[00:00] elopio: its these three [00:00] https://code.launchpad.net/~unity-team/unity8/dash-as-app/+merge/228534 [00:00] https://code.launchpad.net/~mzanetti/unity8/drop-running-apps-from-dash/+merge/227707 [00:00] https://code.launchpad.net/~canonical-platform-qa/unity8/dash-as-app-autopilot/+merge/229123 [00:16] kgunn: the first one is already merged, so I guess it shouldn't be part of the silo. [00:17] it shouldn't affect either. [00:17] I have merged trunk with the other two and it has no conflicts. [00:17] maybe the latest merge I did improved things? Can you try again?? [00:18] elopio: hmmm, musta been merged with saviqs [00:18] kgunn: sorry, it's the second one the already merged. [00:19] elopio: can you merge back trunk once more ? i bet you probably did it right before all the unity8 stuff migrated... [00:19] we caught it on the edge [00:20] kgunn: http://paste.ubuntu.com/7919046/ [00:21] also http://paste.ubuntu.com/7919049/ [00:21] this one I did it when I got your ping. It added some changes but was a clean merge. === shiznix_ is now known as shiznix === _salem is now known as salem_ === salem_ is now known as _salem [07:47] elopio, apps not launching make sense because of https://code.launchpad.net/~unity-team/unity-scopes-shell/drop-appid-workaround/+merge/228427 [07:47] elopio, we need a corresponding branch from the click scope [07:47] that will send us proper appid:///, applications:/// urls [07:55] holy merge batman [07:55] too many branches take a while to merge :D [07:59] :) [08:00] it's been 13 MPs... [08:11] mhr3, FYI: bcmwl builds again [08:11] yey, so i can reboot again :) [08:12] mhr3, yeah, upgrade first ;) [08:15] Saviq: who do we get to merge https://bugs.launchpad.net/ubuntu/+source/qtdeclarative-opensource-src/+bug/1349705 ? [08:15] Launchpad bug 1349705 in qtdeclarative-opensource-src (Ubuntu) "[PATCH] Flickable gets confused when there's other flickables on top" [Undecided,New] [08:15] otherwise merging in the overview is going to be a mess [08:17] Saviq: dandrader: are we fine with top approving https://code.launchpad.net/~unity-team/unity8/dash-as-app/+merge/228534 that has every single autopilot test not passing? [08:18] tsdgeos, leo has a branch for it that fixes the AP tests [08:18] ah cool [08:19] tsdgeos, there's the branch after it that fixes all those [08:19] tsdgeos, let me find out [08:19] (re: qt) [08:19] still 1 test broken [08:19] https://code.launchpad.net/~canonical-platform-qa/unity8/dash-as-app-autopilot/+merge/229123 [08:19] but much more acceptable [08:20] tsdgeos, yeah, Leo reported that's a bug, shell locks up when you try to swipe dash in when app is still loading [08:20] oki [08:20] tsdgeos, for bug #1349705 I'll just prepare a package and put it in there [08:20] bug 1349705 in qtdeclarative-opensource-src (Ubuntu) "[PATCH] Flickable gets confused when there's other flickables on top" [Undecided,New] https://launchpad.net/bugs/1349705 [08:21] same silo with overview, only problem is we'll need to smoke-test it [08:21] dandrader: if there's https://code.launchpad.net/~canonical-platform-qa/unity8/dash-as-app-autopilot/+merge/229123 i guess you can top approve https://code.launchpad.net/~unity-team/unity8/dash-as-app/+merge/228534 again and put https://code.launchpad.net/~canonical-platform-qa/unity8/dash-as-app-autopilot/+merge/229123 as requisite in the description? [08:22] Saviq: cool [08:22] and not cool at the same time :D [08:23] Saviq: ah now that you're here [08:23] shouldn't [08:23] if (!rendererLoader.expanded && !seeAllLabel.visible && target.collapsedItemCount > 0) { [08:23] be [08:23] if (!rendererLoader.expanded && seeAllLabel.visible && target.collapsedItemCount > 0) { === popey_ is now known as popey [08:23] in qml/Dash/GenericScopeView.qml ' [08:23] ? [08:26] tsdgeos, lookin' [08:26] tsdgeos, no, it's only meant to filter the results if see all isn't there [08:27] ah [08:27] ok [08:27] tsdgeos, because you can't see all the results in the grid (because of collapsed-rows: 2 for example) [08:27] makes sense [08:27] and you can't expand it because you have no see all (because you have a header link, which "invalidates" see all) [08:29] ok :) [08:29] Saviq: so what's our next target for landing stuff? dash_overview + friends? [08:29] do we have a silo [08:29] ? [08:30] tsdgeos, we're doing dash as app first [08:30] tsdgeos, then overview [08:31] right [08:31] makes sense [08:38] tsdgeos, do you have dash overview handy btw? [08:38] Saviq: on the desktop yes, phone not up to date [08:38] tsdgeos, desktop's fine [08:39] tsdgeos, I'll forward you images for the favorites, could you please do a few screenshots at different aspect ratios? [08:40] Saviq: what you mean different aspect ratios? [08:40] change the "screen" size? [08:42] tsdgeos, yeah, landscape, portrait etc. [08:43] tsdgeos, try out three or four sizes please [08:43] Saviq: favorites and all or just favorites? [08:45] tsdgeos, just favourites, in All we're not forcing aspect ratio yet (are we?) [08:49] Saviq: no we're not [08:49] Saviq: do you happen to know the paths of the files i have to overwrite? [08:50] because they are not unity8's [08:53] Saviq: i think i may need some extra packages [08:53] mhr3: ping [08:53] tsdgeos, pong [08:54] mhr3: which files do i need to overwrite to show my own dash overview images [08:54] for applications, music, video [08:54] tsdgeos, $ scopes-client musicaggregator "" | head [08:55] icon field [08:56] oki, tx [10:19] tsdgeos, alt nav should be fully working now [10:19] ...ish :) [10:23] mhr3: :D [10:23] ish? [10:23] might be a bug here and there :) [10:29] mhr3: tests don't pass [10:29] can't dpkg pacakge it [10:30] tsdgeos, did you pull? [10:30] ah you just changed it [10:31] meh [10:31] still failing [10:33] tsdgeos, I was thinking... when you look at a dash page, meaning you focused the dash and you are on a certain page... maybe we should just create all the delegates on that page, maybe delaying just the image fetching until they're on screen (to potentially save data) [10:34] tsdgeos, creating/destroying delegates (and whole categories, for that matter) will always kick us in the a$$ [10:34] tsdgeos, can you pastebin LastTest.log? [10:35] Saviq: yeah :/ [10:35] tsdgeos, and well, if you're looking at a dash page, we should not refrain from committing memory to it [10:36] mhr3: http://paste.ubuntu.com/7922784/ [10:36] tsdgeos, as we unfocus, we should destroy all offscreen delegates (at least those vertically off-screen, not sure if we should destroy all that are not visible in the dash right now), but maybe not the categories [10:37] tsdgeos, let's have a think/sleep & chat about it next week [10:37] :( [10:37] :) [10:37] i mean [10:37] lol [10:37] tsdgeos, btw, does that make any sense to you http://pastebin.ubuntu.com/7922748/ [10:37] tsdgeos, weird, it's failing in location test... i didn't even touch that :P [10:37] tsdgeos, that's what I got out of a local build with your patch [10:37] mhr3: i need to more somewhere else :D [10:38] Saviq: beats me [10:38] tsdgeos, do you have python-tornado installed? [10:38] i do [10:38] pete-woods, why would that fail ^? [10:39] tsdgeos, or you know... DEB_BUILD_OPTIONS="nocheck" bzr bd [10:40] taht too [10:42] mhr3: well the location tests only require tornado, and don't do anything clever outside [10:43] tsdgeos, mhr3 is right, it built with 4.9 [10:43] (at least probably) [10:43] we might need to force it to 4.8 for now and do a whole-set transition to 4.9 separately [10:44] Saviq: what built with 4.9? [10:44] tsdgeos, qtdeclarative, the symbol changes [10:44] ahhhh [10:44] ok, thought you were talking about unity-scopes-shell and the test issues :D [10:44] sry [10:44] hi pete-woods, so thostr_ told me you're the man for adding a snippet