[01:16] Hey? There's a thing. about Unity. That is mildly annoying. Is this the proper place to kvetch? [01:38] :( === jhodapp|afk is now known as jhodapp === pierre1_ is now known as f03lipe === duflu_ is now known as duflu [08:54] what [08:54] why does unity-scope-click depend on unity-scope-mediascanner2 ?¿ [09:03] pstolowski: there? [09:03] TheMuso, hi [09:04] pstolowski: about https://bugs.launchpad.net/unity8/+bug/1281709 [09:04] Launchpad bug 1281709 in Unity 8 "Preview with gallery widget and multiple images: image is cut off" [Undecided,New] [09:04] tsdgeos, hi [09:04] pstolowski: is it the thing that happens with let's say ebay? [09:04] TheMuso, sorry, mistake [09:04] or which other click app do i click to see it? [09:05] tsdgeos, I saw it yesterday when implementing proper gallery in click scope [09:05] tsdgeos, I'm now flashing my phone, so I can verify with others [09:05] i think i have a fix [09:05] but need to make sure i fixed what you're complaining about :D [09:06] pstolowski: that branch you mention is said to be merged, so i should have it if i'm using new-scopes? or not? [09:06] tsdgeos, I'll now in ~20 minutes when my phone is ready again [09:06] ok [09:07] pstolowski: are you used to compiling the shell? [09:07] tsdgeos, cause yesterday I was running a private build [09:07] tsdgeos, yeah [09:07] pstolowski: try https://code.launchpad.net/~aacid/unity8/new-scopes-image-gallery-scroll-workaround then [09:07] tsdgeos, I just build a deb on the phone, so if that works, that's ok [09:08] tsdgeos, you can test with "SkimboU" preview [09:08] whatthe [09:08] unity8 crashed when searching ^_^ [09:08] tsdgeos, there's like 8 screenshots, but you can only access the first few, as swiping to the sides changes the preview [09:09] tsdgeos, might be the thumbnailer still [09:09] didn't i just fix that the other day? [09:09] tsdgeos, dunno, did you? :D [09:09] i did [09:09] but maybe not all instances _/ [09:09] tsdgeos, well, there's only really one instance, isn't there... [09:10] Saviq: i mean all causes of the "search crashes" [09:10] Saviq: i don't have a skimbou app [09:10] tsdgeos, in app store [09:10] searching for it gives me a blank thing [09:10] :/ [09:11] tsdgeos, that on desktop or phone? [09:11] desktop [09:11] Saviq: btw i'd say we need some more spacing on the search results [09:11] title seems to be too close to the mascot [09:11] tsdgeos, I think that got fixed already [09:11] i'm running latest new-scopes [09:12] it's still there [09:12] tsdgeos, it got fixed in the scope itself [09:12] ah [09:12] ok [09:12] tsdgeos, i.e. what you're seeing is art in horizontal layout [09:12] tsdgeos, and it got fixed to use mascot instead [09:12] i see [09:12] tsdgeos, btw bug #1281782 [09:12] bug 1281782 in Unity 8 "Getting stuck in overlay scope page and "back" too small a target" [Critical,Triaged] https://launchpad.net/bugs/1281782 [09:12] anyway i'm pretty sure https://code.launchpad.net/~aacid/unity8/new-scopes-image-gallery-scroll-workaround is the fix [09:13] since the ebay app preview is a similar case [09:13] even if it's just one big image [09:13] Saviq: yes, i've seen it [09:14] tsdgeos, cool [09:14] tsdgeos, ugh, so we managed to confuse all the Flickables, did we :/ [09:15] yeah [09:15] we already had this in the "old" preview [09:15] i was hoping we wouldn't need it [09:15] but i was obviously wrong D: [09:15] ;) [09:16] tsdgeos, grr, looks like flashing failed and I have to start over [09:16] pstolowski, no worries, I'll handle it [09:21] Saviq, thanks. I'll try the fix anyway when my phone decides to work [09:40] tsdgeos, pstolowski, fixes, merged! [09:41] tsdgeos, you're looking at the "got stuck in overlay scope"? [09:41] Saviq: yeah [09:41] tsdgeos, ok, marking in progress [09:41] Saviq: the PageHeader is a bit of a weird thing [09:41] tsdgeos, I know :/ [09:42] no idea why we're using flickable in there [09:42] Saviq, cool [09:45] tsdgeos, why did we loose the activity indicator when preview is loading? [09:45] tsdgeos, IIRC so that we move the search entry around depending on available width [09:45] mhr3, "we don't need an activit indicator since we get stuff immediately" [09:45] that [09:46] cause you get empty preview model immediately? [09:46] mhr3, when do you think we should hide the activity indicator again? my initial try was onCount > 0 [09:46] Saviq: doesn't seem the reason i am seeing from the code, but may be [09:46] Saviq, either that or just hook to the loaded property on the preview model [09:47] mhr3, there wasn't one when I implemented it ;) [09:47] Saviq, things move while you're on the hills ;) [09:47] ;) [09:48] tsdgeos, mhr3, I'll bring the activity indicator back [09:48] ty [09:56] mhr3, I never get loaded true :/ [09:57] what.. how? why? [09:58] Saviq: so what do we do with those bugs tagged as mwc that we fix in new-scopes? mark as fix released? [09:58] tsdgeos, fix committed [09:58] ok [09:59] Saviq, wait, no way, i have tests that are waiting for it and they're passing [09:59] QTRY_COMPARE(preview_model->loaded(), true); [09:59] mhr3, http://paste.ubuntu.com/6959127/ [10:00] mhr3, I was only getting "scopes_ng::PreviewModel(0x2945490) false" [10:00] mhr3, nothing onLoadedChanged [10:00] mhr3, feel free to commit/push to new scopes once you confirm it's working for you [10:00] Saviq, hm, is there some convention that i'm missing? should the loadedChanged signal have the boolean as a param or something? [10:00] Saviq, quick mumble when you have time? [10:01] mhr3, nope, not needed [10:01] mhr3, loadedChanged was never emitted for me [10:01] Saviq, which scope did you try? [10:01] Cimi, am there [10:01] mhr3, apps [10:02] * mhr3 wouldn't be surprised if they got it wrong [10:03] mhr3, apparently they did [10:03] mhr3, other scopes are fine [10:03] heh [10:08] Saviq: https://code.launchpad.net/~aacid/unity8/new-scopes-improve-back-button-opened-scopes/+merge/207122 [10:08] tsdgeos, thanks [10:09] * tsdgeos gets into https://bugs.launchpad.net/unity8/+bug/1281707 [10:09] Launchpad bug 1281707 in Unity 8 "Sometime open apps are displayed in the dashboard under the title of the scope" [High,Triaged] [10:09] this one will be harder probably [10:10] mhr3: I see you guys managed to already fix the font colors you reported yesterday. [10:10] mhr3: or at least the "Running apps" title font color. category font color seems to not have changed. [10:12] karni, hmm? category header has your fix in? [10:13] karni, or do you mean the top header? [10:13] mhr3, I'll wait for apps scope to be fixed before merging the activity indicator, k? [10:13] Saviq, sure, jussi is looking into it [10:13] Saviq: mhr3 reported that category header had "lightgrey" instead of "grey" (if I understood that correctly), and that the running apps title font color was wrong. the latter has been fixed for sure, I can see that now. [10:14] yep ^ [10:14] karni, yeah, I did that yesterday, but category header has "grey", but it's Light, maybe that's what you're seeing mhr3? [10:14] ok, I assume that's been done in that case :) [10:14] I think category header looks fine, unless mhr3 compared that against visuals, in which case we could fix it [10:14] or actually [10:15] it's grey at 60^ [10:15] 60% [10:15] Saviq: thanks for fixing it yesterday [10:15] ah [10:15] Saviq, hm, it just didn't look like in the designs [10:15] karni, do you