=== salem_ is now known as _salem [05:20] I am not finding VLC menu in ubuntu unity... can anyone please tell me how to fix it ? === zniavre__ is now known as zniavre [08:05] didrocks, salut! I'd like some Jenkins wizardry on server scopes branches, do you know how I could make that happen? === alan_g|life is now known as alan_g [08:05] davidcalle: ah, here you are :) you want to have upstream merge on those branches, right? [08:06] davidcalle: like merging to trunk? [08:06] didrocks, indeed [08:06] davidcalle: this is for mmrazik ^ [08:07] pstolowski: hi, officially something needs to be done with that timeout test, as it even failed on powerpc now [08:08] davidcalle: can you create a merge proposal for this file/branch: http://bazaar.launchpad.net/~cupstream2distro-maintainers/cupstream2distro-config/trunk/view/head:/stacks/head/unity.cfg ? [08:08] davidcalle: it should be as easy as appending the name of the projects to the file [08:08] sil2100: hmm, ok [08:08] davidcalle: assign the mp to me then (mrazik is my launchpad id) [08:08] I'll approve and create the respective jenkins jobs out of that config [08:09] mmrazik, didrocks, sounds good, thanks :) [08:26] pstolowski: since powerpc failed twice already now with: ERROR:test-scope.c:4078:main_scope_tester_preview_result: assertion failed: (run_with_timeout (ml)) [08:27] sil2100: ok, looking at this test [08:29] pstolowski: thanks! [08:51] mmrazik, https://code.launchpad.net/~submarine/cupstream2distro-config/server-scopes/+merge/165308 [08:52] didrocks: is the stack above (^^^) ok or shall we rather create something like unity-server.cfg? [08:54] mmrazik: I would prefer a separate release and file [08:54] mmrazik: as this is never going to be for daily release [08:55] didrocks: so even release? i.e. out of head? [08:59] mmrazik: yeah, it's not something that will follow releases [08:59] mmrazik: maybe call it server/ ? [08:59] or online/ ? [08:59] didrocks: ok [09:00] davidcalle: I'll take your MP and do what needs to be done [09:00] mmrazik, thanks :) [09:02] didrocks: can you have a look on this (stack membership primarily): https://code.launchpad.net/~mrazik/cupstream2distro-config/dbus-cpp/+merge/165310 [09:04] mmrazik: hum, let's wait for launchpad to answer and show your branch, but platform (from the description) sounds good to me :) [09:04] mhm.. takes ages to generate it [09:04] I'm going to add one more revision anyway... (coverage hooks) [09:04] Saviq, you around? [09:05] nic-doffay, here [09:05] mmrazik: no daily release yet, right? [09:05] didrocks: I don't know. Why would we want to opt out from daily release? [09:06] mmrazik: is it cleaned? checked for daily release? [09:06] tvoss: ^^^ (aboud dbus-cpp) [09:06] Saviq, cool. Quick question, I have an animation which needs to be used in a SequentialAnimation and alone. Can I call it from the SequentialAnimation without triggering the next animation? [09:06] by it's id [09:06] didrocks: what do you mean by cleaned? [09:06] I think we should first ensure that's it's fine and ready for it, checked and released :) [09:06] ok [09:06] didrocks: I'll add daily_release: False then [09:06] mmrazik: remember that if it's failing, the whole stack is blocked [09:06] nic-doffay, I don't see why not, but feels funky :) [09:06] hence the sanity checking [09:06] I can add it to our list of things to do [09:07] but needs to be properly done first :) [09:07] mmrazik: thanks! [09:07] commented [09:07] Saviq, you think I should copy and paste it over? [09:07] nic-doffay, you'd need to make sure the state of the Sequential one after you've ran the one inside is sane [09:08] nic-doffay, it just feels funky, but should be fine :) [09:08] didrocks: thanks [09:08] nic-doffay, after all a SeqAnim is just a wrapper around other Anims [09:08] tvoss: I'll create a dummy MP to