[11:41] tsdgeos: https://code.launchpad.net/~aacid/qtmir/unlikelyResetStartTime/+merge/280569 is not merging into trunk, deliberate? [11:41] greyback: nah i messed up merge into with prerequisite i guess [11:42] greyback: give me a min to fix it [11:42] sure [11:43] greyback: done [11:50] dandrader: hey, I've a few old reviews I need you to quickly look at again, can I add to your todo list: [11:50] https://code.launchpad.net/~gerboland/qtmir/fix-cmdline-args/+merge/277691 [11:50] https://code.launchpad.net/~unity-team/qtubuntu/DPR2/+merge/281664 [11:51] greyback, ok [11:51] and this one needs another go-over: https://code.launchpad.net/~unity-team/qtmir/dpr/+merge/257514 [11:51] I'm in review mode today too [12:18] greyback, got people looking at the DPR silo yet? mzanetti maybe you could spend some time with it? [12:19] Saviq: at the silo, no. [12:19] kk, can give it a test [12:19] https://requests.ci-train.ubuntu.com/#/ticket/776 <- docs in desc [12:19] mzanetti: ^ [12:22] ack, ta === popey_ is now known as popey === alan_g is now known as alan_g|lunch === alan_g|lunch is now known as alan_g === alan_g is now known as alan_g|lunch [13:13] @unity, btw, you can already access https://unity8-jenkins.ubuntu.com/ and see what's happening there if you like [13:17] dandrader, hey, I just came by this https://bugs.launchpad.net/mir/+bug/1517133 and then there is: https://code.launchpad.net/~ken-vandine/unity8/mouse_touchpad_schema/+merge/281643 [13:17] Launchpad bug 1517133 in unity8 (Ubuntu) "[regression] Mouse speed in Unity8 is much lower than elsewhere (to the point of useless)" [High,Incomplete] [13:18] dandrader, what is the plan for mouse speed settings? [13:18] should unity8 read them from config and tell values to Mir? [13:20] mzanetti, afaik, no plan yet. that sounds ok to me. wonder if libinput has its own config file or something [13:20] dandrader, well, we need more than just speed and accel. ken's branch also prepares things like changing button order for left-handers etc [13:21] so unless libinput can do all that stuff on its own, I guess we want dconf & unity8 to handle it. if libinput can, we might want to talk to ken so the settings app controls that? [13:22] mzanetti, also maybe mir wants to hide libinput as an implementation detail [13:22] mzanetti, so mir users should be oblivious to it [13:22] yes, fair point [13:23] kk, thanks [13:23] lets go with unity8 controlling things then [13:23] mzanetti, don't even know if mir already exposes a way for use to tweak those things [13:23] anpok_, ^^ [13:24] s/for use/for us [13:42] dandrader: there is, see include/server/mir/input/device.h [13:45] mzanetti, dandrader: there is and libinput has an API to control this: http://wayland.freedesktop.org/libinput/doc/latest/group__config.html [13:46] even the left/right handedness stuff etc [13:46] ltinkl: libinput isn't exposed through mir though, it's internal only - instead mir has apis which we can use (hopefully all we need) [13:46] mzanetti, so maybe system settings should set this directly? [13:46] greyback, yeah, that's what I meant [13:47] ah ok [13:47] but I don't see the need for unity8 to control these things (if we don't have any config UI) [13:47] system settings has [13:47] or will have [13:47] ltinkl: ordinary apps can't use libinput to configure stuff tho (unless they're root), it has to be via mir then [13:48] unless I'm very mistaken ofc [13:48] * greyback a bit wobbly on this tbh [13:48] greyback, yup, so am I correct in thinking system settings should drive this via Mir? [13:48] ltinkl: yes [13:48] we'll need to create a quick dbus api in unity8 which uses these Mir apis [13:49] ltinkl, system settings is the wrong place imho [13:49] it's not a service [13:49] it's not going to be active on boot/login [13:49] something needs to read the config and apply it on session start [13:49] right, so we need that DBUS bridge [13:49] something should be unity8 imho [13:49] or we needs a settings daemon... [13:49] seb128, ye what gerry wrote, a DBUS service [13:50] wfm [13:50] a service, I