=== _salem is now known as salem_ [05:09] ping mzanetti: if the dash is an app, why can't I launch it with ubuntu-launch-app ? === salem_ is now known as _salem [07:49] Saviq: there? [08:14] tsdgeos, here [08:14] elopio, you should be able to [08:15] Saviq: was wondering if base the scopes activation thing on dash overview or on plain trunk [08:15] tsdgeos, overview [08:15] ok [08:15] tsdgeos, since it has exceptions for it [08:16] Saviq: "on top of" or "inside" overview? [08:16] tsdgeos, on top of please [08:16] okidoki === vrruiz_ is now known as rvr [08:48] mzanetti, can you please have a think and list any cases where you think just url handling for scope:/// won't cut it? [08:49] (in the MP that is) [08:49] ok [08:50] Saviq: what about line 261 in the diff? is that still wanted? I wasn't unsure [08:50] err. I was unsure :D [08:51] * Saviq looks [08:51] mzanetti, 261 is a newline ;) [08:53] mzanetti, ugh, wrong mp [08:53] no, still newline :D [08:54] Saviq: 621, sorry [08:54] mzanetti, yeah, nothing like that [08:54] mzanetti, what's more [08:55] http://paste.ubuntu.com/7893171/ [08:55] mzanetti, but not necessarily in your MP [08:55] ok [08:56] mzanetti, basically we'll only be changing focus inside the dash if you press BFB or left-swipe [08:57] Saviq: atm we're using different animations when pressing the home button [08:58] Saviq: depending on whether the greeter is locked or not, we animate to home or not [08:58] mzanetti, when dash isn't visible, there will be no animation inside it now [08:58] Saviq: ans also, on left edge swipe we reset it to the top, while we don't do on home button press [09:00] so well, I don't really care... can drop the whole communicator thing [09:00] but I'm afraid the url hadler is not in place yet, is it? [09:00] mzanetti, sure it's not [09:00] mzanetti, I'm not saying you should drop it now [09:01] mzanetti, as we can't really drop it, we need to replace it with the handler [09:01] mzanetti, but I just wanted to have a list of doubts to shoot down ;) [09:01] ok. so I'll find a place to write this down as a list [09:01] thanks [09:01] not the MP, that'll vanish once merged [09:01] some doc or something [09:02] Saviq: didn't we patch Qt to read the QT_TESTABILTIY env var? [09:02] mzanetti, QT_LOAD_TESTABILITY=1 [09:02] or do I still need to read that manually in main.cc? [09:03] elopio complains it wouldn't work [09:08] pstolowski: https://code.launchpad.net/~aacid/unity8/scopeActivatePreview/+merge/228624 this should be the activate/preview thing, no? [09:08] Saviq: ↑ [09:10] tsdgeos, the clickscope hack is still needed for applications; also, preview should be called for scope:// uris in scopes scope [09:11] scopes scope does not exist anymore [09:11] it's called dash overview [09:11] and does not go thorugh this route [09:12] pstolowski: so do it for scope:// and application:// ? [09:12] or both store and click scope return application:// ? [09:12] * tsdgeos checks [09:13] tsdgeos, how do I get this new overview page? I flashed today but it's not there [09:13] pstolowski: it's not merged yet [09:13] but this branch depends on that one [09:13] so for this branch purposes it is [09:15] anyway i've brought back the clickscope hack [09:20] tsdgeos, no, clickscope doesn't return urls yet [09:20] Saviq: it does for me :D [09:21] tsdgeos, for all apps? [09:21] ah [09:21] not sure [09:21] for a few i clicked [09:21] tsdgeos, basically click vs. not click is the important part [09:21] oki [09:23] tsdgeos, but yeah, if (clickscope || scope://) is good [09:23] tsdgeos, lgtm. can we have a silo with this and any dependencies? [09:23] pstolowski: someone needs to review dash_overview :/ [09:23] pstolowski, I'll try and get a silo with overview and firends [09:23] but sure it can be silo'ed [09:23] tsdgeos, does preview in overview work? (and not work for non-scope:/// urls?) [09:24] tsdgeos, could you pls updated the commit message, it's entirely correct? [09:26] pstolowski: updated [09:30] Saviq: autopilot borked again [09:30] any idea what