=== fginther is now known as fginther|away === fginther|away is now known as fginther [07:21] dednick, hey, could you review https://code.launchpad.net/~unity-team/unity8/caching-image-provider/+merge/224415 please [07:23] Saviq: sure [07:37] dednick, thanks [07:42] moin [07:44] o/ [07:44] dednick, btw, https://code.launchpad.net/~nick-dedekind/unity8/move-indicator-qml/+merge/223947/comments/539798 [07:44] dednick, the Dash one is actually a crash, probably same as testShell [07:44] mzanetti, could you review https://code.launchpad.net/~mterry/unity8/is-active/+merge/223653 on this fine mornin' please? [07:45] ok [07:50] ohnoes [07:50] mhr3, you landed unity-api without corresponding changes to unity8 and -scopes-shell did ya? [07:52] Saviq, right, but it's compatible [07:52] mhr3, sure, but FTBFS :P [07:52] Saviq, you can bump u8 when you implement support for customizations [07:52] Saviq, what? [07:52] mhr3, pure virtual [07:52] oh.. mocks [07:52] yup [07:52] right... sorry totally forgot about that [07:53] Saviq, do you want me to prep an mp? [07:53] mhr3, doing, you'll review [07:53] ok [07:54] truth be told mocks should live in unity-api [08:01] mhr3, https://code.launchpad.net/~saviq/unity8/mock-customizations/+merge/224757 [08:02] brb [08:06] is unity-mir Unity.Application 0.1, right? [08:06] Cimi: yes [08:06] this Unity.Application comes from unity-mir [08:06] ok [08:06] so weird [08:07] why? [08:07] I need gerry :) [08:08] https://bugs.launchpad.net/ubuntu-welcome-wizard/+bug/1334429 [08:08] Ubuntu bug 1334429 in Ubuntu Welcome Wizard "touch event allowed through OSK to what is underneath" [High,New] [08:08] but we're using OSK from unity mir [08:08] so I don't know why... [08:08] Cimi: I think we have an InputFilterArea in unity [08:09] ah nio [08:09] nevermind [08:09] missed point 3 [08:18] Cimi: should this probably be reassigned to the welcome wizard? https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1334203 [08:18] Ubuntu bug 1334203 in ubuntu-system-settings (Ubuntu) "ubuntu-system-settings-wizard crashes/freezes when using the back button" [Undecided,New] [08:19] mzanetti, do you have a stacktrace? [08:19] doesn't crash for me [08:19] Cimi: its 100% reproducable [08:19] Cimi: did you follow the instructions in the bug precisely? [08:19] step 1 mostly [08:19] I just saw it [08:20] will bootstrap [08:23] mzanetti, if you could share that crash log [08:24] mzanetti, probably not a bug in the wizard, but a broken plugin [08:26] Cimi, could you please review https://code.launchpad.net/~saviq/unity8/fix-card-vjournal/+merge/224727 quickly [08:26] Cimi, I'd like to land this fix with the FTBFS fix [08:26] Saviq, jumped the gun too soon on the ftbfs fix, should bump deps in control [08:27] mhr3, mhm [08:27] need to reboot [08:27] bloody usb not working [08:27] http://paste.ubuntu.com/7710021/ [08:27] Saviq, ^ [08:27] mzanetti, ^ [08:28] this is my dmesg :( [08:28] Cimi, yup [08:28] Cimi, reboot [08:28] mhr3, pushed [08:29] Saviq, yep, +1 [08:29] Saviq, ok works, what was that? [08:30] Cimi: did you reboot the machine or the tablet? [08:30] mzanetti, pc [08:30] really weird... [08:31] Cimi, https://code.launchpad.net/~saviq/unity8/fix-card-vjournal/+merge/224727 [08:31] Saviq, not that [08:31] Saviq, the reason reboot is required [08:31] Cimi, dunno, something in Ubuntu usb gets confused [08:31] Cimi, adb/mtp/blah [08:31] ah ok [08:32] Cimi, to test the MP easily: http://paste.ubuntu.com/7710041/ [08:32] Cimi, and then in tryDash you'll see in trunk 3 columns on top of another [08:32] ok [08:32] with the fix - 3 separate columns [08:32] complete review fail on my part :| [08:33] Cimi, if you have doubts about width/height, Loader assumes the loaded item's dimensions, or forces its own if are explicit [08:33] Cimi, so Loader { width: foo; sourceComponent: blah } [08:34] \i see [08:34] Cimi, basically means blah will be foo-wide, and loader will be blah-wide [08:34] Cimi, FFR http://qt-project.org/doc/qt-5/qml-qtquick-loader.html#loader-sizing-behavior [08:39] Saviq, subtitle is cut at the bottom [08:39] Cimi, you ACKed this fix, remember https://code.launchpad.net/~saviq/unity8/fix-header-height/+merge/224269 ? [08:39] ;) [08:39] cool [08:39] ok [08:39] well fix works [08:40] Cimi, yeah, needs