=== _salem is now known as salem_ === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem === marcusto_ is now known as marcustomlinson_ [07:45] Saviq, is the greeter unsplit in image 84? [07:45] Cimi, yes, even before that (81 I think) [07:45] Saviq, ok [07:45] Saviq, cause I tried autopilot for my branch but greeter is transparent and such [07:45] all failing [07:46] so could be something else [07:46] Cimi, yeah, but it doesn't in trunk ;) [07:46] ahah yes [07:48] weather sucks in this country [08:07] tsdgeos, can you look at the dep issues? [08:07] tsdgeos, thomas wants to land it today no matter what [08:27] Saviq: mhr3: most of the times if run unity8 inside valgrind i don't get the scopes contents, any idea why that may be? [08:28] tsdgeos, timeout probably [08:28] pfft [08:46] bah [08:55] tsd [08:55] eh [08:56] tsdgeos, as Saviq said, timeouts [08:57] mhr3: any easy way to increase them? [08:57] is recompiling scopes lib an option? [08:57] sure [08:58] although... should be possible to do it just by dropping a conf file somewhere [08:58] but i never tried that [08:58] so let's recompile [08:58] mhr3: well i think i don't need it now [08:58] i can get it to work mostly with unity-scope-tool [08:59] it'd be good to know for the future [08:59] need to go, i'm at the doctors waiting room and my turn is next [08:59] bbl [08:59] include/unity/scopes/internal/DfltConfig.h.in [08:59] static const int DFLT_ZMQ_TWOWAY_TIMEOUT = 300; // milliseconds [08:59] static const int DFLT_ZMQ_LOCATE_TIMEOUT = 3000; // milliseconds [09:00] eh.. [09:00] lol [09:05] Saviq, hm, noticed that noone did a proper review of the u8 departments branch, could you look at it? [09:05] mhr3, will do [09:06] mhr3, you found an issue though? with temp scopes? [09:06] yea.. wish that was the only issue :/ [09:06] :| [09:07] Saviq, either run on device or my branch are quite broken [09:07] at least on the phone :\ [09:07] Cimi, easy to verify - try with trunk [09:07] I haven't touched greeter in such way to break it [09:08] anyway will do [09:09] Saviq, can we have a silo? [09:12] ah I see [09:12] Saviq, I have a mousearea on top of the infographics image [09:12] Saviq, which steals events from the swipe to unlock [09:12] Saviq, I just want the double tap [09:13] is it possible to propagate everything apart double taps? [09:13] or shall I connect to the main greeter mousearea and capture taps from here? [09:14] Cimi, double tap is unfortunately just two taps, so not sure how you think you'd discern those... [09:14] Cimi, so yeah, use the same area [09:15] ok [09:15] that is the draggingarea in greeter.qml.. [09:27] Saviq, hey, do you know if that's a known issue? http://paste.ubuntu.com/7657512/ [09:27] Saviq, robert_ancell is hitting that when trying to run unity8 desktop on current utopic [09:28] likely a platform-api issue, but tvoss doesn't seem online [09:28] hum, ricmm is not on this channel, -> touch === dandrader is now known as dandrader|afk [09:51] Saviq, issue is... draghandle has already onClicked [09:51] Saviq, sure we shouldn't have another mousearea on top of the infographics stealing that clicked event? [09:52] otherwise I have to hack the draggingarea [09:52] and check the mouse position before dealing with onclicked to tease the launcher === dandrader|afk is now known as dandrader [09:55] Cimi, the greeter hints on tap [09:56] Saviq, onClicked [09:56] Cimi, it's working fine in the old infographics, why isn't it working now? [09:58] Saviq, because new infographics have a mousearea [09:58] Cimi, the old ones had one, too, double-click to change the source [09:58] Saviq, does tap work? [09:59] Cimi, yes [09:59] on the circle, to reveal? [09:59] Cimi, I tap once, the hint shows up [09:59] Cimi, I tap twice, the notification source changes [09:59] mouse.accepted = false [09:59] ok [10:00] yeah works with this added code [10:00] I guess I should make tests more robusts for that [10:00] apart from that is fine [10:03] Saviq, looks like it is running with my fake mock on the phone [10:04] why is that? [10:04] cannot se a proper prefix [10:06] ah ok calls run.sh === vrruiz_ is now known as rvr [10:13] tsdgeos, http://paste.ubuntu.com/7657681/ [10:14] mhr3: cool, tx === MacSlow is now known as MacSlow|lunch [10:30] Mirv: ping [10:30] tsdgeos: pong [10:30] Mirv: i have a tentative fix for the unity8 crash on scorlling [10:31] Mirv: I have an upstream patch yet though (patch was exchanged with Simon from Digia via pastebin) [10:31] do you want to try to integrate it so more people can try it and confirm the crash is gone? [10:31] tsdgeos: ooh! must be hard debugging, seeing random corruptions at different places [10:31] tsdgeos: yeah sure I'd like to integrate it [10:32] let me give you a diff [10:36] Saviq, does unity automatically prefix mocks? [10:36] preload I mean [10:36] Mirv: http://paste.ubuntu.com/7657771/ [10:36] Saviq, I always have the fake infographics, even id I almost commented run and run_on_device [10:36] i think it might be because I don't have the real plugin [10:36] so it fallsback to the fake [10:40] tsdgeos: thanks, pushing a build to the normal PPA (the best way to have actual test results) [10:40] I hope the gcc breakage doesn't affect compiling of qtdeclarative though [10:41] gcc breakage? [10:42] tsdgeos: if you upgrade your phone now and get 4.9.0-7ubuntu1 version of gcc-4.9 packages, everything C++ stops working [10:42] but aren't we compiling with 4.8 ? [10:42] but I guess the default compiler is still 4.8. I don't really know the details except that all you get to see is Google logo if you upgrade :) [10:43] tsdgeos: yes I think so, and I think that probably the actual breakage comes from libprocess-cpp being compiled with 4.9 and now the new 4.9 broke it [10:43] :/ [10:44] greyback: using mir+unity-mir devel the apps are not getting focused when opened. [10:44] so I put out http://people.canonical.com/~tjyrinki/gcc/fix_gcc.tar if eg. people want to dist-upgrade to Qt 5.3 PPA and still have it functional (relatively easily) [10:58] greyback: nevermind. was my dodgey code [10:59] dednick: sorry am otp === dandrader is now known as dandrader|afk [11:21] Mirv: upstream is now at https://codereview.qt-project.org/87633 === dandrader|afk is now known as dandrader [11:36] Saviq, time for mumble quickly? have issue I cannot resolve [11:47] Cimi, am there [11:52] tsdgeos: compliation essentially done but the nice PNG crunch will take some time still https://launchpad.net/~ci-train-ppa-service/+archive/landing-005/+build/6105619 [11:53] ok [11:53] will keep an eye [11:59] Mirv, that thing should be killed with fire :P === alan_g is now known as alan_g|lunch === dandrader is now known as dandrader|afk === davidcalle_ is now known as davidcalle === MacSlow|lunch is now known as MacSlow === _salem is now known as salem_ [12:37] paulliu, now that you know the dash machinery more, could you have a look at bug #1330957 ? [12:37] bug 1330957 in Unity 8 "Header with no mascot has no margin when card background is enabled" [High,Triaged] https://launchpad.net/bugs/1330957 === alan_g|lunch is now known as alan_g === boiko_ is now known as boiko [13:11] Mirv: so the build is done, can you do some testing on whether you can reproduce the crash or not? [13:11] tsdgeos: yeah, testing already, I haven't been able to crash it so far [13:12] nice, thanks [13:30] Saviq, same issue in the test with trunk [13:31] Cimi, so probably the test does something, like tap too quickly or so [13:33] Saviq, so basically adding mouseDoubleClick(greeter, greeter.width/2, greeter.height/2); at the beginning in test_teasingArea(data) of tst_SingleGreeter.qml