have time to tweak this? [10:15] Saviq: sure. although I was having hard time finding it. any general pointers where to look? [10:15] qml/Dash/* ? [10:15] karni, Header.qml in Components [10:16] perfect [10:17] tsdgeos, 20 + visible: false on a MouseArea? [10:17] ah [10:17] child Image [10:17] yep [10:17] it's reversed so the button is bigger than the image [10:23] Saviq: so you can reproduce https://bugs.launchpad.net/unity8/+bug/1281707 reliably? [10:23] Launchpad bug 1281707 in Unity 8 "Sometime open apps are displayed in the dashboard under the title of the scope" [High,Triaged] [10:31] tsdgeos, not reliably [10:31] tsdgeos, but was able to a few times [10:31] Saviq, why cardcarousel has 1.38 of scaling? [10:31] Saviq: ok, will see once flashing ends [10:32] Cimi, https://docs.google.com/a/canonical.com/document/d/1n880Fih5KyGPcoP5chidnHDG_8TxXUgSuij7f4rHpuk/edit?pli=1#heading=h.b5760jrq75m3 [10:32] Cimi, "On a device that is 40 GU wide, the currently focused card in a carousel has a width of 18 GU. The width of the other cards is as now on the dash." [10:33] tsdgeos, got a reliable way to repro [10:33] tsdgeos, close all apps, scroll to the third row of apps, launch one of them [10:34] Saviq, but then it changes again on different screen size [10:34] Cimi, "On a device that is 128 GU wide (for clarity: Nexus 10 width), the currently focused card in a carousel has a width of 26 GU." [10:34] Cimi, it's a linear scale [10:34] ok [10:34] "For form factors between these widths the currently focused card size should scale. " [10:34] Saviq, but do we want the currently focused card to be that big [10:35] Saviq, or simply having less elements? [10:35] tsdgeos, make that semi-reliable [10:35] Saviq, if we reduce the number of elements per screen, they become bigger [10:36] Cimi, more elements on a wider screen, I'd say [10:36] Cimi, use the screen real estate, don't just scale everything up [10:36] ok [10:39] Saviq: FTR PageHeaderLabel is not that mhr3 meant. I'm looking for the scope category title font color instead :) [10:40] karni, Header.qml, not PageHeader [10:40] karni, Components/ListItems/Header.qml [10:40] d'oh, was looking for exactly that (based on what I saw in GenericScopeView) [10:41] Saviq, so you believe this will be the new standard for all carousels? [10:41] Cimi, there is only one carousel [10:41] Saviq, I noticed we had 1.2 for squared carousel (music) [10:41] Cimi, had [10:41] Saviq, and 1.14 default for something like movies [10:42] Cimi, had [10:42] Saviq, now we will have only squared cards? [10:42] Cimi, no, we will have portrait cards, but they will have the same scaling ratio [10:42] Saviq, I am thinking if I could drop properties [10:42] Cimi, not now please [10:42] Saviq, same scaling on portrait will be very tall [10:42] Cimi, comment on the spec [10:43] Saviq, like, 26x26 is fine, 26x39 maybe too tall [10:43] Saviq, I'll comment [10:45] mhr3, are we supposed to have all the scopes favorited or is that just missing the env var? [10:45] mhr3: Saviq: https://code.launchpad.net/~unity-team/unity8/new-scopes-font-fixes/+merge/207130 [10:45] mhr3, hey, is https://bugs.launchpad.net/ubuntu/+source/libunity/+bug/1274669 a bug for you? [10:45] Launchpad bug 1274669 in libunity (Ubuntu) "scope-runner-dbus.py crashed with signal 5 in g_variant_new_va()" [Medium,Confirmed] [10:46] karni, feel free to commit stuff like that to new-scopes directly [10:46] Saviq: ack, will do [10:46] karni, no time for the MP overhead [10:46] roger that [10:51] Saviq, the envvar is set by the upstart job [10:51] Saviq, scope tool displays everything by design [10:52] mhr3, yeah, I get all of them on both phone and tablet, maybe something wrong in my setup [10:53] Saviq, you using run_on_device? [10:53] maybe that's why [10:53] mhr3, did not [10:53] the envvar isn't set globally [10:53] it's just for the job [11:09] tsdgeos, I can confirm your gallery branch fixes the issue. thanks [11:09] good stuff [11:19] pete-woods: hud-service back on the crazy! [11:19] tsdgeos: pushed out an update about an hour ago, are you running this version? (hopes not) [11:20] pete-woods: do you know the version number? [11:20] should have today's date in it [11:20] 13.10.1+14.04.20140212-0ubuntu1 [11:20] 20140218 [11:20] that's not today i guess [11:20] :D [11:20] ok, let me dist-upgrade and kill it [11:21] tsdgeos: I took out some crazy code from libdbusmenu-qt that was running processEvents in a busy loop [11:21] and replaced it with a simple QEventLoop [11:21] pete-woods: shall i reboot to make sure everything is clean or killing hud-service should be enough once i upgrade? [11:21] you should just be able to restart hud [11:21] it doesn't have state or anything like that [11:21] the maintainer scripts should really restart HUD for you [11:22] ok [11:22] (they don't currently I think( [11:23] is it possible that the ddebs thing that creates dbgsym packages is "behind"? [11:23] by installign the new hud i got hud-dbgsym libdbusmenu-qt5-dbgsym to be uninstalled [11:23] hmm [11:23] could well be [11:23] I don't really understand that system [11:23] me neither [11:24] the new debs were published very recently, so it sounds plausible [11:24] but i guess it is not that weird to think it'd be a bit behind [11:25] mhr3: any idea where this comes from? [11:25] Caught an error from create_query(): unity::scopes::TimeoutException: Request timed out after 300 milliseconds [11:26] tsdgeos, scope not responding [11:26] or responsing slowly [11:28] sil2100, hi! any idea why https://code.launchpad.net/~dandrader/ubuntu-keyboard/outOfSourceBuilds/+merge/204292 didn't get merged by CI yet? [11:29] sil2100, and whom should I poke about it? :) [11:29] as it feels like ubuntu-keyboard is missing an "owner" at the moment... [11:30] sil2100, maybe re-approving would do the trick? [11:53] dandrader: hello! [11:53] dandrader: there is no automerger anymore, that's normal [11:54] dandrader: I'm helping out with maintaining the u-k right now, but anyway the OSK is in CITrain now and the whole idea is that what's in trunk is what's in the archive ;) [11:54] sil2100, because I see that revision 135 has "Committer: CI bot " [11:54] dandrader: ask bfiller to fill you in with the details once he's back [11:54] dandrader: yes, the merger merges in all the targetted merges once we do a release [11:55] yeah, lander should land this; subscribe your lander (aka bill) to MPs that are ready for merging to trunk [11:55] dandrader: so if you want to have a merge (branch) landed, poke your manager (or the lander for the component) to prepare a landing for you ;) [11:56] sil2100, ok. will pester bfiller then. thanks! [12:10] didrocks, when can we have this in main? https://code.launchpad.net/~3v1n0/unity/fix-decorations-crash-1281610/+merge/206954 [12:10] Cimi: you need bregma to land the branch [12:10] didrocks, I see [12:10] Cimi: it's in transit from what I see [12:10] didrocks, qtcreator crashes frequently and I lose the code :( [12:11] hum, already landed even [12:11] Cimi, that branch was supposed to have landed yesterday [12:11] it seems it did [12:11] the other MPs in that landing merged, but that one didn;t, hmmm.... [12:12] bregma: are you sure it wasn't added after the configuration was done? [12:12] didrocks, it was added later, yes, it could have been something I did wrong [12:12] bregma: so