm not sure what the best13:31
mhr3anyway, pushed what i have to lp:~unity-team/unity8/unity-scopes-only13:32
Saviqmhr3, I'd like us to keep as close as possible, to not fork too much13:33
Saviqmhr3, so maybe indeed separating out the unity plugin would be good13:33
Saviqmhr3, you could fork that for the new scopes api13:33
Saviqmhr3, while we could keep all the UI changes in lp:unity813:33
mhr3yea, it does sound sensible13:33
mhr3the question is whether there really won't be any qml-facing API changes13:34
mhr3if there are, things will get complicated13:34
Saviqmhr3, that's when we'll "fork" inside unity813:36
Saviqmhr3, piece-by-piece13:36
Saviqmhr3, maybe we can even use Qt's versioning mechanism at that point13:36
mhr3qt has versioning? :)13:37
Saviqmhr3, yeah, you can tag members, for example, with revisions13:37
Saviqmhr3, and then, when you import a certain version, that one member will get imported or not13:37
mhr3how do you use it then in qml?13:37
Saviqmhr3, "import blah 1.1"13:37
Saviqmhr3, in qmldir you can specify which qml files correspond to which version of the import13:38
mhr3aaah13:38
Saviqmhr3, so a single plugin under Unity.1 can have different imports for Unity 1.0 and 1.113:38
mhr3interesting13:38
Saviqwould require some refactoring on unity8 side, but that's not necessarily bad anyway ;)13:39
mhr3heh13:39
SaviqMirv, another question - is there a way to get the changed files into debian/changelog automagically?13:39
mhr3Saviq, so for the separated-out unity plugin, should i create a new lp project?13:40
Saviqmhr3, yeah, unity-scopes? ;)13:40
Saviqmhr3, unity-scopes-shell?13:40
Saviqmhr3, unity-qt?13:40
mhr3i like unity-scopes-shell13:41
mhr3since the other part is unity-scopes-api13:41
MirvSaviq: maybe, but I don't know about such. I've used dch for changelog modifying otherwise.13:41
sil2100bregma: ping :)13:41
bregmasil2100, how may I help?13:42
Saviqmhr3, while you're at it, extracting abstract classes into lp:unity-api would be nice :)13:42
=== alan_g|lunch is now known as alan_g
mhr3Saviq, that's the util lib, right?13:42
mhr3i know we're using it, but never actually looked at what's in there :)13:43
sil2100bregma: hello! Any progress in getting the fix for LP: #1247787 released?13:43
ubot5Launchpad bug 1247787 in cmake (Ubuntu) "Unity FTBFS from lp:unity on trusty" [Critical,In progress] https://launchpad.net/bugs/124778713:43
sil2100bregma: I think I saw didrocks mention something about the patch not being rebased on the latest ubuntu source?13:43
Saviqmhr3, not even util, just a place to keep API definitions (abstract base classes)13:43
sil2100didrocks: ^ can you comment?13:43
Saviqmhr3, and corresponding API tests that verify that an API is conformant with the requirement13:44
mhr3Saviq, hm, i guess it's time to look at it :)13:44
Saviqmhr3, it's not big :)13:44
bregmasil2100, the patch is the latest Ubuntu source with a patch cherry-picked from upstream, as the debdiff clearly shows13:44
Saviqmhr3, ah and yeah, there's utils from michi13:44
Saviqmhr3, not sure anything uses those, though13:44
didrockssil2100: what should I comment? the branch is on 2.8.11.2-1ubuntu5 and we have 2.8.12-0ubuntu1 in trusty13:44
mhr3Saviq, i know the scopes-api is13:44
didrocksas well, I saw some inline diff, I would prefer a debdiff with only debian/patches/13:45
sil2100bregma: ^13:45
Saviqmhr3, okay, good, then13:45
bregmadidrocks, the Ubuntu source keeps its patches applied -- not my pref, but I wasn;t going to break what's already there13:45
bregmathe debdiff contains only the patch, the MP diff has the patch applied13:46
didrocksbregma: I saw one file applied (or maybe the diff was contracted)13:46
didrocksbregma: but is it on 2.8.12-0ubuntu1?13:46
didrocksI saw 2.8.11.2-1ubuntu5 in the branch13:46
mhr3Saviq, hm, there isn't really much besides the util classes, but from the dir layout it's very much an older copy of current scopes-api13:47
