=== _salem is now known as salem_ === salem_ is now known as _salem [09:28] greyback: good morning. I addressed all your comments in the qtmir merge. [09:28] mzanetti: cool thanks [09:29] mzanetti: ah sorry I didn't realize there was a unity8 branch to go with it [09:30] greyback: no worries [09:30] greyback: if you want to test, it should be enough to copy Shell.qml from that unity8 branch to the device [09:30] no need to build a package [09:30] mzanetti: yep [09:37] huh... why does "df -h" not show my partitions any more? [09:38] only virtual file systems in there [11:10] Hola [11:12] Saviq: hey, want to have a look at those so we can get it approved before you're away: https://code.launchpad.net/~unity-team/unity8/two_see_more_bugfix/+merge/234340/comments/572776 [11:12] its really minor things === MacSlow is now known as MacSlow|lunch === _salem is now known as salem_ === MacSlow|lunch is now known as MacSlow [12:44] mzanetti: lifecycle fixes approved [13:07] Saviq, can you have a quick look at https://code.launchpad.net/~macslow/unity-notifications/synchronous-notification/+merge/229060 [13:08] Saviq, I'm confused about that "automatic" merge that just happened [13:09] MacSlow, you pushed to trunk [13:09] MacSlow, https://code.launchpad.net/~unity-api-team/unity-notifications/trunk [13:09] How? [13:09] Crap! [13:11] MacSlow, will you resolve or shall I? [13:11] just doing that [13:29] Saviq, hm... kind of stuck with fixing the accidental merge... [13:30] MacSlow, can I overwrite trunk? [13:30] MacSlow, ugh, you overwrote? [13:31] MacSlow, https://code.launchpad.net/~ps-jenkins/unity-notifications/ubuntu-utopic-proposed should be the good thing [13:31] Saviq, correct [13:32] MacSlow, so yeah, take that branch and overwrite trunk [13:32] MacSlow, and fix your bash history and push locations ;) [13:33] Saviq, yeah... hell knows how that happened... I always explicitly do "bzr push lp:~macslow/bla..." [13:40] Saviq, hm... fixed... but lp:unity-notifications on launchpad did not yet update [13:43] MacSlow, indeed, it will catch up [13:45] phew... lp also updated now [13:48] greyback: thanks! [14:10] Saviq: we need an opinion: void fn(int var) { Q_UNUSED(var); } OR void fn(int) {} [14:12] Saviq, is that part of our word-of-mouth coding guidelines/ http://paste.ubuntu.com/8327396/ [14:12] ? [14:13] dandrader, greyback, Q_UNUSED generally leads to Q_UNUSED being left around when it is, in fact, used [14:14] I don't have a strong opinion but was leaning towards /**/ [14:14] Saviq, you mean later someone actually starts using a parameter but forgets to remove the corresponding Q_UNUSED() line? [14:14] dandrader, yes [14:15] Saviq, greyback, hmm, makes sense [14:20] mzanetti, do you know how to anchor an element of the shell to the OSK? Normally you can ask for that behavior if you have a MainView, but we don't [14:20] dandrader, greyback, I'm not even sure we need the /* */, after all you're only an F4 away from the header [14:21] mterry: Qt.inputMethod.rect or something like this [14:21] mzanetti, not in shell [14:21] yeah, works in apps too [14:21] unless it's back with qtmir? [14:21] oh... in the shell [14:21] mzanetti, Qt.inputMethod didn't work before qtmir, does it work again now? [14:21] Saviq, "F4", are you talking Qt-Creator? [14:21] Saviq: mterry: there is an InputMethod item in Shell.qml [14:21] you can probably anchor to that [14:21] dandrader, yes, but $your_ide's_shortcut works, too ;) [14:22] * dandrader is old school and doesn't use such niceties [14:22] mzanetti, interesting, will try thanks! [14:22] mostly because vim doesn't do it reliably :D [14:23] Saviq, aren't you supposed to be on holiday? [14:23] mterry, starting tomorrow [14:23] Saviq, hrm... OK then, carry on ;) === Malsasa_ is now known as Malsasa [14:32] Saviq: https://code.launchpad.net/~unity-team/unity8/two_see_more_bugfix/+merge/234340/comments/572865 [14:32] mzanetti, yeah, saw that, will have a look [14:33] Saviq: ok... I see (re sync MP) [14:33] Saviq: is silo 11 about to land? [14:34] mzanetti, you tell me ;) [14:34] ah ok [14:34] mzanetti, but no [14:34] I thought you'll land that still today [14:34] mzanetti, there's a bunch of unapproved branches [14:34] ack === salem_ is now known as _salem === dandrader is now known as dandrader|afk === _salem is now known as salem_ [14:48] mzanetti, satorisany idea how to catch individual role changes for model data in qml? [14:48] Saviq: ^ [14:48] * any idea [14:48] dednick: ? [14:49] emit dataChanged(index, index, QVector