for basic networking to the Qt Creator scopes template :) [10:45] dpm: yep, I'm happy to take on that task when I have a minute [10:45] I'd basically be gutting the one of the video scopes [10:45] I'll probably keep a basic test case, too === vila is now known as vila-lunch [10:46] pete-woods, awesome, yes, that sounds like a plan. So if that helps, this is essentially the code we've got now http://bazaar.launchpad.net/~dpm/+junk/scopetemplate/files - you might want to use it as a basis or as you say, start from one of the video scopes [10:47] pete-woods, in addition to networking, it'd be good to have some basic code to process what the remote service returns (e.g. perhaps json, if that's the most common format?) [10:47] dpm: yep, will do it with json [10:47] sounds good [10:47] dpm: I'll put a basic python based http server in the test case to test it all [10:48] oh, that'd be a huge improvement from what we've got now, yes [10:48] dpm: well I have that already in the video scope tests, so it should be very little effort to copy it over [10:50] great [10:51] Hola [10:51] pete-woods, do you have a rough idea of when you could have this template ready? [10:52] dpm: it's a matter of priorities, if thostr_ says to do it right away, then pretty quick really [10:52] Saviq, so dash-as-an app is not getting merged any time soon because of that AP test failure, right (which pointed to a regression) [10:52] I'm currently managing a release of unity-scopes-api with ABI breaks, though, so after that [10:54] dandrader, "any time soon" meaning until we fix it (we need to) [10:55] dandrader, there's two issues I know of: stuff getting stuck when you switch between dash/app, and failure to launch apps from dash (I believe a click scope issue) [10:56] Saviq, I mean that this will likely come before dash-as-app: https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1350917 [10:56] Launchpad bug 1350917 in Unity 8 "UI freezes when switching to an app which is opening." [Critical,In progress] [10:57] Saviq, or should, as it's a critical bug... [10:57] dandrader, it can come *with* it, not necessarily before [10:57] Saviq, ah, the silo thing [10:57] dandrader, yup ;) [11:04] thanks pete-woods [11:05] dpm: about pete's priorities [11:06] dpm: so, first he needs to finish the unity-scopes release and afterwards he can spend some time on it... would it be ok for you if he'd just "throw" you some code over? [11:09] thostr_, I think so, I'm talking to Zoltan about when it'd be good for this to land on a QtC release. My hunch is that it'd be good to have the template early next week if possible, but let me see what he has to say [11:18] thostr_, pete-woods, so either after the unity-scopes release is finished (if that's today) or early next week would work well to include the template in QtC. Would that work for you? [11:18] dpm: next week :) [11:20] ok, let's follow up next week, then. Thanks! === MacSlow is now known as MacSlow|lunch === vila-lunch is now known as vila [11:27] mhr3: ping [11:27] tsdgeos, poing [11:28] mhr3: can we kill visible in scopves-v4 ? [11:28] or is anyone using it? [11:29] Saviq: ↑↑ [11:29] tsdgeos, visible what? [11:29] mhr3: Scope::visible [11:29] tsdgeos, oh yea, let's drop it [11:29] seems like a bastard favorite [11:29] to me [11:33] +1 [11:48] tsdgeos, and that means you can drop the filter proxy model in shell right? [11:48] yey! [11:48] mhr3: hmmmm [11:49] we may need to add some stuff [11:49] mhr3: we are using the filter proxy model to ask for the index of a scope in the model [11:50] mhr3: do you think we can add to ScopesInterface ? or at least a count [11:50] so i can just iterate over getScope(int row) [11:50] it doesn't have that? [11:51] count not [11:51] getScope(row) yes [11:51] but without count i don't know where to stop [11:51] well i guess i could stop on null [11:52] adding it, we have it on