nic-doffaySaviq, there's one unrelated instability that jenkins has reported with the filters mp13:48
Saviqmhr3, there's more here http://bazaar.launchpad.net/~unity-team/unity-api/trunk/files/head:/include/unity/shell/13:48
nic-doffayhttps://code.launchpad.net/~nicolas-doffay/unity8/filter-selector/+merge/19114513:48
Saviqnic-doffay, yeah, saw that13:48
bregmadidrocks, here's the debdiff attached to the bug: https://launchpadlibrarian.net/155893853/debdiff13:48
Saviqnic-doffay, that's just a flaky test13:48
Saviqmhr3, the shell APIs are not built into a lib even - they're just headers13:48
mhr3oh, didn't check the include dir :/13:49
didrocksbregma: ah nice, sponsoring that one then13:49
Saviqmhr3, the idea is that whoever wants to implement a shell-facing component13:50
mhr3Saviq, not exactly sure that's good... if we want to keep minimal changes, sure, otherwise it will be pita to maintain in there13:50
* bregma thinks packaging branches should not be kept with patches applied in bzr13:50
Saviqmhr3, sure, we should only merge into lp:unity-api once we have a more-or-less stable API13:51
nic-doffaySaviq, I think this needs a rebuild too https://code.launchpad.net/~nicolas-doffay/unity8/search-history-persist/+merge/19393513:51
Saviqmhr3, not saying we should start with that13:51
mhr3Saviq, ok, then i'm fine with it13:51
Saviqmhr3, just extract the api in unity-scopes-shell into headers, and at some point we'll move them over to lp:unity-api13:52
mhr3yep, will do13:52
Saviqnic-doffay, the test checks for entries on the first scope twice, how does that check for persistence across scopes?13:53
Saviqnic-doffay, and it's doing that at too low a level13:53
mhr3but then this doesn't seem like a very good place for the util classes13:53
Saviqmhr3, lp:unity-api? might not be indeed13:53
mhr3seems kinda circular13:54
Saviqnic-doffay, it's just interrogating the same object, which is kind of useless13:54
Saviqnic-doffay, you should type into the search entry, switch scopes, focus the search entry and check that the history is correct13:54
Saviqnic-doffay, what you're testing there instead is whether the SearchHistoryModel works13:54
Saviqnic-doffay, which, granted, it might need (don't remember its coverage)13:55
Saviqnic-doffay, but still not a relevant test for that bug13:55
nic-doffaySaviq, all that would have to be done from the PageHeader.13:56
Saviqnic-doffay, why is that?13:56
Saviqnic-doffay, you just find a PageHeader child in the DashContent13:56
Saviqnic-doffay, based on its objectName13:56
Saviqnic-doffay, then find the text entry and type into it13:57
nic-doffaySaviq, I just presumed that those tests were best done from the PageHeader.13:57
Saviqnic-doffay, you can't do that there13:57
Saviqnic-doffay, as you don't have multiple scopes13:57
Saviqnic-doffay, just one PageHeader13:57
nic-doffaySaviq, right13:57
Saviqnic-doffay, or well, you could just create two PageHeaders13:58
Saviqnic-doffay, but that wouldn't test how it's really used13:58
nic-doffaySaviq, should I get rid of that other test that I wrote prior then?13:58
Saviqnic-doffay, what we need to make sure of there, is that the use case really works13:58
Saviqnic-doffay, the added test in that MP is not in the right place, at least13:58
Saviqnic-doffay, if you want to test that the SearchHistoryModel behaves as defined, that's a tst_SearchHistoryModel (if there isn't one)13:59
Saviqi.e. whether it adds / promotes entries as expected13:59
Saviqnic-doffay, what you need is basically a regression test for the bug14:00
Saviqnic-doffay, and do it on as low a level as possible, yes, but still such that it verifies that the user will see the same behavior14:00
Saviqotherwise you're testing the test setup itself / Qt / QML, which we don't want to do14:01
sam113101Saviq: help14:05
Saviqsam113101, please ask your question first14:06
