=== duflu_ is now known as duflu [07:23] hey Saviq, how was your week-end? === iahmad is now known as iahmad|afk [08:36] Saviq: still sleepy? ;) [08:39] didrocks, wassup? [08:41] didrocks, if it's about the unity8 crash, I haven't been able to retrace it (just ?? everywhere) [08:42] Saviq: we still get it in the latest image [08:42] Saviq: so, seems like a popular crash [08:42] didrocks, it doesn't affect the tests does it? [08:42] didrocks, as it's on exit? [08:43] Saviq: we aren't sure it's on exit, my guess would be it is [08:43] Saviq: are you certain it's on exit? [08:43] didrocks, so it doesn't affect UX either [08:43] didrocks, yes [08:44] Saviq: ok, in that case, better to keep an eye on it, but not a blocker [08:44] didrocks, yeah we are [08:44] Saviq: thanks for the additional infos! [08:45] didrocks, here's one on exit we managed to retrace: bug #1253685 [08:45] bug 1253685 in unity8 (Ubuntu) "unity8 crashed with SIGSEGV in QMetaObject::activate()" [Medium,New] https://launchpad.net/bugs/1253685 [08:45] didrocks, but we don't really want to spend time fixing Qt 5.0 still, when 5.2 is around the corner and solves most of these for us [08:46] (and comes with others that we need to be fixin') [08:46] Saviq: agreed [09:09] sil2100, morning, could you finish the unity-scopes-api packaging cleanup, so we can get it in distro? [09:10] sil2100, there are some comments on the licence branch [09:16] mhr3: sure [09:20] tsdgeos, I didn't kill the headache friday :D [09:20] I had to stay in bed till 8pm [09:20] Cimi: better? [09:20] yeah now I'm great [09:20] \o/ [09:20] there was a beautiful sunrise this morning [09:20] took a pic [09:20] mmm [09:23] http://i.imgur.com/09PG2jQ.jpg [09:23] Cimi, nice one indeed! [09:23] Saviq, it's the 55-210 with the nex 6 [09:24] Cimi, why the vignetting? [09:24] Saviq, because I like it ;) [09:24] Cimi, got it :) [09:24] Saviq, but yeah in this pic I can remove it === iahmad|afk is now known as iahmad [09:25] Saviq, but was 9am and decided to postpone the editing :) [09:25] Cimi, it looks nice, don't get me wrong, was wondering if it's the lens or maybe some filter [09:25] Saviq, I usually put vignette on purpose [09:25] Saviq, this time it's due to the contrast [09:26] Cimi, mhm [09:26] Saviq, I need to bump the contrast and play with lights in a way I keep the black shape of buildings while making the rest brighter [09:26] Cimi, yup [09:26] Mirv, hey, tsdgeos fixed qtubuntu to build against rc1 [09:27] tsdgeos, although it stopped building against beta, that expected? [09:27] https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2/+build/5255710 [09:27] Saviq: hrrmmm, lol [09:28] ah it's the urlinterceptor that was made public, private, public, private... [09:28] yeah [09:28] so as long as it builds against rc1 that's fine [09:28] i can remove the QtQuick include again [09:28] :D [09:32] tsdgeos, let's not target the beta ;) [09:32] Saviq: https://code.launchpad.net/~aacid/qtubuntu/no_qtquick_megaheader/+merge/196503 [09:33] tsdgeos, ah it's only that [09:33] tsdgeos, then yeah [09:33] yeah [10:06] mzanetti, got you a functional review of the expandables https://code.launchpad.net/~mzanetti/ubuntu-ui-toolkit/expanding-listitem/+merge/195602/comments/454602 [10:06] Saviq: thanks [10:07] Ubuntu One - Service Temporarily Unavailable [10:07] ugh :? [10:08] mzanetti, try refreshing, worked for me after a few seconds [10:08] mzanetti, either way also available at http://people.canonical.com/~msawicz/unity8/ [10:10] Saviq: ok. works again [10:11] Saviq: they do collapse when you click outside, given they are used inside a Expandables* [10:11] Saviq: if you use them just standalong, they can't [10:11] mzanetti, which point are you talking about? [10:11] Saviq: "Shouldn't it collapse as soon as you touch outside?" [10:12] mzanetti, touch != tap [10:12] ah [10:12] Saviq: from what I understood from design this is how it should be [10:12] let me dig out the conversation [10:12] mzanetti, those questions aren't really for you :) [10:12] mzanetti, but to someone from design that will review this [10:15] Saviq: fyi: https://pastebin.canonical.com/100949/ [10:17] mzanetti, yup - seems the only missing thing is the 'continue scrolling when reached the end' [10:18] mzanetti, but as we talked