ahppend? [09:30] https://jenkins.qa.ubuntu.com/job/generic-deb-autopilot-runner-mako/2766/console [09:31] tsdgeos, humpf! [09:31] tsdgeos, it didn't return the PID [09:31] "unity8 start/running" should have ", process 2467" [09:31] or so [09:32] (it does here)\ [09:32] :/ [09:32] seems a pretty constant failure [09:32] maybe it was too quick [09:32] maybe it's a change somehwere? [09:32] weeeeirds [09:33] like even "start unity8" returns a PID [09:33] tsdgeos, I'll have a look soon [09:34] oki === Estilanda_ is now known as Estilanda [09:44] mzanetti, ping [09:44] Cimi: peng! [09:44] mzanetti, :D [09:44] mzanetti, I had a look at the count emblem [09:45] want a reassurance on width: Math.min(root.itemWidth, Math.max(units.gu(2), countLabel.implicitWidth + units.gu(1))) [09:45] ok [09:45] what exactly? [09:45] want a test for it or what? [09:45] itemWidth is always more than 2? [09:46] is it? this code wouldn't suggest that [09:46] I am wondering if we could have a portrait like ubuntushape [09:47] finally got a chance to checkout phone-right-edge ppa -- looks awesome! cant wait to see it in image [09:47] in case itemWidth is less than 2gu [09:47] oh.. [09:47] now I see your thought [09:47] let me check [09:47] this piece of code changed and I was wondering why [09:48] Cimi: well, it didn't change much, I just decreased values from 2 gu's to 1 gu. [09:48] Cimi: ok... just checked the code [09:49] Cimi: yes, itemWidth is the size of the whole launcher icon [09:49] Cimi: so yes, that's always larger than 2 gus [09:49] ok [09:49] Cimi: and if it isn't the launcher is so tiny that we need to shrink those emblems too [09:52] Saviq: can you give me a pointer on how the dash was loading fake scopes for AP tests so far? [09:54] mzanetti, it exported QML2_IMPORT_PATH correctly [09:54] Saviq: so that should work just the same with the standalone dash app, right? [09:57] mzanetti, label is not centered vertically but that is qml fault [09:57] mzanetti, we could use an offset [09:57] 1 pixel maybe [09:57] mzanetti, yeah, just need to make sure you launch the standalone dash with the right import path set [09:57] Saviq: ack, thanks [09:58] Cimi: http://i.imgur.com/YFYv6De.png [09:58] I think it looks fine [09:58] mzanetti, tests/autopilot/unity8/__init__.py [09:58] mzanetti, nope [09:58] the 1 might be a bit off, but for instance the 5 works quite fine, no? [09:58] mzanetti, it is at least 1px down [09:58] mzanetti, they're all off [09:59] oh, vertically [09:59] mzanetti, yes [09:59] hmm [09:59] mzanetti, also I'd remove the border of the ubuntushape [09:59] there is like an inset here [09:59] doesn't look nice [10:03] mzanetti, OK? [10:04] Cimi: http://i.imgur.com/3VZMgXZ.png <- better? [10:04] mzanetti, ok for me [10:04] (thats a verticalCenterOffset of -units.dp(.5) [10:04] Saviq: is that ok? ^ [10:04] units.dp(1) is too much [10:05] the theory would disagree :D [10:05] mzanetti, also now that I see [10:05] mzanetti, is there a separator between ubuntu dash button and the rest of launcher? [10:05] mzanetti, we can remove it with this orange design [10:06] Cimi: dunno. need to talk to Vesa/Esti [10:06] Cimi: there is a separator, yes [10:06] looks unneeded [10:06] we can remove it in another branch [10:07] mzanetti, in practice it should be a static fraction of font.pixelSize [10:07] Cimi: pushed the above [10:07] (static assuming the font won't change that is) ;) [10:07] hmm... [10:07] but really we should file a QTBUG [10:08] that QML doesn't know about font centerLine or whatnot [10:08] mzanetti, // FIXME too [10:08] Saviq: ok with having the units.dp(.5) offset workaround? [10:09] with link of qt bug :D [10:09] I can open it [10:09] mzanetti, yeah, nothing better we can do really [10:09] Cimi: is there a bug already? [10:09] writing it [10:09] Cimi: ah ok, cool [10:09] Cimi: yeah, when you have it, let me know and I'll add the fixme [10:14] Saviq, mzanetti https://bugreports.qt-project.org/browse/QTBUG-40479 [10:22] Cimi: thanks. I've pushed the FIXME [10:23] dandrader: https://code.launchpad.net/~paulliu/unity8/reboot_140728/+merge/228485 [10:35] tsdgeos, mhr3, https://docs.google.com/a/canonical.com/document/d/1NmiM4UCnJgf6IEawmfyTOHRNAA5ZGrqpyrPqPOibwc8/edit?disco=AAAAAKGrPoI [10:40] Saviq: looks ok to me [10:40] actually [10:41] Saviq: wouldn't it just be simpler to have threshold as a number not as a widget [10:41] so i don't have to go over the list and find it [10:41] what does it add being part of the list? [10:41] tsdgeos, right [10:41] * Saviq fixes [10:43] tsdgeos, done [10:43] don't like the name :D [10:43] pete-woods, you probably want to be at least aware of this ^^ ;) [10:43] what is collapsed-widgets ? [10:44] i mean doesn't feel very descriptive [10:44] tsdgeos, better name welcome, but it's the number of widgets that are visible in a collapsed state [10:44] ok [10:45] (from the top, of course) [10:45] tsdgeos, you need to remerge https://code.launchpad.net/~aacid/unity8/unfocusSearchEntryOnHistoryPick with the prerequisite [10:45] so 0 means it's jst collapsed totally [10:45] which is probably an invalid value [10:45] no? [10:45] tsdgeos, yeah, it is, has to be > 1 [10:45] erm [10:45] >= 1 [10:46] tsdgeos, so we could say 0 default, but wanted to be more invalid ;) [10:46] Saviq: sure [10:46] Saviq: hmmm, don't get anything on unfocusSearch [10:46] http://paste.ubuntu.com/7893898/ [10:46] seems as it's already merged [10:47] hmm [10:47] what do you get? [10:47] tsdgeos, here's the order it was merging in https://ci-train.ubuntu.com/job/landing-005-1-build/197/console [10:47] tsdgeos, maybe we just need to reorder? [10:47] let me see [10:47] it does complain about criss-cross though [10:50] i guess scopeSearchHintText makes it weird too [10:50] since it also has pageheaderqml [10:51] give me a sec [10:51] tsdgeos, let's see, I kicked it with explicit order, maybe that'll help [10:51] tsdgeos, you might want to confirm bug #1313697 ;) [10:51] bug 1313697 in Canonical Upstream To Distro "Train should try --weave if criss-cross merge is detected" [Undecided,New] https://launchpad.net/bugs/1313697 [10:52] * Saviq must've been annoyed by the issue when I wrote that description :D [10:53] Saviq: is https://code.launchpad.net/~unity-team/unity-scopes-shell/catch-no-search-hint/+merge/228288 part of the silo? [10:53] tsdgeos, yes [10:53] oki [10:53] tsdgeos, http://people.canonical.com/~platform/citrain_dashboard/#?q=landing-005 [10:54] tsdgeos, I think it merged fine [10:54] tsdgeos, with forced order [10:54] ugh [10:54] but then [10:54] https://ci-train.ubuntu.com/job/landing-005-1-build/198/console [10:54] overview [10:55] arg [10:55] tsdgeos, you're conflicting with yourself so much ;D [10:55] i know [10:55] stupid thing [10:55] inception [10:55] i didn't expect us to merge overview with the rest of things at the same time tbh [10:56] want me to merge all the other branches to dash_overview? [10:56] and make it be the last one that depends on the rest? [10:57] larsu, hey, could you take a look at https://bugs.launchpad.net/gsettings-qt/+bug/1349787 ? [10:57] Launchpad bug 1349787 in gsettings-qt "application crashes when using gsettings binding for schema with a{ss} type key" [Undecided,New] [10:57] Saviq: merge or not? [10:58] tsdgeos, I think so yeah [10:58] ok [10:59] Saviq: but then it'll be quite difficult to review in launchpad [10:59] tsdgeos, why, set the prereq? [10:59] Saviq: can't set more than one, no? [10:59] tsdgeos, but the other ones are depending on one another already? [10:59] probably [10:59] tsdgeos, you just need to prereq on the last one, no? [10:59] ok, let's see [10:59] there's also lp:~aacid/unity8/compile-scopes-v3 [11:00] unless it actually conflicts with two, then you need to chain them... [11:00] which you don't really need in the silo if you have dash overview [11:00] it was supposed to land them before [11:00] k, can drop [11:00] but no worries [11:00] i'll just merge it too [11:00] tsdgeos, if it's a conflict I'll drop it from the silo [11:00] tsdgeos, no problem for me [11:00] Saviq: leave it, i'll merge it [11:00] tsdgeos, k [11:06] Saviq: can you try again? [11:06] tsdgeos, did you resubmit? [11:06] push? [11:06] yes [11:06] ah not with the prereq [11:07] tsdgeos, yeah, please change the prereq [11:07] tsdgeos, and you probably want to merge it into scopeActivate (no need to resubmit there) [11:07] done [11:08] both [11:08] let's see if it works [11:08] got it [11:11] gah [11:11] no [11:11] ah no still on it [11:15] pfff [11:15] all fails [11:16] Saviq: failed at the previous error :S [11:16] tsdgeos, ah crap [11:16] forgot to force ordering [11:16] Saviq: i'll chain scopeSearchHintText after lp:~aacid/unity8/showSearchOnSearchQueryChange even if it's not really needed [11:16] and then we have a single chain of things [11:16] tsdgeos, should be fine now [11:17] ok [11:17] * tsdgeos no touchy [11:17] hello, my name is Andrea Cimitan and I text conflict === pete-woods is now known as pete-woods|lunch === MacSlow is now known as MacSlow|lunch [11:21] Saviq: honestly i feel that all the other small fixes should go in before the dash_overview, makes it for a smaller and easier to control landing [11:21] but if you prefer all together, all together it is :) [11:22] tsdgeos, we might still split it out [11:22] tsdgeos, Text conflict in qml/Dash/GenericScopeView.qml :| [11:22] garg [11:23] i think it's scopeSearchHintText [11:23] it is [11:23] so i'll make scopeSearchHintText depend on showSearchOnSearchQueryChange [11:23] and then dash_overview on scopeSearchHintText [11:23] and it's all nice and chained [11:23] tsdgeos, easier → just make hint depend on the end no [11:23] w [11:24] tsdgeos, this way you only need to change one MP [11:24] got lost [11:24] make scopeSearchHintText depend on waht? [11:24] scopeActivatePreview ? [11:24] tsdgeos, yes [11:24] ok that too [11:25] but then it's chained after dash_overview forever [11:25] oh well [11:27] Saviq: ok, all pushed and resubmited [11:35] "Silo ready to build packages" :) [11:37] and gcc default is 4.9 now [11:41] mzanetti, does the infographic have any logic to change once the day/month changes? [11:42] mterry: I don't think so... I think I reported a bug for that a while back [11:42] mterry: not sure what happened in the meantime though === dandrader is now known as dandrader|lunch [11:44] tsdgeos, merged, building [11:44] lunch [11:46] mhr3: pete-woods|lunch: Saviq: what do you think of "threshold-index" instead of "collapsed-widgets": 2 ? [11:50] Hola! [11:51] o/ === pete-woods|lunch is now known as pete-woods === alan_g is now known as alan_g|lunch === MacSlow|lunch is now known as MacSlow [12:39] jgdx: yeah sure === alan_g|lunch is now known as alan_g === _salem is now known as salem_ [13:00] jgdx, hey, I saw you took over bug #1219793 [13:00] bug 1219793 in ubuntu-system-settings (Ubuntu) "[reset] reset launcher configuration" [Low,In progress] https://launchpad.net/bugs/1219793 [13:00] jgdx, did you notice that there are some merge requests up for reset which are waiting for review? [13:01] jgdx, e.g https://code.launchpad.net/~laney/ubuntu-system-settings/reset-api/+merge/208661 [13:01] jgdx, you might want to review that/base your work on it [13:02] seb128, hey, looking [13:02] jgdx, it might be slightly orthogonal to the launcher specific case, but would still be nice to get reviewed [13:13] seb128, right. Looks nice. It's of course very relevant to the second reset button. Do you know if there's a bug for that? [13:13] seb128, cause that would be completely dependant on ^ that branch, as well as implementing the reset function in all plugins? [13:13] jgdx, bug for what? [13:13] seb128, "Reset all system settings" [13:13] right [13:15] jgdx, no bug afaik [13:16] there is the bug you pointed and bug #1292932 [13:16] bug 1292932 in ubuntu-system-settings (Ubuntu) "Add "factory-reset" option" [High,Triaged] https://launchpad.net/bugs/1292932 [13:18] Saviq: we have a big problem with qmluitests not picking up all tests that fail [13:18] we really need someone to have a look at that [13:18] otherwise we may as well not have or run the tests [13:30] tsdgeos, right, I'll talk to fginther today [13:30] Saviq: like trunk has 2 tests failing [13:30] and we don't see