rickspencer3that should be relatively simple, for example00:00
rickspencer3I could probably do that nowish ;)00:00
rickspencer3put some colors along the bottom and set the opacity to like .800:00
mhall119that would probably get a lot of people's interest00:00
mhall119lots of geek parents out there00:01
rickspencer3mhall119, except, err ...00:01
rickspencer3did you see the screenshot at the bottom of the blog posting?00:01
rickspencer3apparently I am pretty good at breaking my system00:01
mhall119does it do that when running in a desktop window too?00:01
mhall119or just on a device?00:01
rickspencer3mhall119, just on my Nexus 700:01
rickspencer3whenever I touch the context object of the canvas, that happens00:02
mhall119could be hardware related00:03
mhall119depending on how Qt's Canvas is accessing the screen00:03
mhall119a bad driver or something00:03
rickspencer3there are some settings on the canvas that I could play with00:03
rickspencer3but would be better to have an engineer debug it00:04
=== chriadam|away is now known as chriadam
mhall119who on the phablet team has small children?00:04
jono_mhall119, I think a finger painting app would be popular00:05
jono_this reminds us, we need to discuss getting some of the cream of the crop of the apps on the default image too00:05
jono_e.g. the stock tracking app00:06
mhall119jono_: we're already started the process, we have a new PPA for those apps, and once we have installable packages bfiller and his team can review them for inclusion00:07
jono_mhall119, oh awesome, nice work :-)00:07
jono_mhall119, did you and discuss anything re. the app showdown while I was away?00:08
jono_I really want to keep this momentum flowing00:08
mhall119not regarding a contest no00:08
jono_mhall119, ok, lets discuss this this week00:20
rickspencer3mhall119, check out ...00:24
rickspencer3            PaintPot00:24
rickspencer3            {00:24
rickspencer3                height: paintPotSize00:24
rickspencer3                width: paintPotSize00:24
rickspencer3                color: Qt.rgba(1,1,0,.5)00:24
rickspencer3                onColorSelected: {inkCanvas.inkColor = selectedColor}00:24
rickspencer3            }00:24
rickspencer3mhall119, lp:~rick-rickspencer3/+junk/FingerPaint00:24
mhall119that was fast00:25
=== SkavenXXI is now known as SkavenXXI-[OFF]
mhall119rickspencer3: wow, that's awesome00:27
mhall119I can't believe that only took 20 minutes00:27
rickspencer3mhall119, I made a slight change to make the background white00:27
rickspencer3mhall119, well, it took me a little while longer to make the InkCanvas ;)00:27
rickspencer3but yeah, good components = productivity00:28
mhall119I can see InkCanvas being very useful00:28
mhall119I wonder if it could be incorporated into the Notes app00:28
rickspencer3mhall119, well, it's pure QML00:28
rickspencer3so should be easy to give it a try00:28
rickspencer3mhall119, should I put up a quick blog post about FingerPaint?00:29
rickspencer3while I wait for my Nexus7 to update so I can try it there?00:30
mhall119rickspencer3: yeah, and if you can give a description of the code and how it works, I'll link it on developer.u.c too00:32
rickspencer3mhall119, which part of the code?00:32
rickspencer3InkCanvas is a bit complicated00:32
mhall119yeah, it probably is...00:32
mhall119how about how to use InkCanvas, and just links to download InkCanvas itself00:33
rickspencer3mhall119, sure00:33
rickspencer3but not tonight00:33
rickspencer3that will take me a bit longer00:33
rickspencer3and ug ...00:33
rickspencer3the touch screen calibration on my Nexus 7 is waaaay off00:34
mhall119that's fine, I'm working on a blog post tonight already00:34
mhall119man, we're on image 80 now? I haven't even flashed to 79 yet00:35
rickspencer3hey mhall119if you do anything with FingerPaint, you might want to pull00:59
rickspencer3I just deleted a ton of the code01:00
aliAnyone know if it's possible to flash ubuntu onto an LG Optimus G (Nexus 4 variant)01:01
mhall119rickspencer3: nothing about FingerPaint it in tomorrow's blog, but I'll be happy to promote yours when it's posted01:15
mhall119oh, you already posted it I see :)01:16
mterryI can't run "autopilot run qml_phone_shell" in lp:unity/phablet... After failing a builddir path issue, all tests fail because they can't find autopilot interface.  Am I running it differently than I should?01:50
mterrySaviq, ^02:01
mhall119rickspencer3: wow, the calibration really is off on the N703:17
mhall119rickspencer3: if you want, I used james_w's work on pkgme to create packaging for your FingerPaint app: lp:~mhall119/+junk/FingerPaint03:21
sergiusensmterry: you need to run in a real shell and not an adb shell04:19
dholbachgood morning06:39
oSoMoNgood morning07:01
mardyis there a tutorial or guide explaining how to setup pbuilder for Ubuntu Touch?07:56