before, that's tricky from the users' perspective, I think [10:18] Saviq: yeah. which I told them in a later conversion about the issues we have with that [10:18] mzanetti, right, I think it's fine to omit this for now [10:18] mzanetti, even when we get "there" it shouldn't change the API, just the behaviour [10:18] so we can fix later [10:19] yep [10:19] * mzanetti nods [10:20] Saviq: the two ThinDivider's is weird. I don't add a second one... [10:21] also I can't see it here. but it's clearly there in your video [10:21] mzanetti, ah! it's only there when overshooting [10:22] mzanetti, hence only visible on the Expandables*, but then it's not there without your branch... [10:22] tricky... [10:23] humm... [10:23] mzanetti, hmm must be some visual glitch maybe [10:23] mzanetti, can't see it when maximized either [10:24] yikes [10:24] Saviq: I can't see it here, no matter what I try. But I can see the UbuntuShape in the expanded area painting a bit weird [10:24] it's actually there on mouseY = foo, not there on mouseY = foo + 1 [10:25] ah. no. now I see it [10:25] mzanetti, right, maybe it's actually the top of the shape [10:25] mzanetti, it must be [10:25] mzanetti, it's a rounding error [10:26] Saviq: where is this mouseY you are referring to? [10:26] mzanetti, my pointer ;) [10:26] ah [10:26] mzanetti, I meant that if I drag by foo pixels, it's not there, but if I drag one more pixel - it's back [10:26] mzanetti, should be solved by the 'content overlays divider' anyway [10:26] mhm... yeah. that does sound like a rounding error indeed [10:29] Saviq: the scopes.cpp code is now in whch bzr repo? [10:29] tsdgeos, lp:unity-scopes-shell [10:30] Saviq: do you mind me add a title property to the scopes model so it's easier to be used with the new SDK tabs? or prefer me to add an identitymodel that adds that new role? [10:30] s/title property/title role [10:30] tsdgeos, add away [10:45] Saviq: fixed the divider overlay glitch. will talk to design about the rest [10:45] mzanetti, cheers [10:48] Saviq: one issue with the immediate expanding on click is that you can't really only collapse all the items any more [10:48] mzanetti, true [10:48] (katie and christina are not around yet) [10:48] but I will ask them about this in any case [10:48] mzanetti, I'm just poking holes, if they plug them that's completely fine [10:49] yeah, sure [10:54] Saviq: https://code.launchpad.net/~aacid/unity-scopes-shell/addscopestitle/+merge/196514 [10:55] mhr3, ↑ [10:59] approved [11:00] Saviq, btw i'll need someone who'll do reviews for stuff in that branch, who's going to be the lucky one? :) [11:01] Saviq, i supposed tsdgeos/mzanetti/dednick are quite familiar with models and qt [11:02] mhr3, there's plenty of us that can do that [11:02] mhr3, just bug us [11:03] ok [11:25] /food [11:29] Saviq: mhr3: updated this: https://code.launchpad.net/~mzanetti/unity8/music-preview/+merge/193803 and created this: https://code.launchpad.net/~mzanetti/unity-scopes-shell/musicpreviewtrackmodel/+merge/196522 [11:32] mzanetti, cheers [11:32] mzanetti, hi, would you mind adding a harmless feature to ^ : not displaying the track length when it's == 0. There are cases where the music provider doesn't expose it and displaying 0 feels wrong. [11:33] davidcalle: makes sense [11:34] is there a way from qml to get the type name of the component as string? [11:36] mzanetti, thx :) [11:36] /lunch [11:44] Saviq, tsdgeos ^ [11:45] Cimi: what you need that for? [11:45] tsdgeos, having a clever renderer name default [11:45] objectName? [11:45] tsdgeos, it weirdly didn't work [11:47] no no [11:47] i mean what do you mean by "renderer name default"? [11:59] mhr3, you might want to look at https://code.launchpad.net/~mzanetti/unity-scopes-shell/musicpreviewtrackmodel/+merge/196522 too === _salem is now known as salem_ === MacSlow is now known as MacSlow|lunch [12:05] mzanetti, https://code.launchpad.net/~mzanetti/unity-scopes-shell/musicpreviewtrackmodel/+merge/196522/comments/454656 [12:07] mhr3: hmm... dunno. I don't use them currently. We could as well drop those 2 roles [12:07] Saviq, what do you think about unity8 publishing a d-bus interface for taking screenshots? [12:07] mzanetti, pls do [12:07] mhr3: ack [12:10] mhr3: done [12:11] dandrader, don't have better ideas, TBH [12:11] mzanetti, thx, approved [12:13] Saviq, so it's a sensible approach then? [12:13] dandrader, wonder if it shouldn't go through the Mir connection, TBH [12:13] dandrader, especially once we say we want screencasts, not only screenshots [12:14] dbus won't be a good choice then [12:14] tsdgeos, a name for the denreder [12:14] dandrader, and it shouldn't be something unity8-specific, really, have you talked to Mir folk? [12:14] tsdgeos, so in genericscopeview I can do if typeName != carousel do... [12:15] property string rendererName is what I have now [12:15] Cimi: typeof ? [12:15] Saviq, just asked alf_ on how to take a screenshot using mirserver api. [12:15] mmm let's see [12:16] Saviq, and why dbus would be appropriate for screencasts? I was thinking about the resquestor saying "hey, take a screenshot and put it on /foo/bar.jpg" [12:16] Saviq: do you have a package for the upstart signals handy now? [12:17] tsdgeos, weird [12:17] the main target for the screenshots feature being autopilot [12:17] tsdgeos, I'll just use rendererName [12:17] atm [12:18] Saviq, so, if I understood you correctly, you think a screenshot feature should be in the mir client API? [12:18] dandrader: I'd say so too, yes [12:18] disclaimer, I don't have a strong opinion on the subject myself, that's why I'm asking around [12:19] mzanetti, almost... [12:20] maybe we should tell the ci guys about all those otto failures :D [12:20] dandrader, it depends on the level of the API we're thinking [12:20] dandrader, but once you want to implement something as functional as kazam [12:20] dandrader, where you can select a window to screenshot / screencast [12:20] the FPS [12:21] dandrader, we'd either have to maintain a bug chunk of code (and API) somewhere between mir and unity8 [12:21] Saviq: right, I meant to answer you but I was preparing for a hangout and then forgot :) yes, I built qtubuntu for armhf already in the morning. I guess more modules rebuilds needed though still for u8 to start. [12:22] Mirv, ok, please let us know if you need help [12:22] tsdgeos, ok I need a different head here [12:22] dandrader, or provide a low-level API and let the "other side" take care of encoding / compression / whatever [12:23] Saviq: well if you wouldn't mind, try upgrading from qt5-daily and see if what kind of problem you hit with starting unity8. I somehow stuck with something already bound to xxx message even though I did initctl stop unity8 first. [12:23] tsdgeos, line 450 of GenericScopeView [12:23] tsdgeos, if you open the preview of the carousel [12:23] swipe between the preview items [12:23] Mirv, tried removing the mir_socket from /run/user/32011? [12:23] tsdgeos, this Y changes [12:23] :-\ [12:24] dandrader, in which case an API of "give me a buffer of region x,y,w,h, 15 times a second" would be enough, and the app would be responsible for doing what it needs doing [12:24] dandrader, but for that dbus wouldn't be appropriate, IMO [12:24] Cimi: shoot! [12:24] Saviq: thanks for the tip, trying [12:24] tsdgeos, tried printing y f the carousel but it's always the same [12:24] 0 [12:24] something moves [12:24] dunno [12:25] Saviq: ok thanks, now I got a symbol lookup failure, no further help needed [12:26] Cimi: yeah it's the list that moves [12:26] the carousel is stuck at it's position [12:26] and since the carousel is inside a loader [12:26] it's always 0 [12:26] maybe the loader pos moves though [12:26] yeah [12:26] but why it doesn't for the flitergrid? [12:29] Saviq, yeah, passing actual pixel data through d-bus is no-go [12:31] dandrader, yeah, that's why I'm saying it depends on how far we wanna go now [12:33] dandrader, if we can define a more-or-less-final API for this already, and just write a small utility around it, that'd probably be better than implementing an interim solution [12:33] dandrader, OTOH protecting that API is important as well - we don't want any app to be able to take screenshots - we need to mediate it through apparmor [12:34] dandrader, and while it's doable otherwise, it's certainly easier with DBus [12:35] Saviq, that's why I thought about d-bus, because you can "easily" add access control to the interface later through d-bus policies [12:36] Saviq, and such an api, at least for screenshots, would be dead-simple. the screencast use