lp:dbus-cpp just to see if coverage works as expected in jenkins [09:08] nic-doffay, so it should be possible to control them independently when the SeqAnim isn't doing it [09:09] nic-doffay, did you try already? [09:10] mmrazik, ack [09:11] Saviq, hadn't tried already, but figured I'd ask your opinion on the better way to achieve it so I wouldn't have to go back during the MP review [09:11] nic-doffay, I'm good either way, really [09:16] davidcalle: noticed you added daily_release: False to unity-scope-imdb [09:16] should it be also in the online part? [09:17] nic-doffay: ping [09:18] mzanetti, what's up [09:18] mmrazik, yes, it's being completely removed client side, it might come back online. [09:18] nic-doffay: hey. any news on the Blur? [09:18] mzanetti, yeah. I've done a shader. [09:18] nic-doffay: how's performance on the phone? [09:18] It scales the image by a multiple. [09:18] mzanetti, haven't tried it yet, been busy putting finishing touches on the infographics [09:19] nic-doffay: ok. where's the branch? [09:19] mzanetti, no branch yet, I'll push one if you'd like. [09:19] nic-doffay: yes please [09:20] mzanetti, my suggestion would be to copy and paste the existing GaussianBlur shader code which Qt uses and add the code from my vertex shader to it. [09:20] Call it ScaleWithGaussianBlur [09:20] Or something. [09:21] mmrazik, got a url for me for dbus-cpp? [09:21] Since it can be done in pass, it will make it faster then calling the Scale shader separately and then doing the blur. [09:21] nic-doffay: hmm... I still think that's too slow. even if scaled down. but lets try [09:21] tvoss: url for what? a jenkins job? [09:21] mzanetti, ok I have another idea if it's still too slow. [09:21] I'll push to a branch now. [09:21] mmrazik, yup [09:21] its not ready yet. There is some weird problem with coverage. Just looking into ti [09:21] tvoss: https://jenkins.qa.ubuntu.com/job/dbus-cpp-raring-amd64-ci/3/console [09:22] I don't understand why I'm getting permission denied [09:24] mmrazik, you sure that gcovr is available? [09:24] mmrazik, the packaging setup only installs lcov [09:24] tvoss: yup. gcovr is in the base image as we use it for most projects [09:24] mzanetti, how did the recursive one cope? [09:24] tvoss: even these hooks are used in other projects [09:25] mmrazik, okay ... [09:25] interesting [09:25] mmrazik, seems like it is not even able to access the coverage-xml.cmake due to permission denied [09:25] mzanetti, a small one https://code.launchpad.net/~unity-team/unity/phablet.fix-run-device-mousetouch/+merge/165321 [09:25] tvoss: I don't think so. gcovr is apparently executed [09:26] and its gcovr that throws the OSError [09:26] Saviq, mzanetti: Is implicitHeight meant to work with expanding/changing layouts? Expanding works, but shrinking does not. [09:26] MacSlow: should work I'd say [09:26] mmrazik, ah [09:26] MacSlow, yeah, it should be used in stead of height [09:26] tvoss: so it looks like the result can't be saved [09:26] Saviq: ack (on the review) [09:26] or something like tht [09:27] Saviq: let me test the recursive one. not much hope tho... [09:27] MacSlow, what's your usecase? column with expanding/shrinking things? [09:30] MacSlow, http://pastebin.ubuntu.com/5693130/ works, so I'd say yes [09:32] hey. I've just apt-get upgraded and now the shell segfaults with: Settings schema 'com.canonical.Unity.Dash' does not contain a key named 'home-lens-ordering' [09:32] I did merge with trunk [09:34] Saviq: any ideas? ^ [09:35] mzanetti, hum, sounds like some version mismatch [09:36] ok... just wiped unity_build and currently rebuilding. lets see [09:36] nope... still crashing [09:37] mzanetti, ah, you mean the qml shell crashes? [09:37] mzanetti, Saviq: hm... I see... that's the issue at work btw... http://ubuntuone.com/0Eh6U1vGofYb6sHQjF3sl6 [09:37] Saviq: yes [09:38] is it about the overall height of the notification? [09:38] I wouldn't use implicitHeight there tbh [09:38] mzanetti, apt-cache policy unity-common [09:39] Saviq: http://paste.ubuntu.com/5693149/ [09:39] MacSlow, that's unrelated to implicitHeight [09:39] MacSlow, it worked in my example after all [09:39] mzanetti, yeah, daily-build-next [09:39] Saviq: not good? [09:40] mzanetti, evidently not [09:40] f*** those ppas [09:40] mzanetti, looks like you got 100scopes [09:41] mzanetti, why wouldn't you use implicitHeight? isn't that exactly what it's there for? [09:41] Saviq: I'd say no. but I know that our opinion on that differs ;) [09:41] ;) [09:43] didrocks, does daily-build-next raring build unity7 off of trunk? [09:43] Saviq: yeah [09:44] didrocks, that on purpose? ;) [09:44] Saviq: why shouldn't it? :) [09:44] Saviq: that's how daily build works :p [09:44] Saviq: works again. thanks [09:44] didrocks, well, it's never going to land in raring, is it [09:44] didrocks, but yeah, I get it [09:44] Saviq: yeah, daily-build-next is while we can't upload to saucy [09:44] so normally until saucy opens [09:44] but the issue is that you had a lot of transitions in touch [09:45] like autopilot 1.3 [09:45] got it [09:45] and all is interelated [09:45] so we'll have to land everything at once in saucy :) [09:45] not the desired behaviour, then [09:45] which is next week I guess [09:45] well, desired behaviour… until saucy opens [09:45] now, converging with touch made it hard… [09:45] yeah, I get it [09:49] Saviq: RecursiveBlur is another order of magnitude slower than the Gaussian one :/ [09:49] mzanetti, yeah, expected that [09:49] +1 === alan_g is now known as alan_g|tea [09:53] MacSlow, got a fix for you [09:53] MacSlow, use "contentColumn.height" instead of "childrenRect.height" in line 33 [09:54] Saviq, already tried that... but then I've have issues with bottom margins not being correct [09:54] MacSlow, still, that gets you closer, doesn't it ;) [09:55] mmrazik, any luck? [09:55] Saviq, sure... but it's trading one issue with another :) [09:55] MacSlow, childrenRect is difficult like that [09:56] MacSlow, height: implicitHeight + spacing * 2 [09:56] MacSlow, line 57 [09:56] tvoss: not really. I think there is something broken in dbus-cpp as I'm unable to run coverage-xml locally as well [09:56] but don't know what it is yet [09:56] Saviq, and when I did the expansion/collapsing once the bottom margin is correct for the collapsed snap-decision... *sigh* [09:57] MacSlow, the two changes above fix everything for me, no? [09:57] Saviq, let's see that second change... [09:58] Saviq, ah indeed... interesting [09:59] MacSlow, only thing now is there's a spacing-high jump at the beginning and end [09:59] MacSlow, and actually, get rid of the height of the column altogether [09:59] Saviq, yeah... that's what I'm seeing here too [09:59] tvoss: oh [09:59] I think I start to know what it is [09:59] MacSlow, and go for "implicitHeight: contentColumn.height + contentColumn.spacing * 2" for the UbuntuShape [10:00] give me a sec [10:01] MacSlow, that jump is because of spacing :/ [10:01] Saviq, ok... the height jump is still there... but since this is just a temp. solution for the expansion until we've the proper UI-element in the toolkit I can live with this [10:02] MacSlow, I'll fix it, gimme a chance ;) [10:02] mmrazik, coverage-html not working either ... seems like a parsing issue [10:02] Saviq, ok... go wild! :) [10:02] tvoss: oh. ok. For a second I thought it is because of non-existing directories from --exclude (copy&pasted from mir) [10:02] but it is not that [10:03] hmm [10:03] mmrazik, might be the case, too [10:03] mhm [10:03] maybe it is in the end [10:03] I'll create a branch and will see if it works in jenkins [10:04] tvoss: oh... I think its the missing CMAKE_GCOV