tests, I know [08:40] Cimi, but let's just get it in, fix is obvious [08:44] mzanetti, can I bootstrap without reflashing? [08:45] Cimi: I don't know... probably you can wipe ~ and /userdata, but I never really tried it [08:45] dednick, replied on the mp [08:51] @unity please grap an MP to review (and claim it in the MP page when doing so), we're drowning here [08:52] Saviq, I have few that can be grabbed :P [08:52] when you have some time / strength to do so, of course [08:52] Cimi, yup, works both ways [08:54] dednick, btw, how do you find the reworked QmlPlugin.cmake? any comments, other than making it support folder hierarchies? [08:58] mzanetti, yeah i can reproduce [08:58] ok [08:58] mzanetti, problem I need to install dbg symbols [08:58] but I have no wifi :) [08:58] :) [08:59] and no rw filesystem [08:59] you'll figure a solution, I'm sure [08:59] mzanetti, I'll remove the back button :) [08:59] Cimi: so the ro file system is not a problem [08:59] Cimi: I also had this when the wizard landed [08:59] and I didn't bootstrap and had a rw image [08:59] Saviq, was there ever solution to :native not working on jenkins? [09:00] Cimi: I guess cleaning up every data that the welcome wizard produces would be enough [09:01] mhr3, I believe they fixed it, yes [09:01] mzanetti, it produces no data [09:01] mhr3, bug #1322294 remains, though [09:01] bug 1322294 in launchpad-buildd "Recipes don't support :native multi-arch dependency" [High,Triaged] https://launchpad.net/bugs/1322294 [09:01] Saviq, not sure about that https://jenkins.qa.ubuntu.com/job/unity-team-unity-scope-click-devel-utopic-amd64-ci/129/console [09:02] mhr3, they did just strip :native before, maybe they reverted this [09:02] right... "fixed" [09:02] mhr3, or maybe they didn't "populate" all the jenkins [09:03] mhr3, well, until they start x-compiling, everything's native ;) [09:03] Saviq, anyway, i guess fginther will know [09:03] mhr3, I *think* it was psivaa who knew about it before, too, so try with him [09:04] Saviq, k, thx [09:09] dednick, reviewed u-s-c menu.plugin [09:17] Saviq: you mean the QmlPlugin.cmake in the suru-switch branch? [09:17] dednick, hm? no, the one you copied to menu.plugin [09:18] dednick, "new" as in the one we have in unity8 for some time [09:18] dednick, was just wondering whether we should put it in cmake-extras already or is there still work on it needed [09:20] Saviq: sorry, getting my branches mixed up. i meant https://code.launchpad.net/~unity-team/unity8/manual-qmltypes/+merge/224578 [09:20] which is the one i copied from [09:20] dednick, well, yeah [09:20] dednick, that one [09:21] but yeah, it's good for usc. should put it in cmake-extras [09:21] didnt have to mod [09:25] Saviq: well, we have that commit hook... its just that noone seems to be using it [09:25] mzanetti, yeah exactly... [09:25] well, I use it all the time... [09:25] and have also updated it here to check for stale tags [09:26] mzanetti, well, also, commit is probably too often [09:26] works fine for me... its less than 1 sec [09:26] mzanetti, yeah, but you might want to split things into several commits [09:26] mzanetti, and then the hook will tell you NO [09:26] the hook doesn't disallwo committing... it just tells you: this will fail CI if you continue [09:27] and keeps a copy of the commit message so you don't have to type it again in case you decide to fix it and recommit [09:28] mhm [09:28] mzanetti, hmm where is it? [09:28] .bazaar/plugins [09:31] mzanetti, I think we need a script in . [09:31] mzanetti, that will do the checks [09:31] mzanetti, and the plugin (if installed) will call it [09:32] mzanetti, this way you can run it without the hook [09:32] yep, works for me [09:32] like a pre_push.py [09:33] also, I'd like to have the hook more visible/easy to install [09:34] but haven't gotten a good idea for that yet [09:34] I'm sure more people would use it if they'd be aware of it [09:34] mzanetti, this script in . could check for the hook [09:34] mzanetti, and say "if you want this to run on commit, do foo" [09:34] yeah, sounds reasonable [09:34] mzanetti, no point for it being in .bazaar, either [09:35] mzanetti, we can move it to tools/ [09:35] right... [09:35] since you can't have a branch-local plugin [09:35] yeah, too