case is likely in a whole different league [12:37] dandrader, sure, but then there's no reason why the utility around the "real" mir screen{shot,cast} API couldn't wrap it and expose the DBus API [12:38] dandrader, FWIW that'd be a transition strategy ↑ [12:38] dandrader, so yeah, I'm of two minds :) [12:39] dandrader, it'd be better / more future-proof to write it proper [12:39] dandrader, but then it'd be easier to just have it going, especially if we don't have a plan for the future+proper solution yet [12:40] tsdgeos, I think it's the fact it's still moving [12:40] tsdgeos, the open effect is not finished [12:40] hmmm [12:40] need more context :D [12:41] Cimi: what is still moving and when? [12:41] dandrader, re: jenkins, there seems to be a problem with the mediumtests runner [12:41] tsdgeos, i's expanding [12:41] dandrader, ev is on it [12:41] tsdgeos, currentIndexChanged of previewListView runs and sets itemY for the first time using a value that is not finalized [12:42] mzanetti, should we not have bumped unity-scopes-shell version? is it safe if unity8 will upgrade without scopes-shell? [12:42] mzanetti, I mean for the music preview? [12:42] Cimi: is that a new bug? [12:42] Cimi: or just happens in your branch? [12:42] Saviq, "ev"? [12:42] mmm I dunno [12:42] Saviq: yes, the upgrade should be safe. still, bumping the version probably would have been a good idea [12:42] dandrader, Evan Dandrea [12:42] tsdgeos, carousel preview is broken in trunk [12:43] mzanetti, ABI changes, doesn't it? [12:43] mzanetti, OTOH nothing's linking against it [12:43] Saviq: it's a qml plugin... it's resolved at runtime [12:43] mzanetti, yeah, that [12:43] Cimi: ah ok, i thought this was on the base-renderer branch [12:44] Saviq: but in any case I think we have 2 TODOs here: [12:44] mzanetti, so yeah, nvm, if u8 won't fail regardless, we're fine not bumping [12:44] Saviq: - add scopes api to unity-api [12:44] mzanetti, +1, when it's stable-ish [12:44] Saviq: - add some check that requires version number bumping on api changes in jenkins for unity-api and friends [12:44] There is something wrong with the order indicators are show. I drag the messaging indicator, it shows bluetooth indicator [12:45] Cimi: it's broken as in "doesn't show up at all"? [12:45] mzanetti, that will be part of the first one - API tests in unity-api [12:45] mzanetti, since that will check for the type of the prop [12:45] mzanetti, we could look at what UITK has - the .api files - /me doesn't know what they're about [12:46] btu they seem to help with API stability [12:46] Saviq: yeah. that's great stuff [12:56] Saviq: we're green again :) https://code.launchpad.net/~mzanetti/unity8/music-preview/+merge/193803 [12:56] mzanetti, good [13:00] tsdgeos, on trunk, no [13:00] tsdgeos, works with my branch [13:00] tsdgeos, but there is this bug and I don't know why [13:14] tsdgeos, yeah contentY of the flickable containing the loader changes with the carousel [13:14] tsdgeos, doesn't with the filter [13:16] might be some stuff here... [13:16] in previelistview === alan_g is now known as alan_g|lunch [13:29] mzanetti, packages for testing sigstop http://people.canonical.com/~msawicz/sigstop.tar.xz [13:30] Saviq: cheers [13:30] mzanetti, real-life thing to test: drop the "sleep 2" script from /usr/share/upstart/sessions/maliit-server.conf [13:30] \o/ [13:31] I don't know === MacSlow|lunch is now known as MacSlow === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader [13:57] Cimi: ok, push everything and i'll have a look [14:03] tsdgeos, thanks [14:03] tsdgeos, I need to change task [14:03] tsdgeos, not going anywhere on this [14:10] Cimi: pushed? [14:10] tsdgeos, it is [14:10] ok [14:10] tsdgeos, not my tests [14:10] tsdgeos, like print out [14:10] that's ok [14:11] i'll have a look today, not now, need to finish some tests in the tabbar+dash thing [14:11] Cimi: can you tell me exactly what i do to see the problem though [14:11] ? [14:11] tsdgeos, open preview with carousel [14:11] tsdgeos, look at theY position of the crusel at the bottom [14:12] tsdgeos, swipe the preview and notice the Y changes [14:12] * Cimi cooks a burger [14:23] Saviq: works with Mir and SF [14:23] mzanetti, cool [14:23] Saviq: and wow! just was reminded how fast the Galaxy Nexus is with SF [14:23] mzanetti, ;) [14:23] * mzanetti wants that with Mir :D [14:23] mzanetti, I pushed some tweaks to the test today [14:24] mzanetti, so please re-review and let's land this [14:25] ack === alan_g|lunch is now known as alan_g [14:46] * mzanetti installs emulator [14:58] greyback: Saviq: both approved. works fine here [14:58] mzanetti: cool, thanks! [14:58] mzanetti, the unity8 won't land before unity-mir builds, though [14:58] sure [14:59] mzanetti, unapproved [14:59] mzanetti, will take care of it, thanks [15:12] dammit. note to myself: don't press ctrl+c in the emulator's console [15:14] mzanetti, yeah just don't use it at all - ssh / adb as usual [15:14] yep [15:15] mzanetti, or well, this is a bug I'd say [15:15] seems so [15:15] you know what sucks :D. When you see the display of your phone lighting up. You know there is an incoming phone call with a hidden number but you just can't accept it. [15:16] mzanetti, there a bug #? [15:16] yes [15:16] lemme search [15:16] mzanetti, feels like one that should be escalated [15:17] Saviq: https://bugs.launchpad.net/telephony-service/+bug/1238933 [15:17] Ubuntu bug 1238933 in telephony-service "Incoming phone calls with hidden phone numbers are ignored" [High,Confirmed] [15:17] mzanetti: it's less that SF is faster, it's mostly that it is consistently smoother which in turn makes it feel a hell of a lot faster I think :) [15:18] davmor2: I would say that for the Nexus 4. But for the Galaxy Nexus Mir definitely is slower. much slower [15:18] davmor2, how do you define smoother vs. faster? [15:19] mzanetti, mhr3 small one https://code.launchpad.net/~saviq/unity-scopes-shell/platform-minimal/+merge/196557 [15:20] Saviq: this only c++ tests? [15:20] mzanetti, yes [15:20] ok then [15:20] mzanetti, there's no other ones there [15:21] Saviq: so things like scrolling is juddery and feels much slower on mir, on sf because its speed and smothness is consistent it therefore feels much faster, where as it may only be slightly faster to display if that makes any sense [15:23] davmor2, I think it does not make sense to me ;) - it's just that "speed" of a display server in that sense can only be measured in how many frames get to the display every second [15:24] davmor2, which equals smoothness [15:26] Saviq, why is that necessary? [15:27] mhr3, otherwise some of the tests try to access the display [15:27] it'd define smoothness vs faster talking about frame drops [15:27] mhr3, without you wanting it, just that the (default) xcb plugin excepts it to be there [15:27] hm, but it was passing on ci [15:27] mhr3, must be it granted access to display somehow - or maybe they're exporting that globally [15:27] I'd [15:27] mhr3, failed for me in sbuild [15:28] Saviq: so for me speed is how quickly things are displayed initially these seem to be on par, ie opening an app, clicking on something in the dash and it expanding and swipes of the indicators etc. The smoothness is down to perceived speed if it is smooth it seems faster, if it is juddery it feels slower [15:28] Saviq, very well [15:28] davmor2, on frame drops maybe [15:28] davmor2, yeah, but that's not related to Mir or SurfaceFlinger, as they're not doing the work [15:28] if everything is in sync, it feels smooth [15:29] davmor2, only thing they can do is say "bleh, I'm not ready for that frame, let's drop it" [15:29] when the framerate is not consistent, it's not smooth [15:29] davmor2, even if the application is ready with it [15:29] davmor2, and that results in judder etc. [15:29] davmor2, so it's not that Mir causes overall slowdown of the device, of course [15:29] davmor2, that would suck big time [15:29] davmor2, it's just that it can't juggle the frames as fast [15:30] davmor2, 'cause it's not optimised for the Galaxy [15:30] meh... someone keeps on calling me with a hidden number... [15:30] * mzanetti switches to his good old nokia phone [15:30] :) [15:30] tsdgeos, reproduced? [15:31] Cimi: sorry, still busy in the other one :-/ [15:43] Saviq, right now to switch between old and new scopes you have to change the import statement in about 5 qml files, is there a way to make it simpler where you'd need to change just one thing? [15:47] mhr3, right, I was wondering about that - easiest would be to have the new scopes in a different