causes the following to fail [13:34] so a double tap/click on the infographics mouse area break things [13:34] I don't know what [13:34] note that if you then add another mouseClick after this, it works [13:34] it's like something gets stuck [13:39] Saviq, ^ [13:39] Saviq: the fact i need "afterLineChildItem: DashDepartments" in two places in https://code.launchpad.net/~aacid/unity8/departments/+merge/221072 shows we need some refactoring there, but i'd rather add a todo than do it right now, what do you say? [13:40] tsdgeos, call it "navigation" btw... [13:40] Saviq: what should i call navgation? [13:40] tsdgeos, "afterLine" [13:41] tsdgeos, they refer to it as navigation (departments + sorting) [13:41] Saviq: well, it's a generic name for the thing that shows in a pageheader [13:41] you can you there whatever you want [13:41] tsdgeos, hmm ok [13:41] tsdgeos, bottomItem? [13:41] tsdgeos, the name is bad :D [13:41] i agree it's not the best name ever, but navigation is worse [13:41] tsdgeos, right, forgot it's generic [13:42] tsdgeos, but yeah, we probably need some refactor between ScopeItem and Dash, am fine with TODO === dandrader|afk is now known as dandrader [13:53] Saviq: so you prefer bottomItem instead of afterLineItem ? [13:53] afterLine is about being after the "signature line" [13:53] tsdgeos, I know... postDividerItem? [13:53] tsdgeos, it's all in PageHeader, too, right? not specific to LVWPH? [13:53] correct [13:59] tsdgeos, so bottomItem would be good, too, your call between postDividerItem and bottomItem [14:00] ok === alan_g is now known as alan_g|tea === alan_g|tea is now known as alan_g [14:43] Saviq: ok. [14:44] Saviq, I'm pretty confident https://code.launchpad.net/~macslow/unity8/sound-hint-support/+merge/218962 and https://code.launchpad.net/~macslow/unity-notifications/sound-hint-support/+merge/218961 are good to go, if you want to take another look... regarding the two branches for the combo-button support there's still a merge-conflict (reported by lp) in debian/changelog I haven't been able to reproduce locally. [14:46] MacSlow, which branch has the conflicts? [14:47] Saviq, this one https://code.launchpad.net/~macslow/unity-notifications/combo-button-support/+merge/221506 [14:52] MacSlow, indeed, ignore LP, must've gotten confused [14:52] MacSlow, if CI is good then we're fine [14:53] Saviq, newer CI-run for unity8/sound-hint-support is currently underway... I expect a CI-approved within the next 30 min or so [14:53] I think my system defaults to qt4 [14:54] I have this error now [14:54] qmlplugindump: could not exec '/usr/lib/x86_64-linux-gnu/qt4/bin/qmlplugindump': No such file or directory [14:54] compiling [14:54] is there a magic env variable? [14:54] Cimi, hm... when did you last pull updates? [14:54] Cimi, `apt install qt5-default` [14:54] MacSlow, all looks good indeed, will run through them and ACK tomorrow morning [14:55] Saviq, what does this package do? [14:55] Cimi, I had a bunch of compiler-errror earlier this day and those went away after pulling updates again this afternoon [14:55] sets the default version to qt5 [14:55] weird [14:55] Saviq, great thx! [14:56] Cimi, why weird? qt4 and qt5 are co-installable, both have the same binaries [14:57] Cimi, so something needs to decide how to deal with that [14:57] mterry: mind taking a quick look? https://code.launchpad.net/~nick-dedekind/unity-system-compositor/trusted-sessions/+merge/221835 [14:57] just some compatibility updates for mir. [14:58] dednick, looks visually fine yeah [14:59] although jenkins should be able to build it now... [15:00] dednick, top-approve when you like [15:00] mterry: thanks [15:00] Saviq, would have preferred a tool [15:00] Saviq, instead a package [15:01] Cimi: Saviq: you don't really *need* the package, you can just export QT_SELECT="qt5" [15:01] like