seb128mardy, ubuntu or ubuntu touch, should be only sources.list differences no?07:59
mardyseb128: mostly yes, but I need to tell pbuilder to build for armhf08:00
pdeyPorting question: is having the kernel sources/config for the target device mandatory?  I'm trying to port Ubuntu Touch to a device with no sources, and CM10 hasn't been ported yet either.08:25
Namidairopdey: well considering that you need to change things in the defconfig, kernel sources would probably be plus :)08:28
pdeyNamidairo: are the chances of getting it running on the stock kernel .. slim-to-nil?08:29
Namidairoit might run, but i suspect that the inbuilt browser won't work08:29
Namidairoor the wifi in general08:29
pdeyI thought Ubuntu was running in a chroot though... wouldn't the wifi drivers,etc already be loaded by the outer Android part?08:30
Namidairo"Without paranoid_network disabled, the browser won't work correctly"08:31
pdeyah, right.08:31
pdeyshame there's no way to extract the config from a kernel binary08:32
pdeythough I presume even if I had the config, I'd need the glue or whatever to link in the proprietary modules08:32
=== ckpringle_ is now known as ckpringle
Namidairomodifying the zimage only gets you so far i think08:35
Namidairowouldn't it be easier to just ASK your manufacturer to give you the source...08:36
pdeyFunnily enough, I tried that... it's available under a NDA with a minimum order of...10,000 units.08:36
Namidairo...and what device is this08:43
pdeyAinol Novo 7 Venus ... Actions ATM7029 SoC08:44
=== SkavenXXI-[OFF] is now known as SkavenXXI
=== ckpringle_ is now known as ckpringle
=== oreneeshy_ is now known as oreneeshy
=== chriadam is now known as chriadam|away
Mirvlool: is the ubuntu-ui-toolkit upload for raring still planned, or will we rely on PPA updates only?11:01
nik90mesq: ping11:08
=== MacSlow is now known as MacSlow|lunch
loolMirv: hey; sorry, I think I got confused because we discussed setting up dailies and then I stopped thinking about uploading it12:07
loolMirv: I just took a look; the changes are fairly large, did this pass QA in some way?12:07
loolMirv: (I'm looking at .38...42)12:08
Mirvlool: all merges are reviewed at least, and everyone uses now the 0.1.42 as it's in the PPA12:08
Mirvfor both precise, quantal and raring users12:08
Mirvmaybe the bzr417 would be a safe bet, ie when 0.1.42 was released to the PPA. there are three commits after that as well.12:09
=== MacSlow|lunch is now known as MacSlow
Mirvalso the raring touch images used the PPA12:10
loolMirv: I'm fine taking r420 or r417; if r420 then we ought to bump the version number12:12
lool(IIUC 0.1.42 was uploaded to the PPA with r417)12:12
Mirvlool: I'd take r417 since it's released as defined by sdk team12:13
loolMirv: well, why don't I just take the PPA .dsc and sponsor that12:13
Mirvlool: true, just fixing the jenkins auto-changelog is needed12:14
Mirvhmm, it's also a native package it seems12:15
loolMirv: which auto-changelog would need fixing?12:15
Mirvlool: removing the "Automatic build of revision 417" from dget https://launchpad.net/~ubuntu-sdk-team/+archive/ppa/+files/ubuntu-ui-toolkit_0.1.42%7Eraring1.dsc12:15
loolMirv: if it needs any change, then it's probably better to have them in bzr and I sponsor from that, but that would suck in r42012:16
loolI can't remove the changelog entry without repacking a tarball12:16
looloh well, nobody released a 0.1.42 tarball anyway12:17
Mirvlool: so how about  lp:~timo-jyrinki/ubuntu-ui-toolkit/bzr417 ? I did bzr uncommit three times from trunk :)12:17
Mirveither that or cleaning the jenkins added entry12:18
loolMirv: erf, I can do that (bzr co -r 417 BTW), I just wanted to avoid generating a new tarball12:18
Mirvlool: right, well that's needed anyhow12:18
loolbut since I'd have to change the one from the PPA, I guess it's fine if I just rebuild from r417 or from the PPA source12:18
Mirvbecause of the jenkins mangling12:18
loolsorry it's bzr revert -r 417 not co12:19
didrocksMirv: after that release, do you mind preparing for daily release in S?12:20
didrocksMirv: then adding the stack to head and so on?12:20
loolMirv: uploaded r417 of lp:ubuntu-ui-toolkit12:20
loolI compared to the PPA .dsc and the only delta was debian/changelog12:20
lool(jenkins entry)12:20
Mirvlool: just out of interested I checked manually what's the difference between pure 417 and the jenkins upload: http://pastebin.ubuntu.com/5712997/12:21
Mirvlool: ok, great12:21
Mirvdidrocks: AFAIK I did all from https://wiki.ubuntu.com/DailyRelease/InlinePackaging some weeks ago already to lp:ubuntu-ui-toolkit12:21
didrocksMirv: hum, it's a native package and I don't see the bootstrap commit12:22
Mirvbut I haven't got the big picture of what the daily release needs12:22