yeah, nobody asked to reconfigure to add it [12:13] bregma: just add to next landing I guess [12:13] (once the current transit one lands) [12:15] we shouldn't have one in transit unless someone went and built line 23 without consulting me first, which appears to have happened [12:15] * bregma changes plans again [12:16] bregma, I triggered a build for l23 since it had an assigned silo and I wanted to test those changes [12:17] bregma, if you have a silo you can as well use the ppa, even if you don't end up doing the actual upload [12:18] bregma, if you don't want a request to move forward, don't set the "ready?" column to "yes"... [12:21] seb128, it's OK, we're just following our original plan instead of the last-minute change we came up with a few hours ago before some appropriate sleep [12:21] no harm done [12:21] k [12:22] sorry if I created issues, I though you had the slot assigned after your eod [12:22] so I though triggering a build would be useful [12:22] next time please let a comment or change the ready to "no" if you want to put on hold [12:23] we 've given up on the EOD concept, it wasn't working for us [12:23] bregma, btw the hidpi stuff seems buggy for me, cf #ubuntu-desktop, it scales the UI on my lowdpi monitor [12:23] lol [12:24] did you give up on the sleep concept as well? ;-) [12:24] I suspect Trevinho has [12:25] seb128, I think we need one more change to hook up the proper scaling values from u-s-c and Unity before it will work correctly on multiple displays [12:25] shouldn't the default be correct though? [12:26] e.g no scaling by default if the other bits are missing/not sending a value [12:26] yes, there's supposed to be no visible change by default [12:27] that part is buggy then [12:27] I'm happy to provide debug infos [12:28] Saviq: you working on bug 1281657 ? The spreadsheet says so [12:28] bug 1281657 in Unity 8 "Keyboard is hard to use on Manta, and occasionally hard to use on flo" [Undecided,New] https://launchpad.net/bugs/1281657 [12:33] greyback, no [12:33] greyback, I think it's more ownership than assignment [12:33] Saviq: ok good [12:34] Saviq, I'm moving some functions inside cardTool, getCategoryLayout (so layout will be defined there) for example [12:34] Saviq, is it ok? [12:35] Cimi, hmm CardTool should only really deal with Card, not category? [12:36] Saviq, problem is [12:36] Saviq, cardWidth depends on category [12:36] but category changes [12:36] depending on results.count [12:37] at the same time [12:37] category depends on cardwidth :) [12:38] Cimi, sure, but card itself doesn't depend on those [12:38] greyback, there's no applicationManager function to make a given application surface disappear (i.e. not be shown), right? [12:38] Saviq, I'll add a function getCardWidth(category) to cardTool [12:39] and rechange categoryLayout from outside [12:39] dandrader: I think MirSurface has a "visible" property, but Application does not. You thinking what I'm thinking? Hide/Show surfaces to fix some animations? [12:39] greyback, to achieve this effect you just make unity8 display something (as unity8 surface is above all app surfaces) on the app surface position [12:39] it's more code though, and looks less nice imho [12:39] greyback, yes [12:39] dandrader: I've a patch for unity-mir that'll help you [12:40] greyback, show me! === pete-woods is now known as pete-woods-lunch [12:41] dandrader: http://pastebin.ubuntu.com/6959723/ [12:42] dandrader: get anywhere with yesterday's problem? The mapRectToScene thing [12:42] greyback, didn't even start. was working on https://code.launchpad.net/~dandrader/unity8/sideStage_lp1281088/+merge/206946/comments/485476 [12:42] dandrader: ok [12:43] Cimi, CardTool *does* give you cardWidth [12:43] Cimi, that's it's whole purpose [12:43] -' [12:43] Saviq, but can give me the not correct one [12:44] Saviq, let's say you have a carousel with 2 items [12:44] Saviq, cardWidth is for a carousel, but it's not [12:44] greyback, ok so I'll just add it to https://code.launchpad.net/~dandrader/unity8/sideStage_lp1281088/+merge/20694 branch [12:44] because we know we would have fall back to grid [12:45] Cimi, should be enough to change category-layout in there, no need for a function I'd say [12:45] greyback, oh, that's to unity-mir, stupid me [12:45] greyback, so, are you prepping an MP with that? [12:45] Cimi, change category-layout to "grid" and call .templateChanged() so that it gets update [12:45] d [12:47] Saviq, sorry let me explain because I am not sure I understood correctly [12:47] Saviq, hold on [12:48] dandrader: I can [12:48] 1. we receive a request for a carousel, at this point in order to know if the real category will be carousel we need to calculate a temporary cardWidth [12:49] 2. with this temporary cardWidth and results.count we decide if it'll be a carousel or a grid [12:50] on the other side, cardTool receives the initial request that will be carousel and cardWidth will be calculated for carousel, despite maybe being grid [12:51] Cimi, I know what you mean, and until now it's fine - one problem that we might need to think about is that this needs changing onCountChanged - as results come in, we might need to convert back to carousel again === MacSlow is now known as MacSlow|lunch [12:52] Saviq, correct [12:52] Cimi, which starts to feel again like this should all be internal to CardCarousel [12:53] Cimi, I don't want to push it all up to GenericScopeView, TBH [12:53] dandrader: https://code.launchpad.net/~gerboland/unity-mir/app-visible-property/+merge/207177 [12:54] greyback, thanks! will start using it now! [12:54] dandrader: np [12:55] Cimi, you can make the cardWidth binding into a getCardWidth function (and making the binding use that function), that's fine [12:56] Cimi, but make the grid fallback happen inside CardCarousel itself [12:56] Saviq, how do I make it happen? [12:56] Saviq, I'll have to have cardcarousel being a loader of a cardcarousel or cardfiltergrid [12:57] Cimi, something of the sort, yes [12:57] I'll try again === alan_g is now known as alan_g|lunch === MacSlow is now known as MacSlow|lunch [13:14] Saviq: is there someone I could borrow to test an OSK fix on N7 and N10? [13:14] greyback, /me [13:15] greyback, im happy to help test as well [13:15] Saviq: thanks. lp:~gerboland/unity-mir/app-visible-property/ is the branch [13:15] greyback, can I just build that or should I build on top of some other branch? [13:15] cwayne: Saviq maybe if you want to divvy up the work between you. The change touches code that positions surfaces, so please test that all apps are positioned ok [13:15] Saviq: can just build that branch [13:16] cwayne, I'll build packages in a few [13:16] sure thing, thanks [13:16] greyback, so it includes n7-rotate? [13:16] Saviq: yes [13:16] greyback, ok, sorry for pulling your tongue ;) [13:16] np, always good to check === dandrader is now known as dandrader|afk [13:28] mhr3, shall I make scopes activated directly yet? [13:31] cwayne, http://ubuntuone.com/6C2RbWS9APnnvSjPNkkNrF for unity-mir .debs [13:32] Saviq, thanks, downloading now. which bug is this supposed to fix again? [13:33] cwayne, bug #1281657 [13:33] bug 1281657 in unity-mir "Keyboard is hard to use on Manta, and occasionally hard to use on flo" [Undecided,In progress] https://launchpad.net/bugs/1281657 [13:34] ah brilliant === alan_g|lunch is now known as alan_g === dandrader|afk is now known as dandrader [13:48] Saviq, yea, do, we have a branch with which it will actually work [13:49] Saviq, it's already in the ppa even [13:50] greyback, looks SO MUCH BETTER on n10 now! === Trevinho_ is now known as Trevinho [13:50] OSK that is [13:50] cwayne: yay [13:51] seb128: so, the support for different-monitors scaling is in unity, but not in the configurations [13:51] Trevinho, right, but default config should be a factor of 1, e.g no change [13:51] greyback, and seems to work fine on n7 as well [13:51] seb128: so, for now we just scale up/down everything following the gnome text scaling factor, but the support for different factors will be land after === pete-woods-lunch is now known as pete-woods [13:52] seb128: mh, oh wait I misunderstood then [13:52] seb128: I hope that brandon didin't left in some debug data... [13:52] let me check [13:52] mhr3, done http://bazaar.launchpad.net/~unity-team/unity8/new-scopes/revision/657 [13:52] Trevinho, I've text-scaling-factor set to 1.25 [13:53] Trevinho, but that's for text scaling, not UI scaling... [13:53] seb128: isn't that value that gnome uses as well? [13:53] Saviq, else if would make it more readable :) [13:53] Trevinho, just for text I think [13:53] greyback, issue with sidestage though, apps supposed to open in sidestage are opening to the left [13:55] mhr3, pffy! [13:55] cwayne: can you give me more info? What device? What app? [13:55] seb128: mh, ok... We can revert that if you want... I.e. we'll just use our own config value [13:56] cwayne: have you the demo-stuff PPA installed? [13:56] greyback, n7, facebook app, yes i have demo-stuff [13:56] greyback, emailed you a picture [14:00] cwayne: ok, lemme repro [14:04] cwayne: fook I gave Saviq the wrong branch, it should be lp:~gerboland/unity-mir/demo-stuff-manta-osk-fix [14:04] give me a minute & I'll have packages built & ready [14:05] Saviq, I think it's better to put cardtool in genericscopeview [14:06] Saviq, and there do the switch [14:06] Saviq, otherwise it's like repeating bindings etc etc [14:07] greyback, pfft! [14:07] sorry [14:07] greyback, I can cross-build, will take seconds now [14:07] * greyback never got his Xbuild to work [14:07] greyback, it got broken in the mean time, but works again [14:08] greyback, so https://wiki.ubuntu.com/CrossBuilding and https://wiki.ubuntu.com/SimpleSbuild if you want [14:08] Saviq: ah really? Will try again maybe when things calm down a little. Yep those are the docs I was working from [14:08] (I mean the former is the usual way, the latter has some additional tips'n'tricks) [14:10] cwayne: https://chinstrap.canonical.com/~gerboland/debs.tar === MacSlow|lunch is now known as MacSlow [14:11] greyback, thank you for tar'ing it up :) [14:11] downloading/installing now [14:12] cwayne: sorry about the f*ck up [14:13] you're fired [14:14] kgunn, NOOOOOO [14:14] cwayne, same link, new packages [14:14] greyback, lol np [14:14] :P [14:15] kgunn: no, you're fired! [14:15] oh, wait...yeah...i guess we need him [14:15] everyone's fired! [14:15] Saviq, multi column layouts, in/out? [14:15] mhr3, didn't you enable it? [14:16] Saviq, shell needs to set the number of columns required [14:16] mhr3, ah [14:16] mhr3, in, I think, where can I test? [14:16] good question [14:16] greyback, so all i need is libunity-mir1? i don't need unity-mir too? [14:17] cyphase: yep, just libunity-mir1 [14:17] cwayne: ^^ [14:17] cyphase: sorry, bad tab complete [14:17] greyback, aweomse, installed, trying now [14:21] damn, i fixed one issue of the header being wronly positioned but it's not the one that seems to be causing the problem people see :( [14:21] well, at least one is down :D [14:21] tsdgeos, ouch [14:24] Saviq, scopes scope is doing layouts [14:25] mhr3, in previews you mean? [14:25] mhr3, will check it out soon [14:25] Saviq, yep, in previews [14:26] greyback, \o/ [14:26] working pretty well here [14:26] cwayne: on your command I'll merge it to the demo-stuff ppa [14:27] greyback, looks good to me, i've got no objections -- Saviq have yu tested it as well? [14:28] cwayne, yeah, worked fine and fixed the OSK issue [14:28] investigating a (probably unrelated) crash, though [14:28] sounds good to me then [14:28] greyback++ [14:29] cwayne: going in, thanks for the testing [14:29] hmm feature freeze tomorrow... so no Qt 5.2 in 14.04 LTS ;( [14:31] :( [14:32] Saviq: somebody hinted we could ignore that FF somewhen [14:32] Saviq: standup btw [14:32] there is no "ignore ff", but you can ask for a ff exception, and I think that was the plan [14:33] Saviq, tsdgeos: pad.lv/1278329 [14:33] https://bugs.launchpad.net/ubuntu/+source/qtbase-opensource-src/+bug/1278329 [14:33] Launchpad bug 1278329 in qtxmlpatterns-opensource-src (Ubuntu) "[FFe] Qt 5.2" [Undecided,New] [14:33] greyback, standup [14:33] Saviq: I'm there [14:33] oh wait, hangout [14:33] greyback, no [14:33] greyback, I'm blind [14:33] dednick, standup [14:39] MacSlow: right... did you try things like waitForRendering() etc? Sounds a bit like findChild() is called before everything is built up completely [14:39] mzanetti, that's in there [14:40] MacSlow: but it fails directly after a findChild() call, right? [14:40] Saviq: if there's anything I can immediately help with, let me know === boiko_ is now known as boiko [14:41] mzanetti, hard to say from the log-output [14:41] MacSlow: show it to me [14:41] karni, think you could take on adding a PageHeader to the Preview.qml? [14:41] Saviq: I'll look into it now. Any visuals for that? [14:41] karni, yeah, any of the preview visuals [14:42] Saviq: gotcha [14:42] karni, but wait, mhr3, do we have the string to display in the header from somewhere? [14:43] like "Book preview", "App preview" etc? [14:43] Saviq, what should it display? [14:43] /food [14:43] mhr3, ↑↑ [14:43] we do not have *that* [14:44] ok then [14:44] weren't those just placeholders? [14:44] mhr3, https://drive.google.com/a/canonical.com/?tab=co#folders/0B-a_7E3tDxOgTTBybG1TZG9GWnM [14:44] * karni follows [14:45] yea, i know, i always thought about it as placeholder, not the actual thing [14:45] karni, in that case, conditionally take out mascotImage out from mascotShape [14:45] mhr3, I think it's meant to be the actual thing ;) [14:45] but i guess it is the actual thing [14:46] karni, so, if there's a card background, mascotShape should be invisible, and mascotImage should be shown directly [14:46] Saviq: ACK [14:46] Saviq: you're looking at 00_Scopes.png ? [14:47] karni, yup, that [14:47] Saviq: ack [14:47] karni, something along those lines should work http://paste.ubuntu.com/6960227/ [14:47] karni, obviously the "true" part is dumb now [14:47] mhr3, so what do we do with text colour on dark background, then? [14:48] Saviq: :D gotcha [14:48] and that is a good question right there ↑ [14:48] Saviq, compute it [14:48] mhr3, you compute it! :P [14:48] background is an image, isn't it [14:48] hahahah [14:49] Saviq, design said the color is either light or dark [14:49] karni, color, gradient or image [14:49] Saviq, let's forget about images for now [14:50] If it's an image, it'd be hard to take that into consideration to compute it. Anyway, will add your change first, then think about font size. [14:50] *font color [14:50] mhr3, hey, did you see my bug-ping earlier? [14:50] karni, as mhr3 wrote - forget about images for now [14:51] seb128, yea, sorry not prio now [14:51] seb128, mwc too close [14:51] ok [14:51] mhr3, that's ok, I just want to make sure it's known/assigned, can be looked at after mwc/ff [14:52] mhr3: http://paste.ubuntu.com/6960251/ ?