[03:17] ah man, mk-build-deps fails for unity8 with this error: dpkg-gencontrol: warning: can't parse dependency g++-4.9:native. [03:18] changing native -> armhf works for me right now, but is this a bug in the debian/control or in mk-build-deps === duflu_ is now known as duflu [06:49] mzanetti: can you share the unconfined app you used to test the qtsystems, if I try to test it on wily? [07:38] mzanetti: ping [07:38] ping is a too dangerous word, half of the times i write pig and have to retype before offendign people :D [07:39] lol [07:39] tsdgeos, no worries... I wouldn't get mad [07:40] mzanetti: does Staged here https://code.launchpad.net/~mzanetti/unity8/inputinfo-default/+merge/263969 mean "one stage" on the Nexus4 and "2 stages" on whatever we support the tablet mode? [07:40] tsdgeos, yes [07:41] and does it still work on the desktop? i.e. do i get windowed there? [07:41] basically it means !Windowed [07:41] tsdgeos, no, one needs to set that to Windowed or Automatic [07:41] tsdgeos, so the reason for this branch is that Autopilot triggers the mouse detection [07:42] yeah i saw the bug [07:42] mzanetti: but then shall this be accompanied by something else so we don't regress on the desktop? [07:42] tsdgeos, well, it has been staged so far, only changed to automatic on friday [07:42] so far people are used to manually change it to windowed [07:43] hmmm, i've never changed it manually and i get windowed mode on the desktop [07:43] maybe some other package is already changing it to windowed? [07:43] oh... [07:43] right [07:43] so that would still work I guess [07:43] like the unity8-desktop-session or something? [07:43] because it's an override [07:45] ok, i'll give it a try [07:45] Mirv, http://notyetthere.org/data/untitled5.mzanetti_0.1_all.click [07:45] mzanetti: shall you link the bug to the MR ? [07:46] it's not a fix for the bug actually. because eventually we will have to enable this [07:46] rather just a temporary workaround to not break peoples AP test runs [07:48] tsdgeos, actually veebers has been looking into this and turns out the call to "Touch.create()" causes this. [07:48] so maybe there's only a small property in that code that needs to be changed to make it a touch device (instead of a mouse) for real [07:48] ok, so don't review it for the moment? or do we still to land this nevertheless? [07:49] not sure yet [07:49] ok, so let's wait :D [07:52] Mirv, if you're using that app, don't look at the label on the screen [07:52] Mirv, instead tail -f ~/.cache/upstart/application-click-untitled5.mzanetti.log and see if it says "enabled/disabled" there [07:52] I didn't bother to make the app pretty [08:10] mzanetti: thanks! [10:35] mzanetti: ltinkl: i thought this was "on hold" pending the other AP fix? https://code.launchpad.net/~mzanetti/unity8/inputinfo-default/+merge/263969 [10:35] tsdgeos, yeah... will drop it, now that the issue is about to be fixed in AP [10:35] sorry ltinkl :/ [10:35] tsdgeos, if that's the case, sorry, it hadn't been marked as such [10:35] mzanetti, no probs [10:35] tsdgeos, btw: https://code.launchpad.net/~mzanetti/autopilot/dont-use-touchpad-for-touch/+merge/264011 [10:36] I guess we'll leave it to the AP people to review/land it [10:36] yep saw it [10:36] +1 === karni is now known as karni-afk === karni-afk is now known as karni === MacSlow is now known as MacSlow|lunch === MacSlow|lunch is now known as MacSlow [14:19] where does unity8 store that it already displayed the tutorial/first run demo? [14:21] seems like it's in accountsservice as demoEdges [14:22] not as easy than touching a file to skip it on an iso === Olli_ is now known as olli === dandrader is now known as dandrader|afk [15:56] cimi: there? [15:57] tsdgeos, sure [15:57] cimi: i don't understand the question you made on the MR [15:57] " I'm missing where this is needed... maybe you wanted to add this check for the model of the repeater?" [15:57] to which line does that relate? [15:57] to readonly property var actions: root.widgetData ? root.widgetData["actions"] : null ? [15:57] tsdgeos, if you grep for it, nothing is row.actions iirc [15:58] damnit the email puts suff in the wrong order [15:58] i should have opened the web [15:58] email is [15:58] > + id: row [15:58] I'm missing where this is needed... maybe you wanted to add this