Mirvah, so that's missing12:23
didrocksMirv: if you wait to complete the guideline to make that more clear ^ :)12:23
didrocksMirv: before doing those changes, we should diverge between a raring branch and the new trunk, isn't it?12:24
Mirvdidrocks: I have a vague memory that I asked you or someone about the need to push that "/inline-packaging", and the answer was (maybe) that not needed since it's already native package or something. but certainly at least the "Getting it ready for autolanding" (bootstrapping) hasn't been done yet12:24
didrocksMirv: would be nice to not have a native package versionning scheme, just to follow all other packages we have12:25
Mirvdidrocks: yes we should12:25
didrocksMirv: I'm happy to revisit that later on, but at least, that enables to have real "upstream" tarballs12:25
didrocks(without the debian/ dir)12:25
didrocksMirv: apart from that, just had a quick look, everything's fine :)12:26
didrocksMirv: once done, just ping me, we can add this together to the S branch ;)12:26
Mirvok, I'm pushing the raring branch first, then preparing a merge request for bootstrapping and non-native package12:27
didrocksMirv: excellent, you are creating a raring series?12:28
Mirvdidrocks: yes12:28
didrockssweet :)12:28
mhall119sergiusens: have we put out a call for testing the raring images?12:38
=== _salem is now known as salem_
=== dandrader is now known as dandrader|afk
mesqhi nik9013:12
nik90mesq: I just wanted to let you of the update to the bug report regarding the milliseconds when it was marked won't fix..13:13
nik90but you fixed that :)13:13
nik90it is being worked on by alessandro13:13
nik90mesq: although when can one expect the visual designs for clock to be announced?13:14
mhall119popey: ping13:15
nik90mhall119, popey: were you able to run the autopilot test for clock-app successfully? If yes, I can merge the branch by renato into trunk.13:16
mhall119nik90: I haven't tried it13:17
mhall119nik90: what's the branch?13:17
nik90mhall119: https://code.launchpad.net/~renatofilho/ubuntu-clock-app/autopilot/+merge/15798213:18
mhall119nik90: hmmm, I'm getting merge conflicts13:21
nik90mhall119: merge conflicts..ah I see it. I think that's because the branch is behind the recent trunk commits.13:22
nik90I will ask renato to update branch13:22
nik90mhall119: renato told me he will update his branch in another 1-2 mins :)13:23
mhall119ok, it's not a bad conflict13:24
mhall119nik90: how do I run the autopilot tests?13:24
nik90mhall119: cd into the tests/autopilot folder13:25
nik90mhall119: and then run "autopilot run clock_app"13:25
=== dandrader|afk is now known as dandrader
renatonik90, mhall119 done13:27
mhall119nik90: I get an error13:28
mhall119RuntimeError: Unable to find Autopilot interface.13:28
nik90mhall119: do you have autopilot-phablet and libautopilot-qt installed?13:29
mhall119probably not13:29
nik90mhall119: those two packages from the official autopilot ppa are required..although in my comp, the tests still fail13:30
mhall119nik90: do you have a link to the PPA?13:30
nik90mhall119: 1 min..let me find it13:31
mhall119thanks, trying to get it from Raring's archives isn't working13:31
nik90mhall119: sudo apt-add-repository ppa:autopilot/ppa13:31
mhall119popey: do you have a few minutes to get me up to date on the conversation you had with the Trojita(imap client) devs?13:31
=== Robin_Watts_ is now known as Robin_Watts
sergiusensmhall119: not yet, we want to tag the quantal one before that and found a couple of bugs that don't look good on demos13:33
jaduncan_Hi, is it possible to run X11 programs in UT somehow? I was wanting to try and run desktop Chromium.13:34
=== juicyjones|away is now known as juicyjones
mesqnik90: ok, the bug is good to go i think13:35
mhall119jaduncan_: not currently, no13:35
ajalkanejaduncan_: I'm under the impression UT runs on top of Canonical's new display driver (Mir) instead of X13:35
mhall119ajalkane: it currently runs on top of Android's SurfaceFlinger, but will start using Mir soon13:36
ajalkanemhall119: ok, thanks13:36
mhall119there will be an XMir layer, capable of running X11 apps on Mir13:36
jaduncan_ajalkane, mhall119: I wondered about a VNC app13:36
nik90mesq: thnx..I have asked another developer to work on it...should be done by today hopefully13:36
jaduncan_but that's admittedly a roundabout way to do it13:36
mhall119jaduncan_: I'm sure some body could write a VNC app in Qt that will work without X1113:36
mesqnik90: visuals handover might take a while, still. it's being explored as a whole. meaning across the suite of apps.13:37
nik90mesq: ah ok..np13:37
jaduncan_mhall119: I mean to connect to the X11 server with. I think this is currently too edge case a thing to do, and I have the desktop image of Ubuntu for the N7. I think I'll just have to try and heavily cut down the memory usage on that.13:38