variable [10:04] its not defined anywhere AFAICS [10:05] mmrazik, nope, issue is that I'm running the tests under dbus-test-runner by default === alan_g|tea is now known as alan_g [10:05] mmrazik, are you building in a chroot? [10:05] tvoss: yes [10:05] tvoss: but the tests are passing [10:08] mmrazik, sure, but that's not the issue. the coverage files are not placed in the right directory [10:10] Saviq, I'll keep going at the test to work again [10:18] MacSlow, here's a diff against your current branch http://pastebin.ubuntu.com/5693233/ [10:19] Saviq, ok... trying it now... [10:22] MacSlow, I'd also do http://pastebin.ubuntu.com/5693239/ to reduce the opacity change weirdness [10:23] MacSlow, as QML by default applies the opacity to all the underlying objects [10:23] MacSlow, which results in the avatar and the red button, for example, apparently disappear later than the other elements [10:24] Saviq, never noticed that before [10:25] MacSlow, do you see the difference, though? [10:26] MacSlow, you can use --slow-animations to qmlscene to notice it more [10:27] MacSlow, you can read about layers here http://qt-project.org/doc/qt-5.0/qtquick/qml-qtquick2-item.html#layer.effect-prop [10:28] mzanetti, btw, re implicitHeight, "Setting the implicit size is useful for defining components that have a preferred size based on their content, for example" [10:28] mzanetti, http://qt-project.org/doc/qt-5.0/qtquick/qml-qtquick2-item.html#implicitWidth-prop ;) [10:29] Saviq: yes [10:29] Saviq: but the notifications don't have a preferred size, they have a definiteve size [10:29] definitive... its not that they would like to be sized like this. they _have to_ be sized like this [10:29] mzanetti, not necessarily, whoever uses them can decide otherwise [10:29] because they want to [10:30] Saviq: they are not components to be used somewhere else [10:30] at least not the way they are now [10:30] mzanetti, yeah, well, I just like the property there to be noticed, I think [10:31] Saviq, using the --slow-animations switch makes the interactive test act all weird here [10:33] Saviq, but I see now what you mean regarding the opacity of avatar-icons and the tinted button... although it's just a very subtle difference [10:33] MacSlow, it's still there, even if subtle :D [10:33] MacSlow, I can see it without --slow-animations, so ;) [10:34] Saviq, no doubt... thanks for the heads up... I'll keep that change and will remember the --slow-animations switch for visual debugging in the future [10:34] MacSlow, we should probably make more use of layers when animating opacity on complex components [10:34] mzanetti, WDYT ^? [10:34] mzanetti, http://pastebin.ubuntu.com/5693239/ [10:35] makes sense to me. BUT: its not something that one should do without understanding what happens [10:35] Saviq: ^ [10:36] Saviq, I added a comment to that, so one (me) doesn't forget why it's there :) [10:36] because it can slow down things if used wrong [10:36] mzanetti, yeah, of course, but should actually speed up in that case [10:36] I'd say, yes [10:36] is it that much of a problem tho? [10:37] Notifications don't seem that complex to me that it would require this already [10:37] didrocks: uuuh oooh, nooo~! Something happened and unity tests did not run [10:37] MacSlow, add populate.running || there, too [10:37] mzanetti, I can see the visual difference between the two [10:37] how can I test? [10:37] mzanetti, when you apply the opacity animation without the layer [10:37] mzanetti, merge lp:~macslow/unity/phablet-snap-decision-action-expansion [10:38] sil2100: mind checking with the QA people? [10:38] and $ qmlscene --slow-animations -I builddir/tests/utils/modules/ tests/qmltests/Notifications/tst_Notifications.qml [10:38] damn ^W [10:38] :D [10:38] Saviq: ping [10:38] dednick, pong [10:38] mzanetti, it's easiest