them [13:30] actually 3 i think [13:31] tsdgeos, it's weird 'cause I run them locally and have no fails... [13:32] Saviq: well xvfbtestListViewWithPageHeaderSection fails for sure [13:32] at least i can reproduce and makes sense since it has leftovers from old headearless code [13:32] proposing a MR for that one soon [13:33] tsdgeos, think it only affects non-qml tests? [13:33] maybe [13:33] it's this and the cardcreator [13:33] tsdgeos, does it write out an .xml file? === dandrader|lunch is now known as dandrader [13:34] sure [13:34] -o /home/tsdgeos_work/phablet/unity8/unity8/builddir/ListViewWithPageHeaderSectionTest.xml,xunitxml [13:35] well, is it there? does it have the fail reported? [13:36] there == where? [13:36] tsdgeos, /home/tsdgeos_work/phablet/unity8/unity8/builddir/ListViewWithPageHeaderSectionTest.xml [13:37] whoa it takes a long time on xvfb... [13:38] well 5 secs for each test [13:38] it's doing tryCompare on something bad [13:41] so fixing it will also make our CI faster :D [13:43] https://code.launchpad.net/~aacid/unity8/fixListViewWithPageHeaderSectionTest/+merge/228679 [13:43] Saviq: what do you think of "threshold-index" instead of "collapsed-widgets": 2 ? [13:44] tsdgeos, not that much better... mhr3 ↑ [13:44] tsdgeos, fwiw testListViewWithPageHeaderSection just hangs here... [13:44] well beter than using collapsed which is consfusing since everything is actaully collapsed [13:44] yea, not really more meaningful [13:45] Saviq: ? [13:45] Saviq: what are you running? [13:45] Saviq: what are you running? [13:45] tsdgeos, /home/michal/dev/canonical/unity8/repo/builddir/tests/plugins/Dash/ListViewWithPageHeaderSectionTestExec [13:45] don't do that [13:45] we have make targets for a reason [13:46] tsdgeos, I started with a target first :P [13:46] not sure if in this specific case it makes a difference or not [13:46] tsdgeos, hanged just the same [13:46] Saviq: are you sure it's not just the tryCompares being slow? [13:46] tsdgeos, nothing happens for a minute... [13:46] worksforme [13:46] :D [13:47] give me one hour so the run of test works [13:47] and i can tell you if the file is there or not [13:47] tsdgeos, so, the qmluitests job doesn't even run that test [13:47] they do [13:47] what you mean they don¡t [13:48] ah n o [13:48] it does [13:50] tsdgeos, oh ok, so it takes so long because it fails all the time, and takes 5s or so to do the tryCompare, right? [13:50] yes [13:50] times 180 [13:51] 15 mins, cool beanz ;P [13:52] yep [13:52] approve the fix ;) [14:19] Saviq: mhr3: pete-woods: when triggering a preview widget inside a collapsible widget, what do i do? [14:19] the ids still need to be unique, so nothing special really [14:20] tsdgeos, yeah, the id of the inner widget needs to be passed [14:21] mhr3: well seems pretty special to me, until now there was just a widget what could be activated [14:21] now you want the inner? the outer? both? [14:21] tsdgeos, why would i care about the collapsible? [14:22] i do not know [14:22] it's just a container [14:22] you care about things i don't always understand ;) [14:22] :) [14:23] ok, so inner widget out [14:23] yea [14:23] out as in, that's the one passed to the scope [14:23] yes [14:23] Saviq: the file is created and yes it contains the errors [14:24] tsdgeos, yeah I saw that, so for some reason it's not picked up as a failure by jenkins [14:28] Saviq: you said I should be able to launch the dash-app with ubuntu-app-launch. But ubuntu-app-launch unity8-dash doesn't work. [14:28] am I doing it wrong? [14:29] mzanetti, can you help elopio ↑? [14:29] sure [14:29] err, actually no... should work :P [14:30] tbh I've never used ubuntu-app-launch [14:30] Saviq: so we are removeing expand/collapse habilities from PreviewTextSummary? [14:30] elopio: but why aren't you using upstart? [14:30] tsdgeos, yes [14:30] Saviq: i.e. if you need expand/collapse you need to use PreviewTextSummary + PreviewExpandable [14:30] ok [14:30] mzanetti: we launch all the apps with ubuntu-app-launch. So that way we could reuse some of the helpers from autopilot. [14:30]