i do [15:01] but then you have to change all control files to not require the package ^_^ [15:01] Saviq, otherwise what happens if you install qt4-defaults then uninstall the packages (and qt4?) [15:01] it should fall back to qt5 [15:01] tsdgeos, I know [15:02] Cimi, sure, it'd probably be handled by alternatives in Ubuntu otherwise, but qt-chooser is an upstream Qt tool, we didn't want to go against that [15:02] Cimi, and it allows you to easily switch between them by just exporting the var as Albert said [15:05] Saviq: platform-api/devel & unity-system-compositor/devel should be using mir staging ppa as well for jenkins CI builds right? [15:05] as well = same as unity-mir [15:05] dednick, yes [15:05] dednick, afaict [15:05] kgunn, can you confirm ↑? [15:05] Saviq: hm. ok, they dont seem to be at the mo. I'll talk to CI [15:10] Saviq, this double click thing is weird [15:10] Saviq, do you have an idea? [15:10] I can try with a testcase but... [15:11] I'm wondering if it will become wasting time [15:14] Cimi, if another click is a workaround, use it for now [15:14] Cimi, with a FIXME [15:16] Saviq, it is weird [15:17] Saviq, I am not sure there is no bug here [15:27] Saviq: what do you mean with "TryVerticalJournal doesn't resize the journal, would be nice to see how it behaves." ? [15:27] it does totally resize here :S [15:27] tsdgeos, as I resize the window [15:28] tsdgeos, the actual journal does not get resized [15:28] i get more/less columns [15:28] tsdgeos, huh' [15:28] you don't? [15:29] tsdgeos, width: units.gu(80) [15:29] tsdgeos, line 359 of the diff [15:29] tsdgeos, and it's only anchored left [15:29] Saviq: that's not tryVerticalJournal [15:30] that's tryResponsiveVerticalJournal [15:30] tsdgeos, sorry, meant that [15:30] does it make more sense now? :) [15:30] honestly didn't think it was needed since the other already does it, but sure, can do [15:30] tsdgeos, but the other does something else [15:31] correct [15:31] tsdgeos, here you'll see that the columns are spaced before there's space for one more etc. [15:31] yep [15:37] mhr3, I dist-upgraded from silo 014, where should I look for depts? [15:37] Saviq, amazon [15:37] mhr3, hmm I only get a single item... wonder if it's geography-based (we don't have amazon here...) [15:38] (single item and no depts that is) [15:38] Saviq, hmm.. maybe, i get three :) [15:38] and depts [15:38] * Saviq shakes fist [15:39] that's gonna be tricky then :| [15:40] Saviq, what do you get with https://dash.ubuntu.com/smartscopes/v2/amazon/search?q=&platform=phone ? [15:40] mhr3, got depts there :/ [15:40] dednick: i saw you got the answer, gcc shenanigans... [15:40] Saviq, when did you upgrade from the silo? i was rebuilding u8 like an hour ago [15:41] mhr3, just now [15:41] hm, should be good then [15:41] mhr3, and it's not there in either favourite or temp scope [15:41] Saviq, pastebin upstart's smart-scopes-proxy.log [15:41] kgunn: that's the issue with unity-mir. usc & platform-api aren't building with mir staging ppa. [15:42] mhr3, http://paste.ubuntu.com/7659027/ [15:42] kgunn: i've spoken to fginther. he's adding the hooks for the ppa to the jenkins projects. [15:42] dednick: ah [15:43] Saviq, hm.. all looks good, check versions of unity-api, u8, scopes-api? [15:43] mhr3, all updated from the ppa :| === gatox is now known as gatox_lunch [15:44] Saviq, music scope not working right? [15:44] mhr3, http://paste.ubuntu.com/7659040/ [15:44] mhr3, yeah, empty [15:44] *weird* [15:45] Saviq: about the 'Should this be "responsive"?' [15:45] it's what we have in the others [15:45] tsdgeos, ok [15:45] Saviq, indeed... last one then - scope-registry.log [15:45] i agree it can be a bit confusing [15:45] mhr3, http://paste.ubuntu.com/7659043/ [15:46] AH [15:46] mhr3, didn't restart smart-scopes-proxy [15:47] oh