to see with the snap decision [10:39] mzanetti, the buttons / avatar "disappear" at a different rate than the rest [10:39] due to the contrast between them [10:39] didrocks: actually, it seems unity got removed before the test, huh [10:40] Saviq: re this indicator-client move. I've moved everything into unity, and also done first pass of the indicator file monitoring. But until we have those .indicator files, we cant really remove the plugin framework that was done (because that's how it determined which indicators to load). unless i just hardcode them to load manually. [10:40] mzanetti, it's not about the speed in this case, it's about the transition [10:40] Saviq: or i can export the files from unity for the time being [10:41] larsu, ^ any comment on when the .indicator files will be available? [10:41] dednick, I think it's fine if we "ship" the .indicator files in the mean time [10:42] dednick, and then just flip the switch to read from the real place [10:42] dednick, when they're there [10:42] dednick, assuming we know its format [10:42] Saviq: cool. thats what i was hoping we could do. [10:42] hmm... can't really see any difference here [10:42] sil2100: is the ppa in a consistent state? [10:42] Saviq, mzanetti: pushed the fixes sofar... thx for the help! [10:44] Saviq: hmm... can't really see any difference here [10:44] MacSlow: found a bug [10:45] MacSlow: click 10 times on "add a snap-decision" [10:45] MacSlow: click 10 times on "remove f1st" [10:45] MacSlow: add some snap decisions again => they are not on top any more [10:46] MacSlow: repeat that a couple of times and notifications stop showing up at all [10:47] mzanetti, the model doesn't seem to get cleared correctly [10:47] didrocks: I just dist-upgraded from it and it's ok here locally, I'll look into what's going on [10:48] sil2100: it's still wrong that it's trying to install everything :/ [10:48] mzanetti, using clear model seems to fix it again [10:48] MacSlow: no... that works around it [10:48] :P [10:48] MacSlow: but yes. clearing the model restores a good state [10:49] mzanetti, :) [10:49] mzanetti, first I need to get rid of warnings and get the tests to work again [10:51] mzanetti, let me make a video ;d [10:55] MacSlow, btw, the "delay behavior until opacity == 1" is a hack, but we don't have a good solution for such delays [10:56] Ah ha [10:56] didrocks: I think I found something, let me take a closer look [10:56] Saviq, I guessed [10:56] ok ;) [10:56] sil2100: fyi, I hope https://code.launchpad.net/~stolowski/libunity/fix-test-timeout/+merge/165343 to fix the issue with occasional test failures [10:58] morning all. Am in a spot of trouble here: http://askubuntu.com/questions/298881/13-04-unity-graphics-suddenly-broken-for-no-reason === mmrazik is now known as mmrazik|lunch [11:00] since my home folder is on a separate partition, I was wondering if it might have stored a faulty setting somewhere. where are these settings stored (if in home) and am I safe to delete the entire folder etc? [11:06] andrewebdev: should be able to reset compiz settings with "dconf reset -f /org/compiz/" (you might need dconf-tools package installed first) === MacSlow is now known as MacSlow|lunch [11:07] pstolowski: thanks! Will review [11:08] andrewebdev: handy way to check if it's a setting issue: log in as a guest user and see how things look there. [11:08] sil2100: please don't globally approve, I'd like jamesh to take a look as well; [11:09] sil2100: and can you run these tests locally? [11:09] mzanetti, http://pastebin.ubuntu.com/5693352/ should work, no? or am I missing something obvious? [11:11] Saviq: can't see anything obvious. Does it print any errors? [11:11] mzanetti, no, it just doesn't work [11:12] mzanetti, I tried a Connection { ListView.view.add; onRunningChanged } < that fails with [11:12] Cannot assign