import path and keep the version number the same for now [15:47] mhr3, as then you'd just export QML2_IMPORT_PATH [15:47] mhr3_, right, I was wondering about that - easiest would be to have the new scopes in a different import path and keep the version number the same for now [15:47] mhr3_, as then you'd just export QML2_IMPORT_PATH [16:04] Hello there I have some questions about unity8 and was directed here. my questions are as follows. How do I add previews to a Carousel ? I see that I can use the model.icon/title ect but it seems that the gridviews that have previews also are in some sorta model. I am a little bit lost any docs on this ? [16:05] or should I just make a loader that has propertys for each of the model types ? [16:07] 2nd question is about filtering out data in the dash. There seems to be the c++ code in the backend and I have exposed most of this to QML but it is still not showing up. is there people working on this and if so where are blueprints. Thanks . === dandrader is now known as dandrader|lunch [16:17] sil2100, ping? managed to fix the license branch? [16:30] Saviq: any news on the indicator thingy? [16:31] didrocks, yeah, doesn't help it seems [16:33] Saviq: can you work on that (with a test of course once fixed?) it seems some people are upset because of it [16:33] Saviq: at least, to know where it comes from [16:34] didrocks, yes, I'll try and pinpoint where this happened [16:34] didrocks, where do we have the image vs. image diff again? [16:35] Saviq: seems it happened in image 27 [16:35] which had: [16:36] http://people.canonical.com/~ogra/touch-image-stats/20131120.2.changes [16:36] so, it can be Mir… sounds plausible to you? (for the matching?) [16:36] didrocks, huh :/ nothing related there [16:37] didrocks, no [16:37] http://people.canonical.com/~ogra/touch-image-stats/20131119.changes [16:37] that was image 24 [16:37] didrocks, and it's fine on 24? [16:37] but as 24/25/26 was respin quite fast [16:37] maybe peopel didn't notice [16:38] people* [16:38] didrocks, ok, I'll start from there [16:38] Saviq: do you have it in image 29? [16:38] didrocks, not sure which one I've flashed [16:38] yeah, I do [16:38] nevermind [16:38] Saviq: do you know what the plans are for the previews in carousel mode ? or adding filters to the dash scopes ? [16:38] Saviq: the toolkit seems to be the most obvious offender [16:39] didrocks, inded [16:39] pete-woods, could you review https://code.launchpad.net/~nick-dedekind/qmenumodel/lp1199423/+merge/196284 pls? [16:40] undefined1, we have branches lying around for those, but will wait for the new scopes API work [16:40] undefined1, so that we don't have to do it twice [16:42] Saviq: I see that is what I thought. I was playing around with the model.icon/dndUri ect to make fake previews for it. Saviq do you know where the blueprints are for the lens/scope work ? [16:43] I have been hacking on the plugins for my own use case [16:43] mainly Unity0.1 [16:43] undefined1, here's the new one https://blueprints.launchpad.net/ubuntu/+spec/client-1311-scopes [16:43] thanks [16:43] undefined1, but anyway, previews with carousel should work fine [16:43] undefined1, might need a long-press though [16:44] Saviq: onPressandHold: but how to call and from where ?> [16:44] do I need to add statement to basescope ? [16:44] I can not seem to figure out how they are getting called in the gridviews [16:45] ATM I am just loading a new page over the dashContent to make carousel previews [16:45] mhr3_: yes, it can get rereviewed/reapproved I guess? [16:45] undefined1, it's there already in GenericScopeView.qml:198 [16:45] Saviq: thanks that is what I was looking for your great ! [16:45] undefined1, maybe let's define what you mean by "carousel previews" [16:46] undefined1, if you mean getting a preview from a carousel, that should work already [16:46] carousel or coverflow view from the dashContent is what I mean. there is nothing in the delegate that lets me fire off previews. Only in the grid views [16:47] mhr3_: I've had a look at it, as commented, I'm concerned about the API break, but the rest of the changes look good to me [16:48] undefined1, there's definitely a branch that does that [16:48] undefined1, https://code.launchpad.net/~cimi/unity8/carousel-music-video/+merge/192118 [16:48]