¿ [14:53] tsdgeos, the expection is not us [14:54] kgunn, sorry I missed standup. Will put my notes in manually [14:54] oh [14:54] mhr3: sounded scope-y [14:54] tsdgeos, sound mir-y to me :) [14:55] oh you mean lines 1 and 2 do have nothing to do with 3 and 4 [14:55] makes sense [14:55] right [14:55] karni, http://en.wikipedia.org/wiki/Luminance_%28relative%29 could help [14:55] Would it be possible for us to set override=true on the online music scope, so our guys can override the name with "Grooveshark"? our management wants to see "Grooveshark" in there for MWC. Unless you guys are fine to accept a patch to replace "Popular online" with "Grooveshark" (as I understand it, that's where there results come from) [14:55] Saviq: yeah, thought about it too. thanks for the link, tho! [14:55] karni, basically, convert to hsl, and base the decision on l [14:55] karni, colors have .a .r .g .b props from which you should be able to determine some threshold [14:56] * karni nods [14:57] MacSlow: http://paste.ubuntu.com/6960266 [14:58] karni, popular online is mostly from ubuntuone/7d [14:58] ack, I can ask them [14:59] mzanetti, this fixes the branch for you? [14:59] MacSlow: yes [14:59] mzanetti, so what about the waitForRendering() three lines up? [15:00] MacSlow: there's a difference in the argument [15:00] mzanetti, ha... the 's' :) [15:01] mzanetti, I'll try to see if that works with jenkins now [15:06] what's the package for installing minimal kde? [15:06] mzanetti, ? [15:06] * Cimi sick of those unity crashes [15:06] or tsdgeos [15:06] Cimi: dunno. I installed "kubuntu-desktop" [15:07] mzanetti, but this brings in kdm I believe [15:07] that's not really "minimal" though :D [15:07] Cimi: might install it. It still starts up lightdm here [15:07] mzanetti, yep, lightdm is default [15:07] tsdgeos: sure. but the awesomeness of KDE are the applications, not plasma. [15:07] so... no point in just installing plasma [15:07] I'll try it out [15:07] mzanetti: it's not what he asked ;-) [15:08] tsdgeos, you have answer? [15:08] nope :D [15:08] I don't need apps [15:08] I use qtcreator, a terminal, browser [15:08] Cimi: kde-workspace "may" be enough [15:09] but not sure if the kubuntu packages are tested to be standalone so it may as well explode in your face [15:09] * Cimi tries [15:10] mzanetti, any idea for dbgsym to improve http://pastebin.ubuntu.com/6960323/ for backtrace? or maybe you can see something interesting there already? [15:10] tsdgeos, you have a look, too ↑ [15:10] Saviq, might create a carouselUtils.js [15:10] In a hangout. will look in a bit [15:10] Saviq, with functions [15:10] Cimi, wasn't there one already? [15:11] Saviq, there's a carousel.js with functions for the carousel logic [15:11] Saviq: mzanetti: dalngling pointer ? [15:11] Saviq, those are not required by the components, but can be as integration for the dash use case [15:11] *component [15:12] Saviq, would shell support data image uri? [15:12] mhr3, no [15:13] owwwww [15:13] why not? [15:13] mhr3, because [15:13] mhr3, you didn't write an image provider for it [15:14] i didn't write one for http either and it works :) [15:14] mhr3, well, this one doesn't ;) [15:14] :/ [15:17] Saviq: does this paste belong to kgunn's bug report? [15:17] mzanetti, it might [15:17] Saviq: right edge? [15:17] mzanetti, no, mine is flo, so demo-stuff [15:20] tsdgeos, mzanetti goood [15:20] ? [15:20] mzanetti, kde works fine [15:20] Cimi: what's goood? [15:20] ah [15:20] kde/workspace [15:20] sure it does [15:21] :P [15:21] hah [15:21] no ita kbd layout [15:21] ---- [15:21] àìè [15:21] test [15:21] :D [15:21] ok [15:23] tsdgeos, you should tell your friends to reduce animations [15:23] there is a fade in for everything [15:24] Cimi: good thing is you can configure *everything* [15:25] mzanetti, indeed I immediately removed fade plugin [15:25] and probably even more sometimes [15:25] mzanetti, but default experience should be less of a trip to the amusement park [15:25] imho [15:26] Cimi: you're welcome to join the design team :-) [15:27] tsdgeos, that could be an exciting opportunity to do some UX :) [15:29] tsdgeos: FYI, on my unity8 MP, I'm actually waiting on the theme itself to test/fill the MP procedure acks [15:30] tsdgeos, I think a cool idea would be creating a plasma profile with my default settings, is it possible? [15:31] didrocks: OK [15:31] Cimi: not a plasma expert myself tbh, but i think not [15:31] Cimi: not sure if there's a ui for it. but settings are in ~/.kde/share/config/plasma-desktoprc [15:32] Cimi: also I'm not sure if this is the right channel for this topic :P [15:34] man, i need to rewrite the LVWPH [15:35] it's getting hard to understand what that millions of ifs do [15:35] i'm not even sure i need them, but.. [15:37] didrocks, is the plan to make the switch before MWC? [15:37] Saviq: the theme one? Depends on when design gives the final assets I guess [15:37] Saviq: the switch can be quick anyway [15:38] didrocks, assuming everything is fine straight up, which James pung me about just now that it might not be (indicator icons again) [15:38] (and the theme will inherit the old one, so no risk of explosion if you call something that doesn't exist) [15:38] Saviq: oh, finally, you need to do some code change for indicator icons? [15:38] didrocks, I don't know ;) [15:38] ahah :) [15:38] didrocks, chicken'n'egg [15:38] they changed the inner margins [15:38] so I bet you will need to change something [15:39] or the icons will "grow" [15:39] we'll have to see [15:39] yeah, let's install the new icon theme [15:39] try to my change [15:39] and see how it looks [15:40] didrocks, James just said he'll upload them within the next few days, so not for MWC anyway [15:40] ah, yesterday it was supposed to be done for yesterday evening :/ === alan_g is now known as alan_g|tea [15:49] how can I use units.gu inside a javascript? [15:50] I guess I cant [15:51] you have ideas apart creating an empty Item and adding JS there? [15:51] Cimi: why would you need gus in JS? [15:51] mzanetti, I wanted to write a utils for carousel [15:51] might just do a qml item [15:52] Cimi: perhaps you could just work with units. As in, you just use integers in js, and when you consume them in QML, you just do units.gu(jsValue) [15:53] I think I'll embed in an Item === alan_g|tea is now known as alan_g [15:54] Saviq, in any case, cardtool uses units.gu but does not import ubuntu components [15:54] Cimi, it works, doesn't it? ;P [15:56] :D [15:58] Saviq, are you still setting .active on the scope that's currently visible? [15:59] Trevinho and whoever else was involved: thanks for http://www.omgubuntu.co.uk/2014/02/unity-spread-keyboard-filtering-14-04 [15:59] mhr3, should indeed [15:59] gtg [15:59] olli_: np ;) [15:59] I only had to scratch an itch last sunday :D [15:59] Saviq: you ever seen a SIGILL in libcrypto running unity8 with gdb? [16:01] mzanetti, ^ ? [16:02] Saviq, on the cardtool (I'm trying to create a similar component for the carousel) [16:02] Saviq, there's width: 0 , height: 0, clip: true [16:02] Saviq, what's the purpose of this? and a fixme [16:03] ah I see [16:03] Cimi: you coming