to non-existent property "onRunningChanged" [11:12] uuuu [11:13] mzanetti, but ListView.view.add is the Transition object alright [11:13] * greyback bbiab [11:13] greyback, YES! Guest session works perfectly! [11:13] Saviq: how I read the docs, yes. But those tansitions are implemented a bit weird [11:14] greyback, nooo wait... spoke too soon [11:15] the moment I set up the dual monitors in the guest session, unity crashes and I lose all window decorations etc. [11:15] mzanetti, obviously that's a poor-man's solution as there's nothing to say that the transition wouldn't relate to some other delegate [11:15] mzanetti, so in that sense the opacity solution is better [11:16] Saviq: oh.... might be that the Notification is constructed and the implicitHeight changed _before_ it's actually hooked into the listview and the add-transition triggered on it [11:17] mzanetti, well, I could actually get access to the running property in onOpacityChanged [11:17] mzanetti, and it held the correct value [11:17] mzanetti, but it seems there's no NOTIFY onit [11:18] Saviq: I would add a bool property to Notification like "animationsEnabled: false" and bind it in the listview [11:18] mzanetti, still, that would be global, not per-delegate, which we're after [11:18] Saviq: opacityChanged is triggered by the add-transition. so most likely later [11:19] mzanetti, ah you mean the binding fails? [11:20] mzanetti, because the ListView.view attached prop isn't there yet? certainly was there in Component.onCompleted [11:20] mzanetti, anyway, that's a slightly scientific discussion ;) [11:20] Saviq: I mean by the time the binding is evaluated its not in the listview yet and thus evaluates to true. The animation is triggered and only then the binding starts working [11:20] yes, it is [11:20] mzanetti, enabled never changes [11:20] hmpf... [11:20] on the behaviour [11:21] anyway === pete-woods is now known as pete-woods-lunch [11:34] didrocks: ok, so it seems that Brandon's latest nux commit bumped the ABI version and he didn't bump the changelog upstream version [11:34] didrocks: yesterday's commit ;/ [11:42] yay, 1 bugfix for ListView in! [11:42] https://codereview.qt-project.org/#change,56583 [11:43] now to bother Mirv :D [11:47] :D [11:48] didrocks: so, we'll need to rebuild all reverse-depends of nux now [11:48] didrocks: and probably get this one in: https://code.launchpad.net/~sil2100/nux/bump_version_due_to_abi/+merge/165352 === pete-woods-lunch is now known as pete-woods === alan_g is now known as alan_g|lunch === mmrazik|lunch is now known as mmrazik === MacSlow|lunch is now known as MacSlow [12:16] sil2100: ok, please take care of that :) [12:17] sil2100: also keep brandon updated about it. Everything is explained in the FAQ… They should start to know it [12:17] bregma: can you pass the message in your team about it ^ [12:17] sil2100: maybe add a message to the changelog as you are bumping it? [12:18] sil2100: this is what made the installation failing? [12:18] didrocks: ok, will add an entry [12:19] didrocks: well, it was because components were built in wrong order, unity was built against the older nux and then nux got built, so when we're fetching the newer nux, it uninstalls unity as the abi-version mismatches [12:19] sil2100: makes sense, so you need to bump the build-dep on the unity side [12:19] sil2100: and then, we rebuild the stack [12:19] Saviq: we already have some, but we can't land them until the current (gtk) panel properly loads them (so that we don't have regressions). I'm working on that this week, hopefully it lands early next week [12:19] (as per the FAQ ;)) [12:19] didrocks: also, we'll have to take care of one thing... as indicators are not releasing, I asked cyphermox yesterday to maybe manually release indicator-appmenu [12:19] didrocks: we need the new