[08:58] dednick: yo, how do I test your trust helper code now? what command(s) do I run? [08:58] greyback: testing mock or real? [08:58] dednick: emm both! [08:59] but I mostly carea about real [09:00] greyback: real is the same as before. using the demo app. [09:01] greyback: trust-session-demo-trusted-helper -p `pgrep -f clock` - when you have clock app open [09:01] dednick: ok. No need for desktop_file_hint since it uses trusted socket I guess [09:02] greyback: for the mock: [09:02] qdbus com.canonical.Unity8 /com/canonical/Unity8/Mocks com.canonical.Unity8.Mocks.Application.addChildSurface "dialer-app" "/home/nick/Work/phablet/qtcompositor/unity8-merged/qml/Dash/graphics/phone/screenshots/facebook@12.png" [09:02] qdbus com.canonical.Unity8 /com/canonical/Unity8/Mocks com.canonical.Unity8.Mocks.Application.removeChildSurface 1 (1=return of addChildSurface all) [09:02] s/all/call [09:03] Do I need to run unity8 with a new env var to define the trust socket? [09:03] greyback: mmm. i think my MP includes the trust socket change. [09:03] dednick: ok [09:04] maybe not... [09:04] I've not tested yet, still flashing my devices [09:05] dednick: hmm... do you know why the messaging indicator icon is a bell instead of the envelope? [09:05] mzanetti: uuuh. no... [09:06] was it changed to be that icon or is there a new feature that paints the bell in some certain cases only? [09:07] mzanetti: has nothing to do with my code. either it's been changed in backend (don't think so) or it's changed in icon theme [09:08] dednick: yeah, I just thought you might be aware of specs that explain it... [09:09] mzanetti: might be a bug. not sure [09:09] ah ok... [09:10] mzanetti: no, looks like a replacement icon [09:10] I guess I liked the envelope better, but I can see that one doesn't really fit for missed calls [09:11] mzanetti: apparent'y it's green when you have a new notification [09:11] so probably that's the reason... the bell just says something rang [09:11] yeah, it is [09:11] that's only why I noticed it :D [09:11] it was there all weekend, but just now I had a missed call and actually saw it [09:11] :) i hadn't before [09:43] pete-woods: i have hud-service at 100% here if you want to come and debug it live [09:52] tsdgeos: will be over shortly [10:19] elopio: ping [10:51] Holas [11:05] dednick: first pass of https://code.launchpad.net/~nick-dedekind/qtmir/prompt_sessions/+merge/227368 complete [11:24] greyback: thanks. fixed some, commented on some. [11:26] dednick: do I get tests? [11:27] mhr3: just pushed something that should fix the openScope from preview, can you double check? [11:27] greyback: er, are there any tests for qtmir yet? make test interestingly doenst do anything ;) [11:28] dednick: there's a tests/ directory. "make check" runs them I think [11:28] greyback: ok. um, yeah... i guess i can write a few :) [11:28] dednick: ty [11:33] tsdgeos, something broke, ping when back :) [11:49] mzanetti, would you have time for a review https://code.launchpad.net/~saviq/unity8/dash-header-style/+merge/227522 ? === MacSlow is now known as MacSlow|lunch === _salem is now known as salem_ === MacSlow|lunch is now known as MacSlow [13:03] mhr3: fix pushed [13:03] that was quick [13:24] Saviq: I'd like to define some constants in QML (FooColors.qml), and reference them from other qml files. I'm not sure how do to that, though. Item { readonly property string blue: "#2ca5e0" } I reference in another QML as FooColors.blue, qmlscene says FooColors is not defined. suggestions? [13:24] should I define those in a .js file instead? [13:25] karni, yeah [13:25] ok [13:25] karni, http://qt-project.org/doc/qt-5/qtqml-javascript-resources.html#shared-javascript-resources-libraries [13:26] Saviq: thanks [13:26] karni, but [13:27] karni, maybe in 5.3 it's better to use http://qt-project.org/groups/qt-contributors-summit-2013/wiki/Evolution-of-the-QML-language actually [13:27] karni, the singleton approach [13:27] * karni looks [13:27] and thank you :) [13:28] karni, this way the "readonly" part can actually be enforced, which you can't in JS [13:28] right :) [13:34] Saviq, for the overlay-color, i'll add "primary-logo-color" to the appearance attrs [13:35] if you're ok with that [13:37] mhr3, what's that? :D [13:38] mhr3, oh in the .ini? [13:38] mhr3, I'm never gonna see it anyway [13:39] Saviq, you will have it in the hints, but yea, you don't really need to look at