mhall119nik90: autopilot runs now13:46
mhall119one test, and it passes13:47
nik90mhall119: wait the clock-app test works for you?13:47
nik90mhall119: I get Traceback (most recent call last):13:48
nik90  File "/home/krnekhelesh/Code/autopilot/tests/autopilot/clock_app/tests/test_clock.py", line 49, in test_label_value13:48
nik90    self.assertThat(label.text, Eventually(NotEquals('')))13:48
nik90AttributeError: 'NoneType' object has no attribute 'text'13:48
nik90and that the test ran but Failed13:48
nik90mhall119: hmm really wierd...so to summarize this, the calculator app test works but clock app doesnt work on my comp...while the clock app test work on ur laptop13:50
mhall119test isolation problem maybe?13:50
mhall119nik90: do you have the latest SDK packages?13:50
nik90mhall119: yeah I have the latest sdk packages..running 13.04 with all updates done13:50
nik90mhall119: I told renato that I will look line by line to see what the change is between calculator and clock app tests to figure this13:51
=== juicyjones is now known as juicyjones|away
=== juicyjones|away is now known as juicyjones
popeymhall119: pong14:03
popeyyes, hangout?14:04
mhall119popey: wanted to get caught up on the conversation you has with the Trojita devs14:04
mhall119popey: I'm in one with dpm right now14:04
mhall119maybe in 15-20 minutes?14:04
ajalkanemhall119: any template we should use in source files for copyrights?14:28
nik90ajalkane: you should have a look at ubuntu-clock-app. Every qml, js, py file there includes the copyright. You can copy just that and modify it slightly like the Author's name to suit your application14:30
ajalkanenik90: Thanks14:30
rickspencer3mhall119, is there still an app developer mailing list?14:35
rickspencer3jono, do you know? ^14:36
jonorickspencer3, if there is, we don't use it any more much14:37
rickspencer3thanks jono14:37
jonorickspencer3, we generally point folks at Ask Ubuntu for support and the G+ community14:38
rickspencer3jono, ok14:38
ogra_(or the ubuntu-phone ML)14:38
rickspencer3ogra_, yeah, that's why I was asking, wondering if it was ok to support that person writing their game on that mailing list14:38
rickspencer3or if it was noise14:38
ogra_i dont think its noise ...14:39
ogra_people discuss apps there all the time14:39
rickspencer3ogra_, thanks!14:39
ogra_and popey and mhall119 make their app announcements there too14:39
davmor2rickspencer3: this one ? ubuntu-app-devel@lists.ubuntu.com14:40
ogra_is that used much ?14:41
mhall119popey: ready now?14:42
davmor2ogra_: occasionally14:42
mhall119rickspencer3: supporting people in the ML is fine, but if they ask questions in AskUbuntu it will help more people in the future14:43
shadeslayerChickenCutlass: if you're around, I wanted to let you know that I kind of made the kernel/initrd work, and by kind of I mean I can change the brightness of the screen by writing values to files in  /sys/class/backlight/14:46
shadeslayer^ writing values from the init script that is14:46
ChickenCutlassshadeslayer: very good14:47
ChickenCutlassshadeslayer: can you share your initrd14:47
shadeslayersure, but its the standard initrd that you get in a armhf chroot14:48
ogra_shadeslayer, still no success with a framebuffer console ?14:49
shadeslayerI just added a call : echo 150 > /sys/class/backlight/pwm-backlight.0/brightness : at the beginning ( after it mounts sysfs and /proc )14:49
ChickenCutlassshadeslayer: ok got it14:49
shadeslayerogra_: I haven't pursued it today, was doing some paperwork14:49
shadeslayerChickenCutlass: the kernel config on the other hand is quite different14:49
ogra_ancient stuff ...14:49
shadeslayerindeed :P14:49
ogra_paper ... i heard of that in my youth14:50
shadeslayerChickenCutlass: I used the config that's in the recovery image14:50
shadeslayerogra_: apparently still used to file for visa's14:50
ChickenCutlassshadeslayer: ok, do you have the diff?14:50
shadeslayerjust used the one from recovery, should just work to let you change the brightness :P14:50
ogra_shadeslayer, you travel to the wrong places (or from the wrong ones) :)14:50
shadeslayerogra_: rather, was born in the wrong country :P14:51
ogra_or that14:51
shadeslayerall the countries ( except a tiny tiny minority of them ) want a visa for Indians14:51
shadeslayerthere should be a passport classification : international, travel to any country in any part of the world :P14:52
shadeslayerChickenCutlass: if you get fbcon going, that'd be superb :)14:56
* shadeslayer doesn't have time to look at it for a couple of days now14:56
sergiusensdidrocks: ping15:11
didrockssergiusens: pong15:22
rickspencer3mhall119, kaleo is it possible to command a Tabs to select a specific tab?15:25
mhall119rickspencer3: I think you can set a property on Tabs15:26