sam113101Saviq: why do I get this?: http://imgur.com/a/vRRwR14:07
Saviqsam113101, no idea, I'm only working on unity814:09
Saviqsam113101, is there a bug? if not - please file one14:09
sam113101Saviq: unity8 is the one that's not using compiz, right?14:10
Saviqsam113101, yes, the phone/tablet one currently14:10
Saviqsam113101, either way, it looks like a gtk theming issue14:10
Saviqsam113101, please file a bg14:10
Saviqbug14:10
sam113101under what "program", gtk? unity?14:11
Saviqsam113101, I can confirm it happens for me, too, under Radiance14:11
sam113101sweet, I thought it was only me14:12
Saviqsam113101, apport-bug light-themes14:12
Saviqthat will send a launchpad bug to the light-themes project, which is where Radiance comes from14:12
Saviqand yikes my eyes hurt ;) it's so bright14:12
seb128Saviq, sam113101: "ubuntu-themes" is the correct project for that14:12
seb128we merged themes back in there some cycle ago14:12
nic-doffaySaviq, right. I have no idea how to access the PageHeader which lives in item which is the delegate of the loader on dashContentList using findChild.14:13
Saviqseb128, sure, apport-bug light-themes will file it there, no?14:13
seb128Saviq, oh, right, we still build that binary ... yes ;-)14:13
Saviqnic-doffay, just find the delegate based on its index (i.e. its objectName should be "scopeView" + index)14:13
Saviqnic-doffay, then you go var scopeView = findChild("scopeView0") or some such14:14
Saviqnic-doffay, and then scopeView.findChild("pageHeader") or similar14:14
Saviqnic-doffay, there's plenty of examples in the current tests14:14
sam113101https://bugs.launchpad.net/ubuntu/+source/ubuntu-themes/+bug/124855814:24
ubot5Ubuntu bug 1248558 in ubuntu-themes (Ubuntu) "Radiance theme has Ambiance "icons"" [Undecided,New]14:24
sam113101is unity8 only available on tablets and phones right now?14:25
kgunndednick: mzanetti joining ?14:34
Saviqtsdgeos, http://pastebin.ubuntu.com/6370774/ that looks like broken docs in qtcreator14:34
Saviqdoes it not?14:34
tsdgeosSaviq: when does that happen? build stage?14:34
Saviqtsdgeos, yeah14:35
tsdgeosweird14:35
tsdgeosCimi: ↑↑↑↑↑14:35
sam113101so, is unity8 available for desktops?14:36
davmor2sam113101: come back at 14.10 and unity8 should be available across the board14:37
sam113101that's a long time14:38
sam113101so unity will still be a compiz plugin on 14.04?14:38
Daekdroomsam113101, yeah, because Mir deployment was delayed.14:38
sam113101well, at least we'll have mir on 14.04, right?14:39
DaekdroomThat's the plan, afaik.14:39
sam113101I'd like to work on mir and/or unity next, where should I start?14:40
sam113101seriously15:08
tsdgeossam113101: there's a team meeting for unity8 happening, so people is not reading here, wait a 30 mins15:10
tsdgeosgreyback: am i the only one that was kicked out of the hangout?15:11
greybacktsdgeos: yes15:11
fgintherSaviq, ping15:17
Saviqfginther, otp15:18
fgintherack15:18
=== alan_g is now known as alan_g|tea
sil2100Trevinho: hi!15:47
sil2100Trevinho: can we get https://code.launchpad.net/~timo-jyrinki/unity/xpathselect-1.4/+merge/194077 approved and merged?15:47
sil2100Trevinho: is it ok?15:47
sil2100Trevinho: if yes, can you review and approve for merging?15:49
=== alan_g|tea is now known as alan_g
dednickSaviq: when you get a second, can you add the comment "will be used by unity8 and ubuntu-system-settings" to the ubuntu-settings-components landing request? I don't have write access.15:53
=== dandrader is now known as dandrader|lunch
Saviqfginther, back16:00
ChrisTownsendsil2100: I'm reviewing that branch right now.16:02
Saviqsam113101, mir+unity8 is going to be an optional non-default session in 14.0416:03
Saviqsam113101, one that will provide mostly the tablet experience16:03
Saviqsam113101, if you want to help - code is in lp:unity8, here's some guidance on how to build it https://unity.ubuntu.com/getinvolved/development/unity8/16:04