it [13:46] mhall119, could you update the scopes docs again pls? [13:56] seb128, shall we land password locking? [13:57] seb128, I'd hijack silo 6 a little and add a few unity8 branches (since one got superseded anyway), wdyt? [13:58] Saviq, +1 from me, we have some other small u-s-s changes ready (mostly a bugfix and some test improvements), do you want to batch them with it or keep that for another landing? [13:59] seb128, good for me, lemme add unity8 stuff to the line and you handle u-s-s [13:59] Saviq, ok [14:00] seb128, hmm the only ~scary thing would be the livecd-rootfs change mterry mentioned in the comment [14:00] to delete the default pwd [14:01] seb128, so we should wait for him (he should be around soon) [14:01] Saviq, k [14:01] seb128, line 14 ready for you [14:02] Saviq, done [14:02] seb128, ok, /me reconf + full rebuild [14:02] Saviq, thanks [14:32] #qtcomp [14:35] mhr3: done [14:39] mzanetti: pong === dandrader is now known as dandrader|afk === greyback is now known as greyback|away [14:44] seb128, crap, mterry is out today, we might need to land without the password things, otherwise people will get (seemingly) locked out of their devices [14:45] Saviq, :-/ [14:45] seb128, I'll drop the locking-hash branches from the silo then and rebuild once more [14:46] Saviq, thanks === dandrader|afk is now known as dandrader === greyback|away is now known as greyback [15:05] greyback, should the window stack report app in foreground if greeter is locked? [15:06] (because it apparently does) [15:06] Saviq: I know it does. I don't know of anyone who cared [15:06] greyback, Chipaca cares [15:06] greyback: 'ullo [15:07] greyback: is it an easy fix? [15:08] Chipaca: hey. It may not be too hard. Need it urgently? [15:08] greyback: define "urgent" [15:08] Chipaca: how soon do you need it? [15:08] i can wait *days*. Like, *three* of them :-p [15:09] Chipaca: ok, please log a bug against unity-mir and we'll take it on [15:09] greyback: will do [15:10] greyback: it can probably wait a week or two too, fwiw :) [15:10] but it's needed for rtm [15:10] one way or another [15:11] greyback: https://bugs.launchpad.net/unity-mir/+bug/1346367 [15:12] Chipaca: ok [15:13] greyback, thanks [15:14] np [15:14] greyback: yes; thank you! [15:15] Chipaca: no worries [15:16] greyback: if it turns out to be hard (or breaks other stuff), a supported way of knowing whether the lock screen is up would work too (if be less convenient for us) [15:17] Chipaca: it should not be too tricky. I'm more curious if WindowStack definitely gives you the info you need. [15:17] Chipaca: push notifications are only supposed to show if no app is focused, or Greeter is open? [15:18] greyback: push notifications for an app are not to be presented if that app is in the foreground [15:18] Chipaca: gotcha [15:18] greyback: i'll tweak the description to be clearer [15:19] greyback: there [15:19] Chipaca: great, thank you. [15:19] Chipaca: can you add a way for us to test a push notification for an app? [15:19] greyback: sure [15:19] Chipaca: magic [15:23] greyback: like so? [15:25] Chipaca: perfect [15:36] Saviq, Can I create a AttributeGrid.qml and then use it inside the CardCreator.js rather than include all the code inside the .js? [15:36] paulliu, surwe [15:36] -w === gatox is now known as gatox_lunch [15:41] * greyback needs to reboot [16:15] mhr3, bug #1346384 :| [16:15] where's the damn bot? [16:16] https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1346384 [16:27] Saviq, hmm, cache crashing? :/ [16:28] maybe it's only during shutdown or something [16:30] dandrader: getting an crash from qtmir. "what(): Failure sending input event : Unknown channel provided" [16:31] dednick, in what situation? [16:32] um. touching surfaceitems which backend may or may not have been destroyed... [16:32] dandrader: my prompt session work. [16:38] dednick, backtrace? === gatox_lunch is now known as gatox [16:38] dandrader: not available. trying to install debug symbols now [16:39] dednick, I'm afraid I would have to build & try out your prompt sessions stuff to be able to do anything useful about it [16:42] that crash should be inside mir [16:42] thostr_: ping === dandrader is now known as dandrader|lunch [16:49] mhall119: pong [16:50] thostr_: hey, for Scopes, what should developers use for accessing the network and location data? [16:50] will it be the same Qt APIs as applications use, or