rickspencer3grammatically, I mean, without user input?15:26
rickspencer3mhall119, well, I want to select a tab on a buttons onClicked event15:26
sergiusensdidrocks: https://code.launchpad.net/~ps-jenkins/phablet-tools/latestsnapshot-0.9daily13.04.15ubuntu.unity.next-0ubuntu1/+merge/15883215:27
mhall119rickspencer3: you can probably set selectedTab or selectedTabIndex on your Tabs component15:27
sergiusensdidrocks: that's my problem... between that and the prev one there are linked bugs and all but the changelog does not reflect that15:27
mhall119that would, I assume, fire an onSelectedTabChanged handler of some kind to actually change the UI to the new tab15:27
didrockssergiusens: is it before rev 79 or after that one?15:27
rickspencer3mhall119, ack, selectTab is read only, but select tab index works15:28
sergiusensdidrocks: 78 and 80 are daily builds, 79 has a bug linked15:28
didrockssergiusens: oh right, I don't list bugs on purpose for feature ppa15:29
didrockssergiusens: stuff that doesn't land on distro15:30
didrockssergiusens: they will be listed once they land on distro15:30
didrockssergiusens: it's part of the big discussion of "what to do with feature branch"15:30
mterryfginther, is this a known issue: ?15:32
fginthermterry, I've never seen it before15:33
mterrydidrocks, ^15:33
fginthermterry, is this a jibel scritp?15:33
mterryfginther, so you landed the quantal CI branch I see15:34
didrocksmterry: no, this is me, I fixed it and didn't rerun it as I thought we'll have the autopilot job fixed15:35
didrocksmterry: which isn't the case15:35
mterryfginther, what PPAs does that use?  (both as source and destination)15:35
* mterry realizes it's probably in config15:35
didrocksfginther: I think we should really discuss about your CI configuration at the sprint15:35
didrocksfginther: I see a lot of duplication and it's making the configuration file horrible to read15:35
fginthermterry, I just restored it to match the existing jobs we had on the ci jenkins. some other job updates were needed15:35
fginthermterry, so the ppas were the old ones until I can get the 'ppa:' value to work15:36
mterryfginther, but notably, not the daily-build-next PPA I see.  So we are going to get out of sync (like the HUD in daily-next is 2.0, but HUD in phablet-team is 1.0)15:37
fginthermterry, I need to examine this for a moment15:39
sergiusensdidrocks: well we probably should or people are going to start complaining15:40
didrockssergiusens: it's just a flag, but this has consequences when upload to distro, that's why I want that to be discussed before15:41
=== pete-woods1 is now known as pete-woods
didrockssergiusens: and notice that bugs won't be closed, just listed15:41
didrockssergiusens: what is closing the bugs is launchpad15:41
didrocksduring an upload to distro15:41
didrocksas there is no "touch" product15:41
didrockssergiusens: if you want, we can discuss that tomorrow over a hangout15:41
sergiusensrsalveti: that's the explanation ^^15:41
sergiusensdidrocks: yes please... we create changlogs for everyday and construct the weekly communication with the changelog entries15:42
didrockssergiusens: ok, discussing that tomorrow? I won't have time today (in a meeting and then leaving)15:43
didrockssergiusens: I need to rethink, there were multiple good reasons to not list them15:43
didrockssergiusens: so I need to draw a map :)15:43
sergiusensdidrocks: yes, let's hangout tomorrow15:43
sergiusensdidrocks: it might turn out to be acceptable loss until we land in distro15:43
=== francisco is now known as Guest86369
didrockssergiusens: yeah, I want to ensure I'm listing every consequences, but for that, I need a fresh brain ;)15:44
rsalvetisergiusens: right15:45
rsalvetisergiusens: the problem is telling what actually changed in an automatic way15:46
rsalvetias the changlog will be kind of useless15:46
=== jhodapp is now known as jhodapp|bbiab
mterrysergiusens, were you working on the branch to add autopilot dep packages to the cu2d configs?15:49
didrocksogra_: btw, from your image you are shipping the ffmepg codec15:50
didrocksogra_: which is not very cool, can you help sil2100 and renato on that one?15:51
ogra_didrocks, how do you mean ?15:51
mterrydidrocks, do you know what happened to the branch to add autopilot dep packages to the cu2d configs?15:51
ogra_didrocks, which image ?15:51
didrocksogra_: ffmepg is shipped through the media-app15:51
* ogra_ isnt aware of any ffmpeg 15:51
didrocksgstreamer0.10 ffmepg codec15:52
ogra_didrocks, well, i dont have anty access to the OBS system, talk to sergiusens or rsalveti ... i'm working on the distro builds15:52
ogra_unless they finally switched to use our seeds15:52
sil2100Adding this to auto-join15:52
fginthermterry, I'm working on changes to add the stack ppa, so hopefully we'll only be out of sync for another day or two.15:52