in tonight? [16:03] it's to avoid card to be rendered, and it's the only way [16:04] dednick, nope, will be in tomorrow [16:04] Cimi: not coming for drink?! [16:04] boo [16:04] dednick, was actually quite sick last week, I'm reducing number of drinks anyway :) [16:04] : [16:04] ) [16:04] I wasn't sick for alcohol [16:04] but doctor prohibited [16:05] dednick, think I gave my tummy too much meat in the last months :) [16:05] I irritated stomach, and it's not a joke [16:06] lol. dont really want to know! [16:06] hah [16:07] dednick, you still invited for mu burgers when I'm better [16:07] dednick, or tomorrow evening if you want to kill zombies with matthieu [16:07] Cimi: cool. as long as you havet given up on meat alltogether [16:08] Saviq: i have a "this shouldn't be needed but doing it is not wrong" kind of fix for the header, want to see and try-test it? [16:08] Cimi: eh. been in twice this week. don't think i can manage another trip into city this week :) [16:08] such a strain actually having to 'go to work' [16:09] dednick, I'll organise a lan party the week of the 10th march then, when matthieu is back [16:10] Cimi: cool [16:10] dednick, mhr3 is invited too if he removes the birth control from his phone [16:12] Cimi, if you insist i'll bring my work n4, which you've unvirginized yourself :P [16:12] * Cimi :D [16:13] Saviq, when I create an item that works as a tool [16:14] Saviq, can I use real width and height? [16:14] Saviq, or better to set that to 0 and use a variable like viewWidth? [16:19] mhr3: btw was it intentional to put card-background in template? that means all cards within category will have same background. [16:20] karni, that's default [16:20] results can still have different background [16:20] s [16:20] mhr3: I mean, the visual contains cards with different backgrounds. oh, ok [16:21] seb128, you know anything about this? https://bugs.launchpad.net/ubuntu/+source/click/+bug/1275828 [16:21] Launchpad bug 1275828 in click (Ubuntu) "click process keeps dying and spawning again continuously" [Undecided,Confirmed] [16:21] it's quite annoying, and I have the CPU fan throttling every 5 seconds [16:22] no [16:22] can you look in ps afx what process is spawning it? [16:23] seb128, http://paste.ubuntu.com/6960649/ [16:24] Cimi, indicator-datetime [16:24] Cimi, what version are you running? charles_ and tedg said that was supposed to be fixed (that was 10 days ago) [16:25] seb128, dist upgraded I believe [16:25] Cimi, dpkg -l | grep indicator-datetime [16:26] Cimi, http://bazaar.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.14.04/revision/301 was the fix I was thinking about (that was on the greeter) === dandrader is now known as dandrader|lunch [16:26] seb128, 13.10.0+14.04.20140217-0ubuntu1 [16:26] k, that includes the fix [16:27] so probably another bug [16:27] talk to charles_ ;-) [16:27] seb128, let me kill the service [16:29] seb128, it's actually datetime service that keeps dying [16:30] Cimi, does it segfaults? [16:30] dednick, that's expected [16:30] seb128, I suppose I have to run it with gdb to see that? [16:30] dednick, it's libcrypto interrogating the CPU [16:30] seb128, it keeps changing PID [16:30] Cimi, yes, just "gdb -p $(pidof indicator-datetime-service)" [16:30] Cimi, read the FIXME? ;) [16:31] Saviq, read it, was asking a more general question [16:31] Cimi, if that doesn't work, "stop indicator-datetime" and run it manually [16:31] It's probably getting respawned by Upstart [16:31] Cimi, CardTool is meant to be invisible, but unless it's visible the results are wrong [16:31] is there anything in /var/crash ? [16:32] Cimi, so I had to actually render it, but hide it with that hack [16:32] tedg, hey, do you plan to do a land request for https://code.launchpad.net/~charlesk/indicator-datetime/alarms/+merge/204420 ? want me to do it? [16:32] seb128, I cannot stop it that way [16:33] Cimi, why not? [16:33] Cimi, what user is owning the process? [16:33] seb128, lightdm [16:33] k [16:33] so it's a bug in the greeter profile [16:34] seb128, Makes sense, but I'd check with charles_ just to ensure where it is. I thought it was supposed to land. [16:34] Seems there's a couple bugs in errors: https://errors.ubuntu.com/?release=Ubuntu%2014.04&package=indicator-datetime&period=month [16:34] seb128, or actually maybe it's the user [16:34] seb128, in any case, stop indicator-datetime gives me unknown instance [16:34] and empty then === gatox is now known as gatox_lunch [16:35] tedg, right, it's annoying that retracing fail that often atm [16:35] charles_, could you look at https://bugs.launchpad.net/ubuntu/+source/indicator-datetime/+bug/1280341 ? [16:35] Launchpad bug 1280341 in indicator-datetime (Ubuntu) "indicator-datetime-service crashed with SIGSEGV in strlen()" [Medium,New] [16:35] tedg, nothing in var/crash [16:35] Yeah, the second one is as a user not lightdm. [16:35] Cimi, Hmm, anything interesting in ~/.cache/upstart/indicator-datetime.log ? [16:36] Cimi, did you try the "sudo gdb -p $(pidof indicator-datetime-service)"? === cyphermox_ is now known as cyphermox [16:36] tedg, it's empyu [16:36] tedg, no file [16:37] That's really weird. [16:37] Cimi, can you "ps aufx |less" and look what user it's running from there? [16:37] look for click [16:38] seb128, Is there someone looking at the retraces? Do we need to convince ev to take his old job back? ;-) [16:38] seb128, http://paste.ubuntu.com/6960711/ [16:38] tedg, I pinged bdmurray yesterday and he said he would have a look, let me ask again on #ubuntu-devel [16:39] tedg, charles_: ^ bt [16:39] That is odd. [16:39] What's the error? [16:39] Cimi, can you "c" it and see if you hit a segfault? [16:39] then bt [16:39] ok [16:40] hm lookint at ApplicationManager in Unity-Mir - should appId be some kind of unique? allowing only a single application instance per application id? [16:40] *logging [16:41] Cimi, Did you use the clock app to set alarms? [16:41] no [16:41] anpok, We use Upstart to ensure we have unique instances per appid. [16:42] tsdgeos, hit me [16:42] anpok, The logging of that app id running then goes to ~/.cache/upstart/application-click-$appid.log [16:42] Saviq: https://code.launchpad.net/~aacid/unity8/new-scopes-hotfix-header-position/+merge/207246 [16:43] Cimi, Do you have an click packages installed? click list ? [16:43] tedg, no [16:43] tedg, I think my laptop must have been eating vodka sauce [16:43] tedg: is there any process that cleans up ~/.cache/upstart/ occasionally? I've 2.3MB in that dir with just 2 weeks usage. [16:44] greyback, Yeah, logrotate does [16:44] tedg: ah good to know [16:44] Cimi, Mmm, your laptop has good taste. [16:45] Logrotate only runs though on login. Perhaps we need to make it part of a cron job as well, since on a phone you could theoretically log in for months at a time. [16:46] * tedg assumes our software has no bugs, ever :-) [16:46] tsdgeos, there's no test we could write for that is there? [16:46] Saviq: if there is a test i could probably fix it "correctly" [16:47] tsdgeos, understood [16:47] Saviq: ideally i need to sit a few days and simplify the codebase of LVWPH a bit [16:47] tsdgeos, ok, later then [16:47] +1 [16:47] Saviq: if you can't reproduce it with this (i couldn't) i'd add it to new-scopes and then see what we do nextweek (or nextweek+1) [16:48] tsdgeos, yup [16:51] Cimi, did you get a bt of the