something different? [16:51] mhall119: location will be passed as query parameter, so no need (but also no possibility) to use something else [16:51] mhall119: network: either net-cpp or Qt... but net-cpp should be more straightforward [16:52] do we have API docs for that which we can publish on developer.u.c? [16:52] mhall119: not sure... I don't think so... need to check with tvoss [16:54] is libcppnetlib the correct package for that library? [16:59] mhall119: no, libcppnetlib is something different... but could probably be used as well === alan_g is now known as alan_g|EOD [17:17] I'm not seeing net-cpp in an apt-cache search [17:18] why are we not using the Qt api? It integrates better with the toolkit we tell people to use [17:24] greyback, Scopes aren't necessarily Qt based, they're C++. You can run Qt if you want, but that's not what we're pushing for non-GUI stuff. [17:26] tedg: Qt is not just for GUI applications. It can do command line apps just as easily. === dandrader|lunch is now known as dandrader [17:28] tedg: if we use one API for scopes and another for apps, now we've 2 equivalent APIs to support [17:28] I don't see how that's a good thing [17:28] greyback, Can, sure. Not sure why you'd want all that for something simpler. [17:30] tedg: "simpler" ? - Qt really does not have that much of an overhead, and it offers so many useful APIs, all of which works nicely together. I fail to see the win from using "pure" C++ [17:31] greyback, moc [17:32] tedg: please explain how some pre-generated C++ code that you don't have to touch is so terrifying [17:33] greyback, Taking my C++ and generating more C++ is terrifying. This is why templates were created. [17:37] tedg: I think you're dismissing it too quickly, and in my time working wit Qt, moc has seldom been trouble. [17:37] instead I've mostly fought with non-Qt C++ libraries and how none of them really quite match up [17:39] greyback, So use Qt. We're saying you don't have to use Qt to use our platform. You're welcome to if you want, but if you don't, that's cool too. [17:40] greyback, You can use netcpp, or just plain ol' unix sockets. [17:44] a consistent message to our developers is all I ask [17:47] greyback, The Linux kernel should be able to load your binary and execute it. :-) [17:51] tedg: the question isn't about whether you *can* use Qt or "pure" C++ APIs, the question is which should we be recommending and supporting [17:52] to greyback's point, we're already recommending and supporting those Qt APIs for apps, so there's little extra cost to us in using them again for Scopes [17:52] however, if Qt itself brings an added cost to developing Scopes, then it may not be the right technology to use there [17:53] so, does the cost of using the Qt APIs for scopes outweigh the benefits of having only 1 API for doign it across the whole platform [17:53] mhall119: would be good to have that cost quantified, instead of just the notion that "Qt is heavy" [17:53] The important thing is that you can develop a scope without Qt. [17:54] tedg: keeping in mind also that Go support will eventually be needed for both apps and scopes, and I assume the Go/Qt bindings will allow those APIs to be used [17:54] I think that's a pretty big assumption :-) [17:55] perhaps [17:55] so we'll have 3 messages for developers? [17:55] sorry but that's just horrible [17:56] mhall119, Just FYI, there's already support for Go scopes. [17:57] I believe (not sure) Soundcloud is a Go scope. [18:03] yeah it is [18:07] tedg: there *are* Go scopes, but from what I've been told it's not ready to officially support yet, and we have no documentation on it on developer.u.c [18:07] bregma: ping [18:07] mhall119, pong [18:08] bregma: hey, I'm working on your survey for convergence feedback [18:08] most excellent [18:08] mhall119, Ah, I see what you mean. [18:08] bregma: http://www.surveymonkey.com/s.aspx?PREVIEW_MODE=DO_NOT_USE_THIS_LINK_FOR_COLLECTION&sm=0DrIewtoCDgDEWXFSUu9VRIHsWY9IYG%2bvNNyfmNzxFw%3d if you want to take it for a test-run [18:09] bregma: I plan to add an introductory page explaining the purpose of the survey, and also a thank you page at the end [18:09] * bregma goes clicky clicky [18:09] bregma: I added 3 "about you" questions as the end to get context about the respondent, but if you can think of any others I will add them as well [21:43] bregma: any feedback on the survey? === salem_ is now known as _salem