you're on desktop? [15:47] then yea... restart all [15:47] Saviq, what's the best way to run autopilot tests? [15:47] dednick: fyi...i'm trying to get a ci-train silo, line 12 of the sheet, to get ppa going for folks to use.... [15:47] Cimi, desktop or phone? [15:47] Saviq, well all [15:47] Cimi, on desktop or phone [15:48] I need to test this infographics [15:48] qmltests are fine [15:48] kgunn: cool. unity-mir not up-to-date yet. just doing some tests now. [15:48] sbuilding now [15:49] dednick: ah...so unity-mir mp needs updating ? [15:51] kgunn: yup. nearly done [15:53] Saviq, so? [15:54] Cimi, install the packages on the phone and use phablet-test-run -n unity8-autopilot [15:54] btw [15:54] sbuilding now, after long time [15:54] libunity-mir-dev:armhf : Depends: libplatform-api1-dev:armhf but it is not installable [15:54] and so on [15:54] Cimi, you can take the packages from CI [15:54] known issue or I delete and start from scratch? [15:55] Saviq, would like to have a silo for that [15:55] Saviq, with infographics too [15:55] Cimi, right, we lost the silo [15:55] kgunn: ok, i've just updated the unity-mir trust session branch [15:56] Cimi, let me try [15:58] * Saviq updates the chroots [15:58] Saviq: the departments, you tested amazon scope? [15:59] tsdgeos, yes [15:59] it's the amazon scope that breaks [15:59] not my fault :D [15:59] tsdgeos, you sure? it looked like something broke internally? [15:59] i'm pretty much sure yes [16:00] can you tryDashContent [16:00] or maybe i don't understand what "click on header to close while it's loading" means [16:00] that's "header" in there? [16:01] the department button [16:01] tsdgeos, it looks like it's waiting for it to load at that point [16:01] tsdgeos, I could reproduce in tryDash, but it opened after it loaded [16:02] i'll need you to drive me with what you do there [16:02] but tomorrow, today fix VJ [16:03] tsdgeos, sure [16:09] and my wifi exploded [16:09] Saviq: i think the problem is items are changing size after created and that confuses the VJ [16:11] booo [16:11] tsdgeos_, :| [16:11] Cimi, u8 cross-compiles fine here [16:11] Saviq, weird [16:11] Cimi, make sure to update/upgrade your chroot [16:11] Saviq, I will delete the chroots [16:11] Saviq, how can I? [16:12] Cimi, sbuild-update [16:13] Cimi, also, https://wiki.ubuntu.com/SimpleSbuild#Maintaining_the_schroots [16:13] cool [16:14] Saviq: i can fix it, but tomorrow, too tired now to tackle that big-ish change now [16:17] Saviq, ap seems to run fine... I'll leave it... [16:20] ah [16:20] the phone app has the new bottom edge [16:21] looks quite confusing to use [16:21] when it's in, the header changes weird [16:22] Saviq, did you try the new bottom edge in the phone app? [16:23] am I alone thinking it's weird layer of navigation? [16:23] it comes from the bottom, but the header suggest to navigate left not down [16:24] i think that if we have a gesture from the bottom we should be able to revert it with an action similar but inverted === salem_ is now known as _salem === gatox_lunch is now known as gatox === alan_g is now known as alan_g|EOD === _salem is now known as salem_ [18:06] mzanetti, Saviq: here's a fake launcher, that exposes via dbus the interfaces we discussed, and that uses libubuntu-download-manager-client to fetch progress and error information: https://code.launchpad.net/~alecu/unity-scope-click/launcher-integration/+merge/223461 [18:07] I'll be using that to test the scope branch that will call startInstallation and completeInstallation [18:08] I figured it might be useful as a skeleton for the launcher implementation. === salem_ is now known as _salem === _salem is now known as salem_ [18:43] Is there a 'FunctionSpy' like there is a SignalSpy? === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem === boiko_ is now known as boiko