roles) [14:49] mzanetti: doesnt seem to work with Connections { target: modelData; onXXXChanged: {} } [14:49] ah [14:49] no. that won't do [14:49] hmm [14:49] dednick: you can do this: [14:49] or onModelDataChanged [14:49] property var foobar: model.barbaz [14:49] onFooBarChanged [14:49] mzanetti: yeah. though so [14:49] not totally elegant though [14:49] mzanetti: ok, thanks [14:53] MacSlow, https://code.launchpad.net/~unity-team/unity8/notifications-audio-no-play-empty/+merge/234486 [14:54] Saviq, on it... [14:56] MacSlow, no worries, not pressing [14:56] @unity ↑ important to note; bool("") === false, but bool(QUrl("")) === true [14:56] I'd say a QTBUG really [14:57] wtf... [14:57] I think QUrl("") actually only sets the path [14:58] so there might be a valid authority or protocol string somewhere [15:01] dandrader|afk, mterry, paulliu http://pastebin.ubuntu.com/8327743/ [15:01] mzanetti, think it defaults to file:// ? [15:01] Saviq, it is a bit odd [15:01] Saviq, I mean the need for this kind of fix [15:02] Saviq, sigh... fixed [15:02] Saviq: likely, but I didn't verify [15:02] MacSlow, well, that's a different thing, our media components didn't deal well with playing an empty string (to be fixed) [15:02] MacSlow, but still we shouldn't even call play() on it [15:02] Saviq: actually file:/// to be precise :D [15:03] mzanetti, you're kidding? [15:03] ah no... "" might be 2 / only [15:03] mzanetti, that would be your slash [15:03] mzanetti, meaning your root directory [15:03] :D [15:04] Saviq, btw... approved [15:04] MacSlow, t [15:04] x [15:04] Saviq, jenkis will take a while I guess [15:04] MacSlow, yeah [15:05] Saviq, but since your branch is so fresh it should cleanly merge [15:05] MacSlow, yeah well, let's see if something in silo 11 conflicts [15:11] dandrader|afk: FYI https://code.launchpad.net/~dandrader/unity-api/splash/+merge/233288 fails on CI [15:15] Saviq: hey, do we actually have the gesture cancellation in place yet? [15:16] we don't, right? [15:16] Saviq: ok. removed tags. Thanks. [15:16] Saviq: IMO bugs like these are are misleading because they aren't so much of an issue any more when that is fixed: https://bugs.launchpad.net/ubuntu-ux/+bug/1368702 [15:16] Launchpad bug 1368702 in ubuntu-ui-toolkit (Ubuntu) "[SDK] HEADER ISSUE: Back arrow needs to be moved further to the right and also have a larger invisible touch target" [Undecided,New] [15:17] mzanetti: we don't have gesture cancellation [15:20] mzanetti, no [15:21] mzanetti, we don't have it I mean, and correct, that bug would be less important if we had it === dandrader|afk is now known as dandrader [15:21] Saviq: https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1368702/comments/1 [15:21] Launchpad bug 1368702 in unity8 (Ubuntu) "[SDK] HEADER ISSUE: Back arrow needs to be moved further to the right and also have a larger invisible touch target" [Undecided,New] [15:22] dandrader, do you have the new python version of the strip tags script? [15:22] dandrader, 'cause you seem to have the most of the two bad tags that are still travelling around [15:22] Saviq, I'lll download it again [15:23] dandrader, http://people.canonical.com/~msawicz/unity8/strip-u8-tags.py [15:23] dandrader, and remember you need to strip both local and remote, as tag deletion doesn't travel with commits [15:24] josharenson, please strip tags on lp:~josharenson/unity8/fix_panel_minute_count and any local checkouts you might have [15:24] josharenson, http://people.canonical.com/~msawicz/unity8/strip-u8-tags.py [15:24] saviq, ack... [15:24] has that script been updated recently? [15:25] Saviq, could you name a branch that has bad tags? [15:25] Saviq, I'm running it and it's not printing anything [15:26] Saviq, ah, finally got some tags in lp:~dandrader/unity8/real_header_splashscreen [15:26] s/got/found [15:29] Saviq, ahhhh! strip-u8-tags.py doesn't run on the current dir by default. you have to specify "strip-u8-tags.py ." [15:29] dandrader, http://pastebin.ubuntu.com/8327743/ [15:29] Saviq, now it's explained why I wans't finding anything locally [15:29] dandrader, oops, sorry, should probably make it work like that :) [15:31] Saviq, or print out an error like "please specify a directory" [15:31] dandrader, yeah, you know... that script was only supposed to live for a week or so :P [15:31] then it got rewritten in py, and was supposed to only live another week... [15:32] Saviq, all my branches should be clean now [15:33] dandrader, thanks [15:33] greyback, ready when you