Saviqsam113101, process is as usual - grab a bug you like, try and fix it and submit a merge proposal16:04
sam113101thanks16:05
Saviqdednick, done16:09
dednickSaviq: thanks16:09
* Saviq recovers his throat16:09
dednickSaviq: /var/local/autopilot/setup.log:  unity8-autopilot : Depends: libautopilot-qt (>= 1.4) but 1.3+13.10.20130814-0ubuntu1 is to be installed16:14
dednickSaviq: known?16:14
Saviqdednick, yeah, we're transitioning from 1.3 to 1.416:15
dednickSaviq: ok16:15
Saviqdednick, but it's not in the archive yet16:15
Saviqdednick, if you want you can add a hook to the job to add daily-build PPA with ap 1.416:15
fgintherSaviq, I noticed unity8 CI failures becuase of missing ap 1.416:16
fgintherahh16:16
Saviqfginther, yeah, we don't have daily-build as default, was waiting for it to get into distro16:17
fgintherSaviq, Ok, just checking. I won't touch anything16:17
sil2100ChrisTownsend: thanks!16:19
sam113101kind of weird, lol16:30
sam113101I don't really see myself using that on a desktop computer16:30
sam113101is there any working app?16:31
ChrisTownsendsil2100: I approved Timo's branch, and went ahead and globally approved.  I'm sure the merge will fail since we are still waiting on the cmake change, but I wanted to approve it for historical purposes:)16:32
sil2100ChrisTownsend: so the cmake change is still not there?16:33
* sil2100 hoped it's released already16:33
ChrisTownsendWell, I think it's been sponsored, but I don't know the status of the upload and inclusion into the archive.  It may already be there for all I know.16:33
sil2100ChrisTownsend: thanks for approving anyway!16:33
sil2100Oh, ok, let me check that16:34
nic-doffaySaviq, the scopeView object doesn't contain a PageHeader in the test. It's FakeScopeView.16:34
sil2100ChrisTownsend: I see it's there, in -proposed, so depending if lp:unity is using -proposed or not for CI and merger, we might get it working16:35
ChrisTownsendsil2100: Ok, cool.  And I heard that Nux and Compiz are being removed temporarily from the daily-build PPA, right?16:36
Saviqnic-doffay, right...16:37
Saviqnic-doffay, ok, let's let it be - tsdgeos is doing things to the header anyway16:37
Saviqnic-doffay, we'll try and get the test in there16:37
sil2100ChrisTownsend: yes, we're holding of the transition for now, we need to finish 1.4 AP transition first16:37
ChrisTownsendsil2100: Ok, great!16:38
nic-doffaySaviq, can we land the branch without a test then?16:39
Saviqnic-doffay, yeah, not today though - still waiting for GO after the autopilot 1.4 transition16:39
nic-doffaySaviq, cool.16:39
sam113101"qml phone shell", is it what it's going to be like on the desktop, too?16:39
Cimitsdgeos, weird, I subscribed myself16:41
Saviqsam113101, that's an old name, where did you find it?16:42
sam113101Saviq: that's the title of the app I've just run16:42
ChrisTownsendsil2100: Autolanding failed with the cmake error, so I guess it doesn't pull from -proposed.16:42
Saviqsam113101, right ;)16:42
Saviqsam113101, that can be your first contribution :)16:42
Saviqsam113101, it's the window title in main.cpp16:42
ChrisTownsendsil2100: How wonder how long cmake will be in -proposed...16:44
sam113101https://unity.ubuntu.com/getinvolved/development/unity8/16:44
sam113101what's the name supposed to be?16:44
sil2100ChrisTownsend: tahdah! It's in release already, re-approving the branch ;)16:52
ChrisTownsendsil2100: Alright!16:52
=== dandrader|lunch is now known as dandrader
sam113101seriously though, the thing I'm using right now will be used on the desktop too?17:36
sam113101serious question17:36
cwaynesam113101: eventually.  as i understand it though, it won't look like that18:27
cwayneit will look like unity7, but be the same codebase as unity818:28
=== salem_ is now known as _salem
=== dandrader is now known as dandrader|afk
=== dandrader|afk is now known as dandrader