all models [11:52] cool [11:52] lunchy lunchy [11:56] same [12:06] larsu, are you able to build gsettings-qt for armhf? I get dep issues I am unable to solve (first mesa, then gio2.0 dev) [12:08] jgdx, is that crossbuild? [12:11] seb128, yes [12:15] jgdx, likely an archive issue rather a gsettings-qt one then... [12:18] seb128, yeah, and jenkins build it seemingly fine for armhf. [12:19] jgdx, well, jenkins doesn't crossbuild [12:19] seb128, it is all clear now [12:19] weird === alan_g is now known as alan_g|lunch [12:20] seb128, no, "it's all clear to me now [why it passes on jenkins but fails for me]" [12:20] oh, ok [12:20] seb128, this is new to me, how is archive issues like these resolved? [12:21] s/is/are [12:21] by somebody looking what the issue is and fixing it [12:21] likely buggy multiarch support somewhere in the depends of what you try to install [12:21] xnox might be able to help you/give more hints on how to debug [12:21] right, thanks [12:24] hm?! =) [12:24] * xnox reads back [12:24] xnox, gsettings-qt doesn't seem to be crossbuildable and jgdx is not sure how/where to report the issue [12:25] seb128: by default, one should assume that cross-building has not been enabled. Most packages need tweaks to get it working. [12:26] xnox, well, in that case jgdx says he got gio not installable, that should work [12:27] seb128: my laptop seems to lack cross-sbuild chroot, please wait =) [12:27] k [12:27] thanks for looking ;-) [12:39] seb128: jgdx: all build-dependencies are coorectly installable for cross-compilation. However gsettings-qt appears to be using qmake, which is not supported build-system for cross-compilation. [12:39] oh, ok [12:39] seb128: jgdx: please use CMake, or some other sensible build-system like autotools. [12:40] we have better things to do for rtm than changing build systems [12:40] so I guess that's just not cross buildable [12:40] it's not like we were changing it every week anyway, so shouldn't be a real issue [12:41] jgdx: if you want to build it locally, $ mk-sbuild --arch armhf utopic; sbuild --arch armhf *.dsc should work to do "native" compilation locally, powered by qemu-user-static. === alan_g|lunch is now known as alan_g [12:48] xnox, seb128, k, thank you [12:49] jgdx, yw! === _salem is now known as salem_ [13:05] mhr3: you removed visible but did not add count [13:06] tsdgeos, just got back from lunch [13:06] same here :) [13:06] mhr3: ping me when you're done [13:09] tsdgeos, pushed [13:09] cool tx [13:21] tsdgeos, ah, test reports should be good now... it was a lettercase issue [13:21] Saviq: ?¿ [13:21] tsdgeos, Test vs test [13:22] lol [13:22] really? [13:22] yeah [13:22] mterry_: answered your question [13:23] Saviq: in our side or on CI side? [13:23] tsdgeos, well, depends ;) [13:23] tsdgeos, CI expected test, we had Test [13:23] i mean there's no standard, is there? [13:23] tsdgeos, but got fixed CI side [13:23] nope [13:23] did they tell us? [13:23] tsdgeos, cool updated [13:24] tsdgeos, it was just miscomm, fginther added Test now, too [13:24] cool :) [13:24] tsdgeos, *but* it still didn't end up in http://s-jenkins.ubuntu-ci:8080/job/unity-phablet-qmluitests-utopic/730/testReport/%28root%29/qmltestrunner/ yet [13:24] tsdgeos, so we might need to follow up [13:26] did we find another bug in qtdeclarative ? (dash as app silo) [13:26] kgunn, found it earlier than that [13:26] kgunn, and related to overview rather than dash as app [13:27] kgunn, but we thought we can just put it in anyway, we're working on a fix for dash as app anyway [13:27] or dandrader is working on a fix [13:27] sure [13:27] * Saviq is about to fly out [13:27] kgunn: it was there for a while, caused using the "old combo in header" not work sometimes, never was able to pin point it [13:27] it is this one [13:27] * kgunn hopes Saviq's arms are well rested for the flappign [13:32] o/ [13:32] \o/ —o— \o/ —o— \o/ [13:34] lol [13:35] /°\ [13:35] weird guy [13:35] mhr3: for mocking purposes can i make navigationid() and query() be the same? [13:36] tsdgeos, do whatever you want in mocks [13:36] well i need to make them do similar something to what you do :D [13:37] in my head query and navigationid are just the same since i don't do weird jumps, so performquery will just use navigationid to internally change the current navigationid and that's it [13:38] tsdgeos: hey, any progress on table widget for previews? [13:39] cwayne: it's done [13:39] tsdgeos, oh that reminds me, i'm not doing anything with currentAltNavId [13:39] cwayne: needs to be reviewed and merged :D [13:39] mhr3: meaning? [13:39] tsdgeos, that i got a bug :) [13:39] tsdgeos: :D nice, thanks! [13:40] cwayne: i can give you a branch if you want to try it [13:40] tsdgeos, but we'll need to think things through, to deal with the bug we've been talking about for the past few days [13:40] tsdgeos: that'd be great [13:41] cwayne: https://code.launchpad.net/~aacid/unity8/tablePreviewWidget/+merge/229006 [13:44] tsdgeos, for one, you should consider that performquery for a particular navId will change currentNavId, but asynchronously [13:45] tsdgeos, and maybe it won't if something goes wrong (query fails or something) [13:45] tsdgeos: awesome, thanks man [13:46] mhr3: i see [13:48] tsdgeos, btw if you want to test it against a scope, you can build my sort-order branch and point scope-tool to ./builddir/tests/data/mock-scope-double-nav/*.ini [14:01] mhr3: so what the current department code does is [14:02] on pressed, get the departmentid the role returns and assume that is the next department id we will go [14:02] is that wrong? [14:02] tsdgeos, sounds good to me [14:03] mhr3: so what's performquery doing then? [14:03] it goes there [14:03] ok [14:03] asks the scope to get data for that dep_id [14:03] and that can fail [14:04] but i'll be showiing that department already [14:04] tsdgeos, but i think the bug was that you assumed that when a scope page is shown the first time, you're in root department [14:04] and that's not always the case [14:05] it's the only way it can work [14:05] i can't go more left than first [14:05] kind of tricky :D [14:05] tsdgeos, unfortunately you might need to [14:05] unfortuantely i might need to rewrite the whole thing then [14:06] tsdgeos, consider that you click on a scope link that takes you to amazon's "Music DVDs" right away, there's quite a lot of stuff to the left [14:06] sure [14:06] there were no scope links [14:06] and noone told me i should have to go left [14:06] so it's not developed for that [14:07] tsdgeos, you know things change and get extended all the time [14:07] i'm just saying "that small bugfix" may mean re-coding everything [14:07] ouch [14:07] it's not your fault nor for you to care [14:07] but it's what it is [14:14] just complaining to the air [14:15] tsdgeos, in ~aacid/unity8/implementOverlayColor, how does one test the branch? [14:16] mterry: make tryCard [14:16] tsdgeos: was that supposed to rhyme? Cuz it sounded good :) [14:16] tsdgeos, I tried altering the default #99000000 value, but it didn't give any visible chang [14:16] and do some stuff [14:16] mterry: let me see [14:16] tsdgeos: even has good rhythm [14:16] greyback: :D [14:16] no it was not [14:16] will try tryCard [14:16] mterry: ah you need to be in overlay [14:17] What was the reason again for unity8 not starting on the desktop and just idling around after run.sh? [14:17] mterry: are you there? [14:17] tsdgeos, I don't know unity8 terminology outside of the greeter ;) [14:17] tsdgeos, where is the overlay? [14:17] mterry: use the combo, select overlaid [14:18] then go to the lower editable part [14:18] tsdgeos, are you talking about tryCard? [14:18] mterry: yes [14:18] mterry: there's a combo topright, no? [14:18] tsdgeos, oh I was talking about looking