bad... [09:35] that'd be useful [09:36] ooh guess what, git does! ;P [09:36] you can? [09:36] ah [09:36] I read "it does", missed the g [09:36] of course git does [09:36] ;) [09:39] Saviq: btw. when you touch this, if you can think of a better to figure the branch, one that doesn't break when we move shell.qml to another subdir :D [09:40] mzanetti, that's the thing [09:40] mzanetti, the hook will just check for a script in . [09:40] mzanetti, if there - run it, otherwise ignore [09:40] mzanetti, script with a known name [09:40] ok... it'll break when you move the script to another dir... but I guess already better than some "random" qml file of the codebase [09:41] mzanetti, yeah, and we won't move the script to another dir, since we want it to be visible [09:41] mzanetti, and available at your fingertips [09:41] ack [09:48] Saviq: maaan... that new-header branch develops to a beast... [09:49] mzanetti, story of my life ;P [09:49] figured that we actually could get rid of the PageHeader::childItem and put the uitk's header in there... [09:49] then there's already a back button [09:49] which doesn't match design [09:49] mzanetti, indeed [09:49] then it has like a thousand states I mustn't break :D [09:50] mzanetti, you probably could [09:50] mzanetti, since the states are meant to deal with the search entry [09:51] mzanetti, which with the new header should probably behave differently [09:51] hmm... should it? [09:51] mzanetti, well, ultimately we should take the behaviour from UITK [09:51] mzanetti, but since it's not there yet... [09:52] mzanetti, I think we should get the UITK header as is and replace its contents with {text,image}, text entry, just sliding up/down on search activation [09:52] mzanetti, since the search icon doesn't make sense when the search entry is on screen [09:52] mzanetti, and vice-versa [09:53] Saviq: right... that's what it does now... but it hides the back button, which is a probme [09:53] problem [09:53] mzanetti, right, because it replaces the whole header from UITK [09:53] so I could obviously just leftMargin the textfield [09:53] but I fugred that wouldn't pass your review [09:53] mzanetti, I mean it should be UITKHeader { contentItem: OurComponent { } } [09:54] mzanetti, where OurComponent would be {text|image} + text entry [09:54] I don't think the uitk's header has such a thing [09:54] it does [09:54] * mzanetti checks again [09:54] it might be __ [09:54] but will be public later [09:55] I must be blind [09:55] * Saviq looks [09:56] ah... there's something [09:56] contentsContainer might do that [09:56] yeah [09:57] mzanetti, oh well, PageHeader still needs to wrap it [09:57] mzanetti, because we need to add departments to it [09:57] true [09:58] also the landscape behavior would break [09:59] Saviq: thanks for review. pushed update. [09:59] mzanetti, well, that's fine, we don't want to bring the text entry in in landscape any more [09:59] mzanetti, only on search press [10:02] dednick, kk [10:05] dednick, hmm, I'd have left find_package(Qt5Core REQUIRED), otherwise you have no qt5_use_modules [10:06] dednick, in your case it basically gets included somewhere else, but I think it's good to have the one for Core [10:07] Saviq: it's in the root cmake [10:08] Saviq: but mk [10:08] dednick, yeah, I know, probably depends on how redistributable we want it to be [10:08] dednick, hmm, why LGPL? [10:08] dednick, you don't need LGPL for dlopening [10:08] Saviq: because i don't really know which one to use.... [10:09] and the other files around seemed to use it [10:09] dednick, you only need LGPL for libs that are linked against [10:09] Saviq: does it matter? [10:09] dednick, IANAL ;) [10:09] it is lesser for some reason :) [10:09] :) [10:09] dednick, we default to GPLv3 [10:10] dednick, fall back to LGPLv3 when required [10:11] dednick, so please make it GPL, if others are already LGPL, let's fix in another branch [10:11] Saviq: doing [10:11] dednick, also, check what's in debian/copyright [10:12] dednick, hmm it's LGPL there [10:13] dednick, sorry, let's leave it be then [10:13] hehe [10:13] dednick, and fix it across the board after verifying that's needed [10:14] Saviq: ok. pushed the fin_package change [10:14] dednick, can't top-ack [10:15] Cimi, add me to