indicator-appmenu for unity-gtk-module to work [12:20] sil2100: yeah, but with jenkins shutting down yesterday… I think cyphermox will redo it today? [12:20] didrocks: but neither daily-build-next nor next have the new indicator-appmenu there as well [12:20] didrocks: ok [12:21] Saviq: mzanetti: I've pulled out the window management stuff in shell into a separate component ready for testing. I think it makes sense to rename SideStage directory to something like Stages, and move Components/Stage.qml and the new WN component in there too. Opinions? [12:22] it centralizes all the window management related code into one directory [12:22] greyback: +1 [12:23] sil2100: ping me once you have the unity branch [12:23] sil2100: approved that one [12:23] greyback: do you already have information about running apps etc from Mir? [12:24] didrocks: could you have a look, please: https://code.launchpad.net/~mrazik/cupstream2distro-config/online/+merge/165355 [12:25] mmrazik: approved [12:25] didrocks: thanks [12:25] mmrazik: I think I'll add a hook to cu2d-update-stack with a general config file in a directory [12:25] didrocks: oh.. [12:25] mmrazik: just to prevent deploying this directory [12:25] wait a second [12:25] didrocks: it will not land [12:26] mmrazik: sent back to "needs review" [12:26] there is duplicate target_branch... forgot to remove the imdb scope from head [12:26] ah ok :) [12:30] mzanetti: not yet, the platform api needs to be stabilized, but it should be soon [12:30] didrocks: https://code.launchpad.net/~sil2100/unity/change_libnux_req/+merge/165357 [12:30] didrocks: but I guess CI will fail on this one due to libnux 4.0.2 not being available in the PPA, right? [12:34] Lunch [12:35] didrocks: new diff finally generated: https://code.launchpad.net/~mrazik/cupstream2distro-config/online/+merge/165355 [12:37] sil2100: they have a local repo, so it should work, not sure it will depwait, so just tell me once the nux merge is… merged :) [12:37] mmrazik: are you sure imdb won't be client side? [12:38] sil2100: do you know? ^ [12:38] didrocks: from what David said, no... [12:38] If I remember correctly [12:39] * sil2100 gone [12:39] didrocks: this is what davidcalle says: [12:39] ok :) [12:39] mmrazik, yes, it's being completely removed client side, it might come back online. [12:39] perfect, approved! :) [12:39] thanks === alan_g|lunch is now known as alan_g === _salem is now known as salem_ [13:01] dandrader, huh, wrong channel ;) [13:02] dandrader, there's an issue with the mouseTouch [13:02] dandrader, it gets stuck when you try to swipe the dash, for example [13:03] Saviq, you mean flicking horizontally to switch from dash to dash? [13:03] dandrader, yes [13:03] dandrader, the whole UI gets stuck (probably some mouse event does not get converted to touch) [13:03] dandrader, it's only when mousetouch is enabled, of course [13:04] dandrader, also I failed the review https://code.launchpad.net/~unity-team/unity/phablet.fix-run-device-mousetouch/+merge/165321 [13:05] ah [13:09] Saviq, couldn't reproduce the issue so far? How easy is to reproduce it? [13:09] s/far?/far. [13:12] dandrader, 100% [13:12] weird [13:12] dandrader, just enable mousetouch [13:12] dandrader, and release your mouse cursor when flicking [13:13] so yeah, flicking, not swiping [13:13] dandrader, actually [13:13] dandrader, flick + click when still moving [13:14] dandrader, so, grab, drag quickly & release, try and grab again [13:14] dandrader, like you'd try to speed it up or reverse the movement [13:14] dandrader, any Flickable, really [13:14] Saviq, now I got it! [13:15] dandrader, must be some event that we don't convert to touch or something [13:15] greyback, +1 [13:15] Saviq, it got stuck into a kind endless stream of MousePress and MouseRelease [13:15] Saviq: cool, almost done by now anyway :) [13:16]