didrockssergiusens: rsalveti ^15:53
mterryfginther, ok15:53
didrocksogra_: I thought you have deep knowledge of the image, sorry :)15:53
ogra_if thats in our seeds i'll happily change it though :)15:53
ogra_didrocks, not of the black hole that the OBS image build system is15:53
ogra_didrocks, ask me anything about official builds :)15:53
didrocksogra_: I think it's a dep of the media-app15:53
didrocksisn't it sil2100?15:54
ogra_didrocks, i think thats actually jhodapp|bbiab country15:54
didrockssil2100: yep, what's pulling it?15:54
sil2100didrocks, ogra_: the mediaplayer-app package has a hard dep on it15:55
ogra_right, jhodapp then15:55
sil2100I poked renato about that, since he's one of the devs there, so I had a chat about the usage of that15:55
sil2100And as mentioned on a different channel, they're using it for thumbnail creation for now because of the various codecs it supports15:56
ogra_if it is only used for that, whats wrong with having it ?15:56
ogra_or does it get in your way anyhow ?15:56
ogra_(seems to be fine license wise)15:57
sil2100Well, it's in universe, so we'll be pulling universe deps for a main package - also didrocks mentioned that the license seems to be confusing?15:57
ogra_nearly everything of touch is currebntly in universe :)15:57
didrocksogra_: we can't ship it, it's having some codecs that are legally weird15:58
didrocksogra_: it's like vlc15:58
ogra_and if it is in universe the lic3ense needs to be fine15:58
ogra_didrocks, if they were actually legally weird that package would be in multiverse15:58
ogra_unless someone made a fatal error nobody noticed over the last years15:59
didrocksogra_: do you mind checking with the release/security team? for me always ffmepg always was -> danger15:59
didrocksogra_: that would be helpful, I'm not fully knowledgeable about this, but seb128 confirmed as well15:59
seb128ogra_, the issue is not license, it's patents16:00
ogra_well, i'd like to leave it to the guys that depend on it ... after all they will need a MIR in S16:00
ogra_seb128, aha ! thats different :)16:00
sil2100Same here, not really super knowledgable about that,16:00
seb128ogra_, still a good reason to not have it on the CD16:00
ogra_well, tallk to jhodapp ... he maintains the media stack afaik16:01
sil2100ogra_, didrocks: maybe we could ask the mediaplayer-app guys to rewrite it that it would Suggest -ffmpeg -> if not installed, no thumbnails would be available16:01
ogra_(but really dont worry about usniverse vs main atm, that will only become relevant once everything can be built from the archive)16:01
didrockssil2100: ok, let's add everything, can you add a big warning for the mediaplayer-app in the spreasheet and starts the conversation?16:02
ogra_(jhodapp seems to be back  soon, i saw him saying ~1h in another channel, just wait fior him)16:03
ogra_might be it is easy to just drop it ... who knows16:04
* ogra_ has no clue about multimedia :) i'm just the image build bitch 16:04
sil2100As mentioned before, I had a talk with renato and he said something like: "maybe I can use other gst codes but I am not sure, and depends of what kind of video the user want to play"16:04
ogra_well, dont we support video thumbnails in the desktop today ?16:05
ogra_what does nautilus use atm ?16:05
ogra_seb128, ^^16:05
seb128ogra_, totem-video-thumbnail16:05
ogra_and what does that use in the back ?16:06
ogra_there must be soem gst16:06
ogra_i think the mediaplayer should just use what we use today ... so wont need a MIR etc16:06
ogra_not sure how much work that is16:07
didrockssil2100: ok, approved, let's deploy that tomorrow, ok?16:11
sil2100didrocks: oh, ok ;) Let's see if we can get it removed as well16:17
didrockssil2100: indeed :)16:17
didrockssil2100: thanks for the note on the spreadsheet :)16:17
tintinhi frnds16:21
rickspencer3mhall119, bzoltan1 is there a magic setting that I need to debug QML?16:22
tintinhello.. anybody thre16:22
rickspencer3hi tintin16:22
tintinhi rick16:22
tintinis it possible to instll ubuntu mbl on n 900?16:22
mhall119!devices | tintin16:23
ubot5tintin: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices16:23
mhall119if the n900 is listed there, there should be a link to instructions for installing Ubuntu Touch on it16:24
=== mhall119 is now known as mhall119|lunch
=== dandrader is now known as dandrader|lunch
tintinrick..its nt listd there16:35
tintintx for the help16:35
ogra_the n900 is definitely not powerful enough16:39
=== jhodapp|bbiab is now known as jhodapp
jhodappogra_, back16:45
ogra_jhodapp, sil2100, seb128 and didrocks had some concerns about using ffmpeg16:45
ogra_seems there are possible patent issues with using it16:46
sil2100Regarding mediaplayer-app16:48