https://launchpad.net/~ubuntu-settings-components-team please [10:15] Saviq: need packaging approval? [10:15] Saviq, done [10:15] dednick, no, just not on the team ;) [10:15] since i removed the -assets package [10:15] better [10:16] dednick, we'll get pkg review in silo [10:16] Saviq: ok. thanks [10:16] dednick, unless you find someone who will take a look ;) [10:16] /fiid [10:16] /food that is [10:18] mzanetti, know what? just do the simplest transition possible for now, we'll get a proper one from UITK when it's there anyway [10:19] ack. on it :) [10:26] mzanetti, thinking here... [10:27] don't :D [10:27] I *know*, sorry [10:27] I'm nearly done [10:27] mzanetti, we have basically two header configurations → [back], {text|image}, favourite, search [10:28] back, text_entry [10:28] yep [10:28] mzanetti, and if we just replace the actions in the current UITK header [10:28] mzanetti, they will just appear/disappear as they do now [10:28] erm [10:28] I mean because there's no transitions in the UITK [10:29] mzanetti, let's use two UITK headers and swap them (slide them in/out)? [10:29] phew... :) [10:29] that's what I'm doing [10:29] good [10:29] or well, not really [10:29] I just attached a back button myself to the textfield [10:30] so for now it keeps behavior as it was... hope that's fine [10:30] well, you'll see in a minute [10:30] will just have to update the preview header [10:30] mzanetti only problem we'll end up with the search entry *and* the search button [10:30] no [10:30] in landscape? [10:30] oh... right [10:31] mzanetti, I think what I'm asking will simplify this quite a bit [10:31] mzanetti, like we get rid of the Narrow states in PageHeader [10:32] mzanetti, anyway... I know you'll get there [10:33] * Saviq focuses on the cereal, or it will be keyboard food [10:40] how do I write http://paste.ubuntu.com/7710506/ so that returns 0 or null when unitymenumodel is not valid? [10:45] Cimi, you probably want dednick for this [10:45] mzanetti, I think that is the crasher [10:46] fix it :P [10:46] mzanetti, model being destroyed but property being manipulated [10:46] * mzanetti rewrites PageHeader.qml :/ [10:47] Saviq: so in widescreen, we behave the same as in portrait, ok? [10:48] having a super long textfield [10:48] mzanetti, no, make it 60GU max, falling back to 40GU in > 60GU screens [10:48] mzanetti, same behavior as we have with other width-dependant things [10:49] Saviq: and the title would then stay? [10:49] mzanetti, no [10:49] mzanetti, replace the title with 40GU wide text entry [10:49] meaning empty space at the left of the text entry [10:49] right [10:49] mhm... ok [10:49] at the right [10:49] I mean [10:50] oh... moving the label to left [10:50] err. the textfield [10:50] | < [text entry] | [10:50] right [10:50] ok [10:50] | < Title f s | [10:51] two configs like that [10:51] ack [10:51] that sounds like a looong friday :D [10:52] mzanetti, I really don't think it should be that complex... [10:52] more red than green, at least [10:53] yep... its not too bad actually [10:57] dednick, ping :) === MacSlow is now known as MacSlow|lunch [12:04] Cimi: sup? === alan_g is now known as alan_g|lunch [12:16] dednick, I have a crash in the wizard [12:16] dednick, happens with http://paste.ubuntu.com/7710506/ [12:17] dednick, if I am in the wifi page then I tap "back" [12:17] dednick, it might be that unitymenumodel is deleted or something like that [12:25] Cimi: could be related to: https://code.launchpad.net/~charlesk/qmenumodel/lp-1282282-dangling-UnityMenuAction-pointer/+merge/211211 [12:33] can we trigger https://code.launchpad.net/~charlesk/qmenumodel/lp-1282282-dangling-UnityMenuAction-pointer/+merge/211211 ? [12:37] Cimi, wdym trigger? [12:37] Saviq, ci [12:37] Saviq, cannot sbuild [12:38] https://code.launchpad.net/~charlesk/qmenumodel/lp-1282282-dangling-UnityMenuAction-pointer/+merge/211211/comments/497978 [12:39] Cimi, done [12:39] Cimi, it was rather old indeed [12:45] xnox, hey, could you please look at qmenumodel's x-compiling? it's definitely missing :any for python, but not sure -gi and -dbus should also get :any? === alan_g|lunch is now known as alan_g === _salem is now known as salem_ [13:06] Cimi, reviewed infog [13:11] oh wow, suggestions in gdocs [13:12] handy [13:24] greyback, mzanetti