around unity8 while it's running [14:18] mterry: ah, then you look at the carousel [14:18] but tryCard is just easier [14:18] tsdgeos, I thought we dropped carousels [14:18] you can change stuff live [14:19] select art, header, summary - overlaid [14:19] and then in the lower editable box [14:19] put "overlayColor":"#99990000", [14:19] after title for example [14:19] and change it to [14:19] "overlayColor":"#99009900", [14:19] and see how it changes [14:20] k [14:22] tsdgeos, brilliant [14:23] works by chance [14:23] i just played with the shader code until it did what i wnted [14:23] :D [14:30] mhr3: ping [14:31] nvm, I can look it up myself [14:34] actually I'll ping again, do you know of any tutorials for using net-cpp? [14:44] mterry: if you're doing small reviews, you could do https://code.launchpad.net/~aacid/unity8/pageHeaderBackgroundOverflow/+merge/228881 [14:45] mterry: very easy to test, run make tryDash, go to the third scope (the one with the ubuntu logo) and click on the search icon [14:45] see difference with and without patch [14:47] Cimi: and maybe you can do https://code.launchpad.net/~aacid/unity8/collapsingPreviews/+merge/228857 and https://code.launchpad.net/~aacid/unity8/tablePreviewWidget/+merge/229006 since they have to do with previews? [14:47] not doing reviews until I got mine approved :O [14:47] :P [14:48] it's give and take! [15:02] Cimi: any eta on settings UI support landing? [15:02] tsdgeos, will review that one [15:02] cwayne, I will start monday [15:02] cwayne, might take a week === gatox is now known as gatox_lunch [15:42] tsdgeos: ping [15:42] dednick: yep? [15:42] tsdgeos: hey. did you do the work on scopeStyle? [15:42] dednick: it's mostly saviq's [15:42] but i know some [15:42] why? [15:43] tsdgeos: hm. ok. don't think it belongs in PageHeader since it's in Components, not Dash. Causing me pain since i want to use it in indicators. [15:43] all i need to do it define QtObject with known params right? [15:44] i remmeber arguing with saviq about that yeah [15:44] yeah a QtObject would do [15:44] now [15:44] you're using a PageHeader in indicators? [15:45] tsdgeos: yes. getting rid of tabs [15:45] dednick: why you need PageHEader? [15:45] and not the SDK's one? [15:46] i mean PageHeader gives you the search thing [15:46] do you need search? [15:46] tsdgeos: no. [15:46] then you should probably use SDK one [15:46] tsdgeos: doesnt the sdk header require mainview crap? [15:46] integrated with the pages [15:46] well you can use the "private" stuff like our PageHeader does [15:47] PageHeadStyle [15:47] see inside pageheader [15:47] PageHeadStyle is "most" of the stuff [15:47] and then there's all the saerch and etc [15:47] but if you don't need that [15:47] PageHeadStyle should be good for you [15:47] tsdgeos: hm. does it look the same as the dash headeR? [15:47] dednick: well it's what the PageHeader is using ;) [15:47] dednick, yeah, the PageHeader component is actually dash-specific... [15:48] dednick, it should actually be moved to Dash/ [15:48] and you just use the SDK one [15:50] tsdgeos: Saviq: ok, thanks. i'll take a look [15:53] Saviq: is there a size limit on the icon for an attribute? I'm trying to use an image I get from yelp API showing the ratings, and it's not showing up on the device, but working in scope-tool [15:55] cwayne, nope, as you can see in https://code.launchpad.net/~paulliu/unity8/attribute/+merge/223242 around line 49 [15:55] cwayne, we just force height to 2gu and use natural width [15:56] Saviq: any idea why mine wouldn't show up on device, but does show up in scope-tool? [15:56] cwayne, paulliu might be able to help, he wrote that [15:56] cwayne, but nothing comes to mind, any interesting unity8.log output? [15:56] or am i abusing this, and we're expecting an actual icon name instead of a .png... [15:57] krieky. we should really get a basic header from sdk... [15:57]