jhodappogra_, yeah, as someone above mentioned, we're using it for creating previews16:48
jhodappogra_, however, we have nearly every license paid for and officially licensed is what I was told16:49
mptJust posted initial design for accessing SIM services: https://wiki.ubuntu.com/PhoneApp#Settings16:49
ogra_jhodapp, well, talk to the guys complaining :)16:50
jhodappogra_, lol...and this might get tricky...there could potentially be that we can ship certain things with a phone or tablet but not on the desktop16:51
sil2100jhodapp: maybe we could somehow make it a loose-requirement?16:52
sil2100i.e. if it's there, use it - if not, maybe disable previews or use something different?16:53
jhodappsil2100, I can't answer that, that's up to the design team16:54
=== arthurm is now known as artmello
bzoltan1 rickspencer3:  not as I know. Just use the "Debug" menu items... you can set a breakpoint to your code and do the normal debugging16:58
sil2100jhodapp: who from design should I poke for some answers? Is JohnLea the right person? ;)16:58
rickspencer3bzoltan1, I clicked on projects in QtCreator and changed the debugger port to a different port, and it started working16:58
rickspencer3sil2100, what's it concerned?16:58
bzoltan1rickspencer3:  that is good :) for me it works out of box16:59
jhodappsil2100, I'd poke Calum16:59
rickspencer3bzoltan1, are there any XmlListModel mavens?16:59
sil2100rickspencer3: hi! So, the thing is that mediaplayer-app currently is using gstreamer0.10-ffmpeg to generate the previews (thumbnails), while we would prefer not to require gstreamer0.10-ffmpeg if we are to ship the mediaplayer-app and others by default17:00
rickspencer3I'm struggling to make what looks like simple xml parsing to work17:00
jhodappsil2100, and if necessary, he'll poke someone else...certainly from a technical perspective your idea is possible, but we shouldn't just try to solve this via technical means in my opinion as it is a legal problem to solve17:00
sil2100Since we would basically have to ship gstreamer0.10-ffmpeg by default - so I've been wondering if we could somehow disable thumbnails or something when it's not installed17:00
sil2100jhodapp: ok, thanks!17:00
rickspencer3sil2100, calum and johnlea are both on my list of contacts for core apps17:00
bzoltan1rickspencer3:  zsombi I would suggest to ask... but he is off already17:01
rickspencer3so, yeah, what jhodapp said17:01
sil2100rickspencer3: thanks, will poke them ;)17:01
jhodappthanks rickspencer317:01
* rickspencer3 shakes fist at XmlListModel17:02
ogra_for convergence it needs to be the same on phone and desktop17:02
rickspencer3it easy when it works, but when it doesn't work, it's really hard to debug17:02
jhodappogra_, exactly17:02
ogra_so we need a solution thts legal17:03
rickspencer3kaleo, bzoltan1 the Tabbed application project template is awesome17:03
jhodappogra_, supposedly there's a list of the codecs that Ubuntu can legally ship somewhere that we have officially licensed...I haven't been able to find that17:03
sil2100jhodapp, ogra_: could you guys take a look into that?17:05
ogra_theoretically all thats in main is freely usable17:05
t1mpjppiiroi1en: ^ :)17:05
t1mprickspencer3: I think jp did those templates17:05
rickspencer3jppiiroi1en, o/17:06
jhodappsil2100, look into what part exactly?17:06
t1mpjppiiroi1en: indeed, I'm checking it now, and great stuff to get started with :)17:07
sil2100jhodapp: if we can freely and legally use gstreamer0.10-ffmpeg with no problem (and the codecs that are provided by it)17:08
jhodappsil2100, yeah, it's on my list of tasks somewhere :)17:08
sil2100I'll try to poke design about possible workarounds tomorrow, as I see everyone is EOD already17:09
jhodappsil2100, this is always tricky, wish it wasn't17:09
sil2100jhodapp: big t17:09
sil2100big thanks17:09
sil2100(unexpected enter press!)17:09
* rickspencer3 gives up and uses json17:13
jhodappsil2100, np17:17
=== SkavenXXI is now known as SkavenXXI-[OFF]
=== mhall119|lunch is now known as mhall119
=== dandrader|lunch is now known as dandrader
mterryWhat is the correct way to run autopilot tests in trunk?17:35
mterry(In unity/phablet trunk I mean)17:35
mterrysil2100, Saviq ^17:35
mzanettimhall119: hey! Whats the proposed way of publishing packages for my apps if the code is not hosted on launchpad?17:39
mzanettimhall119: too early to really release something. but I'd like to have at least a ppa where snapshots can be easily installed on the devices17:40
pmcgowanmzanetti: do you have a debian source package?17:40
pmcgowanthere is an apps ppa17:40
mzanettipmcgowan: yes... I have packaged stuff up17:40
rsalvetiogra_: I think the issue with ffmpeg is just that we might not want it in main due security and maintenance issues17:40
rsalvetibut I believe the license should be ok17:40