fgintherSaviq, can you assign https://blueprints.launchpad.net/ubuntu/+spec/core-1311-upstream-merger-20 to ~canonical-ci-engineering? or do you want to own this BP :-)19:16
Saviqfginther, doing :)19:32
fgintherSaviq, thanks19:33
fgintherSaviq, by the way, now that we have https://blueprints.launchpad.net/ubuntu/+spec/core-1311-ci-airline, I'd like to retool core-1311-upstream-merger-20 some to not overlap as much19:34
Saviqfginther, feel free19:34
Saviqfginther, it does cover a lot of the same ground19:35
sam113101cwayne: good thing, unity7 is kind of buggy20:09
=== seb128 is now known as 3JTAAAD4U
veebersSaviq: are you still around perchance?21:58
Saviqveebers, here22:07
veebersSaviq: hey, I'm experiencing issues with unity when I stop/start maliit-server. Is this a known/expected thing?22:07
veebersSaviq: specifically, this is with the autopilot tests that starts the maliit-server with testability. I can run a single test fine one, if I try run that single test again it will fail22:08
veeberseither the tap won't register on the test app (nor can I tap it myself) or there is some sort of crash22:08
Saviqveebers, not sure, what exactly are you testing?22:09
veebersSaviq: this is the ubuntu-keyboard test22:09
veeberss22:09
Saviqveebers, the only thing I can think of22:10
Saviqveebers, is the OSKController from unity-mir22:10
Saviqveebers, it might not have connected to the keyboard yet22:10
Saviqveebers, you should see some messages about KeyboardInfo in unity8's log when you restart maliit22:10
Saviqthat it disconnected from the socket22:10
Saviqveebers, that's because OSK needs to communicate its geometry to the shell, so that it sets everything up for it22:11
veebersSaviq: I see something like: UbuntuKeyboardInfo - socket error: "QLocalSocket: Remote closed"22:11
veebersI also had this once, repeated heaps:22:11
veebers__pthread_gettid -222:12
veebersterminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<std::runtime_error> >'22:12
veebers  what():  display factory cannot create fb display22:12
Saviqveebers, that means something's hogging the display and unity8 can't start22:12
Saviqveebers, like an app (or maliit itself, for that matter)22:12
Saviqveebers, here's the keyboard info receiver http://bazaar.launchpad.net/~mir-team/unity-mir/trunk/view/head:/src/modules/Unity/Application/ubuntukeyboardinfo.h22:13
Saviqveebers, it's just a socket in XDG_RUNTIME_DIR22:13
Saviqveebers, over which ubuntu-keyboard sends some info about itself22:13
Saviqit's not ideal, but it does the job while we can't come up with a better solution22:14
Saviqveebers, it might be that you need to instrument it and wait for it to reconnect after restarting maliit22:15
veebersSaviq: when you say 'instrument it' you're referring to the better solution?22:17
Saviqveebers, no, the current one22:17
Saviqveebers, I don't think it exposes a property you could listen on22:18
SaviqOSKController.qml has an "enabled" property though22:19
veebersSaviq: right, and the ubuntu-keyboard side would also know when it's connected right? I could also do it on that side?22:19
Saviqveebers, yeah,22:19
Saviqveebers, m_clientConnection22:20
Saviqveebers, in http://bazaar.launchpad.net/~phablet-team/ubuntu-keyboard/trunk/view/head:/src/plugin/ubuntuapplicationapiwrapper.cpp22:20
veebersSaviq: ok understood. I need to wait until the keyboard/shell are connected. I'll see what I come up with22:22
veebersSaviq: hey, if you have a couple more minutes. I might be being dense, but what exactly is unity-mir. It's not a process that's running is it, I'm wondering how (if at all) i might be able to introspect it22:38
Saviqveebers, it's a plugin loaded into unity823:20
Saviqveebers, i.e. import Unity.Application 0.123:21
Saviqveebers, from which OSKController is instantiated in Shell.qml:73923:21
Saviqveebers, and that you can introspect talking to unity823:22
veebersSaviq: ah right, thanks :-)23:22

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!