don't care much if it's a dbus one [13:50] which gets started together with u8 and reads/applies those settings [13:50] or a systemd unit or whatever [13:50] dbus handy as quick, and appArmor can manage permissions [13:50] (quick to implement with Qt that is) [13:51] greyback, yeah and there might come a moment when we actually use the stuff to read some config from within the shell :) [13:52] ltinkl: whoa, let's not get too optimistic! :D === alan_g|lunch is now known as alan_g [14:08] ltinkl, greyback, seb128, won't we need a keyboard indicator at some point anyway, shouldn't that be who drives it? [14:09] Saviq, the way it currently works on unity7 is that the indicator changes the gsettings config and the settings daemon watch the config and apply it to the server [14:10] but I guess the indicators could to that themselve [14:10] I don't have a strong preference either way [14:10] Saviq: I think it's pointless to make a real keyboard indicator. All the relevant state for keymap is in unity8 anyway. We'd just be export that state to separate indicator process, for it to send that state back [14:11] greyback, right, in that sense we could inject a dummy indicator we, in fact, own [14:11] the UI for it, I mean [14:11] yeah [14:12] which roughly coincides with what Wellark_ wants us to talk about for indicators, where they would supply us with QML instead of us interpreting the ever-growing GMenuModel approach [14:13] indeed === dandrader is now known as dandrader|afk === lool- is now known as lool [15:04] greyback, testing the dpr silo. when disabled, I couldn't spot any issues. when enabled, I can confirm your finding. not exactly sure why the terminal would behave like that [15:04] if I change font size in the terminal settings to make it look normal, things get quite blurry [15:06] I like how the scrolling speed is fixed [15:10] mzanetti, terminal likely deals in real pixels, and gets scaled up, similar to canvas [15:10] I wonder if we shouldn't fix the shell rotation animation at least [15:10] given that's in our hands [15:11] but apart from that, I guess we should land it, and we should push for fixes for the rest [15:11] mzanetti: that's my strategy === dandrader|afk is now known as dandrader [15:12] mzanetti: I found another issue that occasionally locks up the shell at DPR2 (indicator image scaling x2 for infinity), so until I've that fixed, we're leaving it disabled [15:15] greyback, ah, the Icon in the indicator is still an issue? [15:16] greyback, the Icon is a bit broken anyways, it frequently prints binding loops about implicitHeight without DPR too [15:16] mzanetti: yep [15:17] greyback, I've update th issues description [15:17] e.g. UbuntuShape is fine, but ProportionalShape isn't [15:17] all ActitivyIndicators are too big [15:18] but I've ran the test plan with DPR=1, couldn't spot an issue there [15:20] o_O [15:20] Saviq, I did a (successful) indicator injection in my desktopBrightness branch (brightness slider) [15:20] i will be back next week.. [15:21] usc in 0.18 has a DBUS API to confgure input devices.. the interface is com.canonical.Unity.Input.. [15:21] it coves everything u-s-s exposes in the gsettings schema.. [15:22] for 0.19 or 0.20 I will add a mir client API so nested servers and clients can change the input settings too .. additionally they will be able to notice which devices are plugged unplugged ... [15:23] dandrader, mzanetti, greybac ^ [15:23] ack [15:24] anpok_, I think the speed config stuff is gaining priority now... but lets talk next week when you're back [15:24] anpok_, I guess it doesn't cover (yet) the new stuff proposed in https://code.launchpad.net/~ken-vandine/unity8/mouse_touchpad_schema/+merge/281643 ? [15:30] ltinkl: it does .. double click times are a toolkit thing [15:30] the other ones should map directly === dandrader is now known as dandrader|bbl === alan_g is now known as alan_g|EOD === dandrader|bbl is now known as dandrader === Saviq_ is now known as Saviq === dkessel_ is now known as dkessel === bschaefer_ is now known as bschaefer