segfault? [16:52] seb128, no [16:52] Cimi, can you get one? [16:52] seb128, it segfaults immediately [16:52] well, you had it in gdb [16:52] seb128, or maybe it doesn't segfault but it closes [16:52] just to "c" and "bt" [16:52] seb128, with continuing, it exits normally [16:52] seb128, exits normally [16:53] k [16:53] seb128, then respawns [16:56] weird [16:57] Cimi, did you play with any phone app before that started? or with alarms from the toolkit? [17:01] mzanetti, one more seemingly related stack: http://pastebin.ubuntu.com/6960818/ [17:02] Saviq: so, I did the mascot shape part, but I'm having trouble converting template['card-background'] to color type [17:02] to calculate font color [17:03] karni, just take off of UbuntuShape.color instead [17:03] karni, it should be a color there === bschaefer_ is now known as bschaefer [17:03] Saviq: I mean I need the Card background color [17:03] Saviq: interesting... this is a different one. and actually one that makes sense [17:04] the Card is AbstractButton. that is an UbuntuShape as well? [17:04] karni, no, the background is [17:04] Saviq: seems a bug in applicationmanager. it emits applicationAdded() but then returns 0 when being asked for that application. will look at it [17:04] karni, background.color [17:04] mzanetti, k [17:05] Saviq: any hints on reproducing it? [17:05] mzanetti, I merged new-scopes into right edge and run_on_device it [17:05] seb128, don't remember [17:06] mzanetti, every time I wanted to launch app it crashed [17:06] Cimi, well, you need charles_, but he doesn't seem to be around atm, maybe try again later? [17:06] Saviq: ah... hmm... I've seen that when unity8 and unity-mir were not matching [17:06] mzanetti, think a clean build needed? [17:06] Saviq: as in unity-mir from trunk used with unity8 form right edge for example [17:06] mzanetti, nope, all right edge [17:07] hmm... I know that I've seen this every time when I reflashed the device and did a make install on a unity-mir builddir compiled with a previous image [17:08] Saviq: please give it a try with clean rebuilding unity-mir [17:08] mzanetti, unity-mir is from ppa there [17:08] mzanetti, only thing I'm building is unity8, but will do a clean build [17:09] cwayne: ping [17:09] mzanetti, pong [17:09] cwayne: trying mwc-flash here [17:09] 404 on the first 3 files [17:09] mzanetti, you have a device plugged in and powered on? [17:09] it's probably failing to do the adb shell getprop ro.product.device [17:10] ah right... it says device not found in between the lines [17:10] righto, i thought i'd made it error out if that's the case, but i guess i didn't, will add today [17:11] cwayne: much better :) [17:11] works now [17:11] great :) [17:11] thought having it in the bootloader would be enough for getprop [17:12] nope, i guess i could maybe try and do that too with fastboot getvar [17:15] hhm shouldnt e.g. upstart-app-launch dialer-app start the application? [17:23] maybe i broke it [17:30] mzanetti, yeah, clean rebuild got rid of that issue [17:35] Saviq: ok. thanks [17:40] my code sucks [17:41] cwayne: the tomtom app crashes here [17:41] mzanetti, known [17:42] cwayne: hmm... seems whenever something tries to access the GPS [17:44] mzanetti, hmm, try adb shell restart ubuntu-location-service [17:46] greyback, no problem about that bad tab complete; though it hasn't happened as often recently (for some reason), i do get it relatively often :) === gatox_lunch is now known as gatox [18:06] Saviq, something like that is very ugly, right? http://paste.ubuntu.com/6961175/ [18:07] Cimi, cardtool should not know about carouseltool [18:07] Cimi, but other than that this'll have to wait until tomorrow, I'm well past my bed time [18:08] Saviq, I added it to cardtool cause the calculation of card width is inside carouseltool [18:08] but we can chat about it tomorrow morning, early [18:09] Saviq, i remember you said the UbuntuShape was being rewritten to do async loading, that that will result in smooth scrolling in the dash. Do you know if thats inprogress ? [18:09] om26er, it is being rewritten (and actually almost complete since December AFAIK), whether it will help with scrolling in dash we'll have to see [18:09] om26er, loicm or Kaleo should have more data on when will that land [18:10] Saviq, ok, will ask them [18:11] Saviq: that change you asked for (toggling ubuntu shape, font colors based on background) - should this be against trunk? [18:11] or new-scopes is easier/faster? [18:12] karni, better against trunk, but higher standards :) [18:12] ok === dandrader|lunch is now known as dandrader [18:14] Saviq: I cleaned up font colors including when in overlay, but that is only new new-scopes. However, if you prefer, I'll omit like 3 lines, and get my changes against trunk. [18:14] (overlay is not in trunk) [18:15] cwayne: still around [18:15] I've packages available here to fix the left edge swiping bug: https://chinstrap.canonical.com/~gerboland/mwc/ [18:16] Saviq, seen the comment from katie? no double shaped stuff :) [18:16] mhr3, yeah I did [18:16] mhr3, but IT'S NOT THAT EASY [18:16] noooooo [18:16] i don't want to hear that [18:20] greyback, absolutely [18:20] cwayne: thanks [18:21] greyback, this is the n7 one right [18:21] cwayne: both n7 & n10 [18:23] kgunn, you duped bug #1127665 [18:23] bug 1127665 in Unity 8 "[SIDESTAGE] Dash isn't visible when only sidestage app running " [Medium,Triaged] https://launchpad.net/bugs/1127665 [18:24] and o/ [18:24] later [18:25] damn it...i actually searched before logging.... [18:25] i'm not a good searcher [18:26] Saviq: g'night! for tomorrow: https://code.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/+merge/207278 [18:27] karni, 93 + font.strikeout: true ? [18:27] Saviq: oldPrice is striked out on designs [18:27] on visuals, I mean [18:27] karni, ah, you did some of that, k [18:27] karni, not used, but hey... [18:27] it's black, and striked out/through [18:27] :) [18:28] Saviq: that ff990000 is red color of price attribute [18:28] cwayne: I'm hitting eod, would you mind emailing me if you find any issues or not. I'll try to land it if not [18:28] in case you wonder. [18:28] karni, why ff? [18:28] Saviq: so it's opaque [18:28] AARRGGBB [18:28] karni, isn't that default? [18:28] greyback, it works [18:28] karni, it has nothing to do with the opacity prop [18:28] Saviq: maybe it is. maybe, so I guess it doesn't hurt? :) [18:29] sorry, just finished testing it [18:29] karni, raises eyebrows ;) [18:29] it works quite well ont he n7 at least, i still need to try n10 greyback [18:29] Saviq: well, it has to do with the color. well.. ok, next time I'll avoid it. [18:29] :) [18:29] karni, we generally only add AA if it's != ff [18:29] karni, template && template["card-background"] !