check for the model of the repeater? [15:58] > + readonly property var actions: root.widgetData ? root.widgetData["actions"] : null [15:58] which made me think it was about id: row :D [15:58] tsdgeos, and since the model of the repeater is root.widgetData["actions"] I thought you planned to do model: row.actions [15:58] sorry ^_^ [15:59] cimi: yeah good question i've no idea what i wanted to do with that :D [15:59] tsdgeos, lol [15:59] cimi: i guess i'll do what you said model: row.actions [16:00] and i'll set the sourcesize [16:01] tomorrow though, i'm going to try to find somewhere in the house that is not so hot and try to not melt there :D [16:01] * tsdgeos waves [16:23] is anyone looking at making the messaging menu "delete" action use the red icon rather than the old label? === dandrader|afk is now known as dandrader [16:37] dednick, there are some approved ubuntu-settings-components merges (and some pending reviews), do you know if anyone is looking at landing those? [16:37] mzanetti, ^ [16:37] https://code.launchpad.net/~ubuntu-settings-components-team/ubuntu-settings-components/trunk/+activereviews [16:38] seb128, I didn't really have them on screen tbh [16:38] seb128: no, i don't know who is looking after that. mzanetti? [16:38] thanks for pointing it out [16:39] yw [16:39] ok. added to my list [16:39] thanks === dandrader is now known as dandrader|bbl === dpm is now known as dpm-afk === dandrader|bbl is now known as dandrader [19:18] tedg, you might be interested in that: https://docs.google.com/document/d/15yq0ziuINo7zLflvcLCuThjY2grftcFMynMcGFLl0uk/edit [19:40] dandrader, erm... my krillin's greeter is locked to landscape :) [19:41] mzanetti, have you been abusing your phone? [19:46] dandrader, not really, no [19:46] dandrader, just using it normally [19:46] I think the last thing I used was telegram [19:48] dandrader, anything I can do in terms of information gathering before I restart? [19:49] seems the focused app is the dash, according to the logs [19:49] mzanetti, maybe check for anything interesting in unity8.log. other than that, I don't know [19:50] nothin interesting in there... [19:50] mzanetti, we do have a bug on that already, right? [19:50] dandrader, something like this on N7. but I can't be 100% sure it's the same thing [19:50] also popey reported this on a krillin today morning [19:50] I'll try to repro [19:51] hah! found a workaround [19:51] launch the settings via indicators, then turn off the screen and on again, it'll rotate [19:51] mzanetti, thing the only sane next step for debugging this is me prepping a special unity8 version where you can tap somewhere on the screen to have the values of some key properties printed in the log [19:51] mzanetti, s/thing/think [19:52] dandrader, I can run such a thing, no prob [19:52] yeah, i found it happens if the screen times out while you're playing a landscape game [19:52] if the area is not preventing me to use it as a phone [19:52] popey, you have a way to repro and didn't tell us? :D [19:53] i thought i said how in the bug? [19:53] but now that you're saying it, seems I had machines-vs-machines open indeed [19:53] oh... [19:53] lol [19:54] changed all my pin numbers so they dont have zeros in :( [19:54] haha [19:54] popey, saw the workaround ^^ [19:54] ? [19:54] no [19:54] launch settings from the indicators, then turn off/on the screen [19:54] i cant [19:54] indicators are off for me [19:55] huh? [19:55] on the lock screen [19:55] on the greeter? [19:55] yeah, i switch them off in settings [19:55] oh. I see [19:57] popey, can you intentionally reproduce it? [19:57] seems I can't [19:57] i think you have to let the phone lock itself [19:57] not lock it with pwr button [19:59] yeah, just tried twice didn't work either [19:59] but I did see this issue now twice by now [20:00] yeah, i have seen the issue twice but can't reproduce it [20:01] maybe it's something weird, that you a) need to let the screen time out in landscape and then b) unlock it in InvertedPortrait or so :D [20:01] dandrader: don't forget about a tool like Gammaray, allows you to inspect the internals of a qt app [20:03] never used it [20:03] greyback_, does it work with QML? [20:04] dandrader: sure. they're just qobjects under it all [20:04] I've a reasonably recent build in my PPA: https://launchpad.net/~gerboland/+archive/ubuntu/unity-mir [20:04] greyback_, so you run gammaray on the desktop to inspect some qt process on the device? [20:05] greyback_, can it work remote to the phone? [20:05] dandrader: that is possible yes. [20:05] nice [20:05] greyback_, have you done it? [20:05] I think so. It was ages ago [20:05] sadly the full qml integration does not work from device-to-PC [20:06] but you can still get the tree of qobjects and inspect their state [20:06] that should be enough I think [20:06] it's not too hard to figure which qobject corresponds to what qml [20:06] * mzanetti installs [20:07] greyback_, are you running wily? [20:07] mzanetti: no [20:07] odd [20:07] gammaray : Depends: qtbase-abi-5-3-2 but it is not installable [20:09] hmm, will need to bump what's in the ppa [20:09] dandrader, anyhow, if you prepare some qml file with a MouseArea that prints stuff I'll run it on my phone and send data your way [20:09] 32 weeks old. Where does time go [20:09] dandrader, if possible, enable it only when the greeter is visible, so I can still use the phone normally otherwise [20:10] maybe even just hook it up to the tease() method that passes through Shell.qml [20:11] greyback_, indeed :D [20:11] I still remember the conversation we had when you uploaded those packages [20:11] can't believe that's 32 weeks :D [20:52] mzanetti: available? Your branch errored on a flake8 issue, if you're not I'll branch, fix, push and approve etc. [20:58] veebers, fixing [21:04] mzanetti: thx awesome. I'll get the silo wheels turning once it's approved && merged [21:18] veebers, hey, having an issue here :/ [21:19] seems with this change click_object() doesn't work properly any more [21:35] mzanetti: just click_object? or any touch/click/tap etc. [21:36] veebers, I was just trying the webbrowser tests and they use click_object() everywhere... the _uinput file has tap_object... but I can't access that from the test [21:37] mzanetti: when you say doesn't work properly that's as in there is no touch event logged? [21:45] veebers, it seems it presses but never releases [21:45] veebers, sorry, was distracted in another channel [21:46] mzanetti: no worries. That's really odd as you've only changed the touch device, nothing else. Only what gets created [21:46] mzanetti: and if you change it back it all works? I know there was some input work/fix gone into mir and or unity8 recently [21:47] yeah... it does... [21:48] veebers, I might have found something [21:48] oh? /me waits on the edge of his seat [21:51] veebers, hehe. I think I found it [21:51] finger_down() finger_up() [21:51] they used BTN_TOOL_FINGER instead of BTN_TOUCH too [21:52] replacing all makes it work [21:52] mzanetti: oh no way, ugh what a bug. Good catch [21:54] veebers, pushed [21:54] mzanetti: awesome, thanks [21:55] veebers, this is actually quite cool. should be easy to create a touchpad/mouse/touchscreen/wacomtablet and whatnot input methods [21:55] mzanetti: I'm tempted to suggest to use _get_touch_tool in the finger_up|down methods, but not sure if it's worth it [21:56] I thought about that too [21:56] mzanetti: yeah ^_^ [21:56] but decided if you want to refactor parts of the code it's probably faster if you do it yourself instead of playing ping-pong with a python noob [21:57] oh... [21:57] that [21:57] I thought about getting rid of that method [21:57] I can change the other I guess [22:00] veebers, pushed [22:02] mzanetti: that will fail, _get_touch_tool isn't an instance method (no need for self.) [22:10] hmm... really.. I did run it tho and it didn't complain [22:13] veebers, ok... removed it.. still wondering why it didn't complain before [22:13] mzanetti: not sure, I just tried using it and got the expected error [22:13] ok... then I'm doing something wrong [22:14] ok... verified. now I'm getting same results at least [22:15] mzanetti: yeah, https://jenkins.qa.ubuntu.com/job/autopilot-wily-amd64-ci/37/console [22:17] mzanetti: just to bombard you with requests :-) any news on being able to cross build unity8 for vivid-ar, [22:17] err vivid-armhf [22:18] veebers, not really. But I've reported a bug pete said he'll look into it [22:18] coolio, thanks mzanetti [22:34] mzanetti: which device where you running your test on? [22:34] veebers, mako [22:34] mzanetti: sweet thanks