should we keep using this doc to track qt comp bugs? https://docs.google.com/a/canonical.com/spreadsheets/d/1Balt7jlHQlNuf7RBuXYjnYJWaICd_wTRpSVfTYYb3cQ/edit#gid=0 [13:24] I think we should, so I will update it with the latest problems caused by qt 5.3 and latest mir/devel [13:24] dandrader: ok with me [13:24] greyback: mzanetti: dandrader: do you guys have any idea when actual N7 landscape support might land? [13:25] cwayne: we plan within 2 weeks from now [13:26] greyback: wonderful, thanks [13:26] dandrader: PPA in working order, patch for input crash added to Mir [13:26] cwayne, we had it but qt 5.3 caused a regression on that [13:26] greyback, great! [13:27] Saviq: in order to have a flake8 make build task, we would have to add it to build-deps, right? [13:27] dandrader: I want to clear the fixed bugs from that spreadsheet [13:27] dah, thanks a lot qt5.3 [13:27] greyback, yeah. they are just polluting the view [13:30] elopio, yup [13:30] elopio, but since I wrote that [13:30] elopio, we discussed a pre-commit check / hook that we'd add this to instead [13:30] elopio, so let's leave it be for now [13:31] Saviq: ok. We had a problem on the toolkit adding flake8 as a build-dep because it's not on main. [13:31] a pre-commit check sounds good. [13:31] elopio, k [13:32] Saviq: the branch is ready for review again, btw. Thanks for your comments. [13:32] elopio, tx! [13:42] Saviq, yes I copied, because I expect UITK delay things [13:43] Cimi, that's not a solution I'm afraid [13:43] Saviq, I can put the changes when we agree on API [13:43] Cimi, if we copy it now, we'll be stuck with it forever, you know that? ;) [13:43] Saviq, why? [13:43] Cimi, because there will be no pressure to fix [13:43] Saviq, well fix comes from our side anyway [13:45] Saviq: I can try to make the hook. I'll take a look at the existing one. [13:45] elopio, we want to change stuff a little [13:45] elopio, make a script in . [13:45] elopio, that we can run manually [13:45] elopio, so that it's visible [13:46] elopio, and the hook would just check for that file and execute if there, ignore otherwise [13:46] elopio, this way it can be used without installing the hook [13:46] elopio, and hook could be reused [13:46] if other projects include such a script in their root [13:47] Cimi, UITK has been very receiving recently - they've opened v1.1 of the API [13:47] Cimi, so please prep an MP to include the different fade modes [13:48] Cimi, see https://code.launchpad.net/~saviq/ubuntu-ui-toolkit/add-icon-source-property/+merge/224122 for how they want the new APIs put in === MacSlow|lunch is now known as MacSlow [14:10] Saviq, how shall we call that API? [14:10] Cimi, mode / fadeMode [14:10] Saviq, and the fade modes? [14:10] Cimi, overlay, cross [14:18] Saviq, did you mean to abstain on the platform-api MP? [14:25] Saviq, https://code.launchpad.net/~cimi/unity8/card_emblems_single_container/+merge/223899 merged === alan_g is now known as alan_g|tea [14:33] Are you speaking on the mumble? [14:34] paulliu, yes [14:34] paulliu, still talking [14:35] Cimi: yes. Thanks. Now it is ok. === alan_g|tea is now known as alan_g [14:40] dednick, unfortunately doesn't fix it [14:41] Cimi: you'll need to get me a crashdump in that case === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === gatox is now known as gatox_lunch [17:01] dandrader, does lp:unity-system-compositor/devel-mir-next replace lp:~mir-team/unity-system-compositor/development-branch [17:01] kgunn, ^^^ [17:02] fginther, I would say so [17:02] dandrader, ok, I'll go with that === alan_g is now known as alan_g|EOW [17:12] mhr3: ping [17:12] mhall119, pong [17:13] mhr3: cwayne pinged me about https://bugs.launchpad.net/ubuntudeveloperportal/+bug/1328559 [17:13] Ubuntu bug 1328559 in Ubuntu App Developer site "14.04 scopes api docs use missing framework "ubuntu-sdk-14.04-clibs"" [Undecided,Confirmed] [17:13] the example click manifest on there uses "framework": "ubuntu-sdk-14.04-clibs", [17:13] which isn't actually a framework on the phone [17:14] I'm also not sure if the apparmor templates "ubuntu-scope-local-content" and "ubuntu-scope-network" exist [17:15] mhall119, hm, we were updating that recently, should be fixed in latest docs [17:15] latest docs package in utopic? [17:20]