== undefined [18:29] cwayne: /me gives tired yay - thank you :) [18:30] karni, template doesn't necessarily need the default defined [18:30] karni, components["background"] is more important [18:30] ah [18:30] my bad [18:30] karni, generally, just make it a prop [18:30] karni, and pass from card [18:30] ok [18:30] karni, from background.visible [18:31] karni, background.color in CardHeader is reaching out of scope [18:31] Saviq: background.visible could be a color? [18:31] karni, should be a prop again [18:31] karni, no, visible is bool [18:31] greyback, no, thank you :) [18:31] ack. Saviq, can you please leave a comment (or just paste this convo), I'll get to it in the morning. super tired. [18:31] that's a huuuge fix [18:31] karni, k [18:32] Saviq: appreciated! :) [18:32] karni, I'll actually fix it up and merge, since it's there, I'd like to get it in people's hands tomorrow [18:32] Saviq: I didn't mean to cause you more work in the evening [18:32] karni, you didn't, you did it all :) [18:32] I'm just falling asleep here [18:32] Saviq: ok man, thanks so much [18:33] It's owned by unity-team [18:33] greyback, seems to work well on n10 as well [18:34] cwayne: oh goodie [18:34] cwayne: if you discover anything bad, please mail. I'll be back for a little in ~4 hours [18:34] * greyback out [18:50] robru, tested, looks good, pls publish once the script realizes it actually built :) [18:50] * mhr3 out [18:50] mhr3, sure [18:51] gonna run, bbiab === dandrader is now known as dandrader|afk [20:20] tedg: i have some more questions on starting.. [20:20] anpok, Cool, what's up? [20:20] I currently work on https://bugs.launchpad.net/unity-mir/+bug/1240400 [20:20] Launchpad bug 1240400 in mir (Ubuntu) "dialer-app crashed with SIGSEGV in __GI___pthread_mutex_lock()" [Critical,In progress] [20:21] in the bug case dialer-app was started from cli withoug upstart.. just /usr/bin/dialer-app --desktop_file_hint.. [20:21] and on incoming call through url-dispatcher i think a second instance of dialer-app is started again [20:21] i got unity-mir not to crash on that case [20:22] but still it misbehaves [20:22] Yeah, you really shouldn't launch an app like that. It won't work for very long. [20:22] Mir should start blocking apps launched like that. [20:22] Which might be part of the problem [20:23] Yeah, "Could not create application instance" that's Mir rejecting the app. [20:23] yeah with that the issue would go away [20:23] right now .. it starts the second instance. and does not allow to start it since there is already one running [20:23] but it manages to open the mir session anyhow [20:23] so there is another buffer on screen [20:24] Why can't the app be started with Upstart? [20:24] but the already running dialer takes the phone call.. [20:24] sure it can [20:24] I thought I could still support that "legacy" [20:24] of apps started on the cli [20:24] without upstart [20:24] i mean [20:25] No, we're not supporting that. [20:26] so it is safe to assume that an app-id is always unique [20:26] Uhm, for click yes. Legacy apps that don't use Upstart for single instance will have multiple connections for that appid. [20:27] ah [20:27] That's nothing on the phone today, but in the convergence cases those can come up. [20:27] ok then the changes I did were reasonable [20:27] They'll still have an upstart instance for each running, but the instance ID for that instance has a timestamp as part of its instance ID. [20:27] ok.. the code I am working on only has an app-id [20:28] or receives triggers that only contain an app-id [20:28] so the stopping of a second instance can easily teardown the first instance [20:29] Hmm, okay. In the legacy case you'd want to be able to close the instances independently. [20:30] there is probably an upstart-api for that? [20:31] Hmm, I think the UAL API only allows for shutting down a full AppID. [20:31] * tedg looks [20:31] Hmm, yeah. [20:31] hm right now unity-mir registers some callbacks [20:32] So you're saying we need a way to break apart the instances in the API. [20:32] hm and somehow gets the appid from that [20:32] Yup, those callbacks tell when an app starts, etc. [20:33] hm I thought something like .. also getting the PID [20:33] but instance id should do the same trick I guess [20:34] o_O there is no api right now [20:34] for that.. [20:35] Yes, there isn't. [20:35] But frankly, we don't have a convergence case really working anywhere :-) [20:35] But we should not loose that work item. [20:36] kgunn, Are you guys going to have a "convergence" blueprint for this upcoming UDS? [20:36] I think I have a few work items on it :-) [20:37] tedg: technically we already do....and granted not the most gardened blueprint...but if you've got "stuff" feel free to dump here [20:37] https://blueprints.launchpad.net/ubuntu/+spec/client-1404-mir-converged [20:37] or here [20:38] https://blueprints.launchpad.net/ubuntu/+spec/client-1404-unity-ui-windowmanager [20:38] that way they don't get lost [20:38] and we'll do the good old cleanup when we go to plan 14.10 [20:38] and i'll be sure to carry them over [20:38] Yeah, I can do that. [20:39] We're not talking 14.10 in a few weeks, right? [20:39] It's still 14.04 [20:39] ok then i would leave it as is - crash fixed - but no support for multiple instances [20:39] tedg: yes... [20:39] we're 14.04 all the way thru the 04 :) [20:40] Heh, to the 4! [20:40] tedg: you'll notice a couple of things i've dumped in there as placeholders as well... [20:50] kgunn, do you know if greyback's unity-mir sidestage fixes made it into the archive or ppa? [20:50] the ones here: https://chinstrap.canonical.com/~gerboland/mwc/ [20:50] cwayne: which specifically... ? [20:50] ah [20:50] checking... [20:50] i never saw an MR, just tested the debs directly, so didn't know where to look [20:51] yeah he's not MP [20:51] ing [20:51] he's just pushing to the n7 unity-mir branch [20:51] checking... [20:53] cwayne: i'm gonna say he did not put those into the branch...i'm gonna guess he probably wanted you to test ? [20:53] or i can [20:53] once tested he'll probably merge the code [20:54] kgunn, i tested, it was just bad timing cus it was right at his eod :) [20:54] plus he'll be back on irc in about 2 hours or so [20:54] kgunn, it seems to work well on n7 and n10 btw [20:54] ah right, he mentioned that, i completely forgot [20:54] i'll wait for that then :) [20:54] cwayne: ok great...we can ambush him [20:54] :) [21:00] after using the new unity heavily for like 3 days, the old one seems so.... inadequate [21:04] cwayne, glad yo read that :) [21:08] Saviq, :) === jhodapp is now known as jhodapp|afk === dandrader|afk is now known as dandrader [23:22] mterry: you still around ? [23:22] kgunn, yup [23:23] hey...so i hadn't looked yet at details...but it seems https://code.launchpad.net/~mterry/unity-mir/alpha-greeter/+merge/204069 [23:23] is conflicting [23:23] needs a rebase [23:24] kgunn, I could believe it. Let me check [23:25] mterry: thanks....i got it in a silo right now to land...so, if you're not leaving super quick... [23:25] only if its easy... [23:26] kgunn, it's easy. Fixed, but just confirming still builds nice and all [23:26] cool [23:28] * mterry wants to enforce a programming law that #includes be alpha-sorted [23:28] kgunn, updated [23:29] kgunn, I am about to run, anything else? [23:33] * mterry runs [23:42] When I disconnect my external monitor, and reconnect it. Sometimes windows 'disappear'. So right now I have 4 gnome-terminal windows that I can see when I hold alt+tab, though if I select them.. they don't appear. [23:42] How can I fix this and get it to stop happening? [23:43] Also, I have a specific window layout each time I setup my computer in the morning. I have chrome/sublime/hipchat/4 terminal windows, all in a specific layout.. how can I get that layout to be 'setup' fast, without manually dragging? The control/alt/num key pad stuff doesn't fit my use case, since I allocate screen space differently.