are === salem_ is now known as _salem [15:34] josharenson: https://plus.google.com/hangouts/_/canonical.com/gerry [15:34] josharenson, no, as long as you had the .py version [15:40] greyback, oh, right. should be fixed now (let's see what jenkins says next) [15:49] greyback, how do I find out where the OSK is on screen (from the shell code) -- we used to have OSKController, now we have InputMethod, but it doesn't seem to give info on where on the screen is actually being drawn [15:59] dandrader: thanks [16:00] mterry: hmm. OSK is actually a maximized surface like any other (non-fullscreen) app surface, with lots of transparency. Shell doesn't know from that how big the OSK is [16:01] greyback, that's why we have the keyboard communicator socket? [16:01] mterry: if InputMethod doesn't tell you, I'm not sure there's any other place to get that info right now [16:01] or... had... [16:01] Saviq: yep [16:01] greyback, I looked into how anchorToKeyboard works and it's doing "Qt.inputMethod.visible ? Qt.inputMethod.keyboardRectangle.height : 0" [16:01] greyback, do we have access to ↑ now? [16:01] greyback, does that seem likely to work? [16:01] Didn't in my quick test, but I'm probably doing something wrong, since that's what the SDK uses [16:02] mterry: yeah that works for apps. I don't see any reason why it would not work for unity8 too, there may be somethinkg missing in the QPA for it [16:02] greyback, it didn't work before qtmir because shell was never focused [16:02] mterry: if it's what the SDK uses, I'm sure we can make it work for shell [16:02] and it *did* work before that because Qt didn't care [16:02] and then it started caring [16:03] Saviq: now the unity8 "Window" should be focused, or at least we can persuade Qt that it is [16:03] greyback, so you're saying there may be qtmir work for this? [16:03] greyback, yeah, that'd be grand [16:03] mterry: a little yeah [16:04] greyback, OK. I think I'll go with a workaround for now, then file a qtmir bug about it for proper solution [16:05] mterry: thanks === dandrader_ is now known as dandrader|lunch === _salem is now known as salem_ [19:32] greyback, btw that Qt.inputMethod.keyboardRectangle.height trick did work after all -- forgot to get back to you on that [19:32] mterry: it works? Oh great [19:32] * greyback draws satisfying line through task in his TODO list === dandrader_ is now known as dandrader [20:22] greyback, what would be the best way to refer to an image in /usr/lib/$ARCH/qt5/qml/Ubuntu/Components/Themes/Ambiance/artwork/ [20:23] greyback, ^^^ that's about your comment https://code.launchpad.net/~dandrader/unity8/real_header_splashscreen/+merge/233286/comments/571898 [20:24] from unity8 [20:37] dandrader: nothing easy springs to mind. But it's late here, so I'm not thinking hard right now [20:38] greyback, I think the only way would be to have a wrapper component there in Ubuntu/Components/Themes/Ambiance so unity8 could import it [20:38] like just an Image with a preset source [20:39] greyback, so easier to copy over like Florian did [20:39] dandrader: could this be a component in the SDK, instead of copying all this stuff over? A kinda private component [20:41] greyback, are you talking about PageHeadStyle and MainViewStyle? [20:41] yep [20:41] greyback, they are :D [20:41] greyback, but for the splash screen we needed modified versions of them [20:41] greyback, so they were copied over and modified [20:42] greyback, I believe mainly so that it's not tied to the theme of the qml scene but to a hardcoded one [20:42] greyback, florian left some todo's there about how it should be ideally in the future [20:43] greyback, but the way the sdk is implemented at the moment does not support such flexibility [20:43] greyback, thus the copy-over-and-modify approach [20:44] dandrader: is a pity. I just forsee design changes in SDK happening that break this trick [20:44] greyback, error prone as it will have to be updated every time their sdk counterparts change, but it-just-works(Tm) [20:44] greyback, ditto [20:54] dandrader: I'm not going to block this. I just dislike copying the images, so would like to explore all options before just copying === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === PyAr is now known as alecu === zoktar_ is now known as zoktar === irsol_ is now known as irsol === zoktar is now known as 77CAAFYCY === salem_` is now known as _salem === jamesh is now known as 6JTAAFV9I === zbenjamin is now known as 6JTAAFWR0 === hggdh is now known as 6JTAAFWBI