mzanettipmcgowan: should I just dput it to the apps ppa?17:41
pmcgowanmzanetti: PPA is https://launchpad.net/~ubuntu-touch-coreapps-drivers/+archive/collection17:41
pmcgowanI'd say so17:41
ogra_rsalveti, well, desktop team complained about possible patent issues17:41
mzanettipmcgowan: ok, I'll give it a shot, thanks17:41
rsalvetiogra_: right17:41
rsalvetiogra_: that's hard to fix, and don't think we have any replacement for that17:41
rsalvetito be used in a clean and patent-free way17:42
ogra_but we need something that works on the desktop as well17:42
pmcgowantrying to remember this conversation17:42
rsalvetiwe're probably violating patents all around :-)17:42
mhall119mzanetti: if you can give popey or I link to the source package, we can upload it to the PPA17:42
pmcgowanis the issue only related to certain decoder implementations?17:42
rsalvetihello world probably violate patents17:42
pmcgowanI though ffmpeg itself was ok17:42
ogra_yeah, looking forward to be sued by apple :P17:42
rsalvetithought so as well17:42
ogra_pmcgowan, gstreamer-ffmpeg is bundled17:42
pmcgowanI assume thats just a plugin for gstreamer17:43
ogra_it contains a huge amount of en/decoders17:43
ogra_it is17:43
ogra_apt-cache show gstreamer0.10-ffmpeg17:43
rsalvetijhodapp: is there a way to not use ffmpeg?17:43
popeyffmpeg has been discussed by the techboard in the past17:43
ogra_look at that output and cry :)17:43
mzanettimhall119: I have pushed the debian package structure to the git repository. so all it takes is to clone it and run debuild somehow on it. is that enough for you or do you need the already built source package?17:44
rsalvetipopey: and what was the conclusion?17:44
jhodapprsalveti, probably, I didn't design the preview code17:44
ogra_rsalveti, totem-thumbnailer apparently works on the desktop today ... i guess if our mediaplayer can mimic what it does it would be ok17:44
rsalvetiogra_: right17:44
pmcgowanrsalveti: ogra_ why are we talking about ffmpeg anyway17:45
rsalvetijhodapp: who is taking care of that? renato?17:45
pmcgowanneeds to read back17:45
ogra_pmcgowan, because we depend on it in phablet :)17:45
rsalvetipmcgowan: needed to generate thumbnails at the media-app17:45
ogra_pmcgowan, mediaplayer-app uses it for thumbnailing17:45
popeyi thought we'd sufficiently stripped ffmpeg of the problematic items17:45
popeywhich is why we have so many libav* packages and libav*-extra ones17:45
ogra_popey, its sufficient for universe17:45
pmcgowanpopey: thats definitely possible17:45
popey(it's libav these days, not ffmpeg fwiw)17:46
ogra_popey, but that doesnt make it patent free or easy to maintain security wise17:46
jhodapprsalveti, yes17:47
rickspencer3mhall119, hey, can you please paste me a link to your reddit app code17:58
rickspencer3I am terrible at making nice looking ListViews :/17:58
nik90rickspencer3: https://code.launchpad.net/~mhall119/+junk/uReadIt18:00
rickspencer3thanks nik9018:00
mhall119rickspencer3: https://code.launchpad.net/~mhall119/+junk/uReadIt18:01
mhall119ah, nik90 beat me to it18:01
rickspencer3thanks mhall11918:01
nik90mhall119:  hehe18:01
mhall119rickspencer3: there's some code that isn't committed yet, around HUD and sharing18:01
rickspencer3mhall119, ok18:01
rickspencer3I'm just looking for help formatting a ListView of articles and stuff18:01
rickspencer3I'm working on a Yahoo! local app18:01
rickspencer3I have all the data and the tabs set up18:02
rickspencer3I just can't make it look good at all :)18:02
mhall119rickspencer3: look at ArticleListItem.qml then, it's my derived component18:02
rickspencer3that's right where I am ;)18:02
mhall119mzanetti: that should be enough, where is the git repo?18:04
mzanettimhall119: http://gitorious.org/xbmcremote/xbmcremote18:16
sergiusenspmcgowan: rsalveti ogra_ so I have http://iso.qa.ubuntu.com/qatracker/milestones/243/builds/42217/testcases18:39
sergiusenssame for mako, grouper and maguro18:39
sergiusensgiven those two prior quantal bugs, what's the plan?18:39
pmcgowansergiusens: what is that exactly18:40
pmcgowanthe smoketest results?18:40
sergiusenspmcgowan: the community tests18:40
sergiusenspmcgowan: not run yet though18:40
ogra_sergiusens, looks fine ... will it be attached to the pending/ current/ moving ?18:41
sergiusenspmcgowan: with hand holding steps http://iso.qa.ubuntu.com/qatracker/testcases/1543/info18:41
ogra_or is that just to give the community a checklist18:41
sergiusensogra_: today it's a call for test... then it's going to be attached to whatever practice is already used in distro but following a 2 week cadence https://wiki.ubuntu.com/Testing/Cadence18:42
sergiusensogra_: we need to also talk about /pending /current with plars18:42
ogra_in distro nothing is actually used anymore since we expect automated testing :)18:42
ogra_the iso tracker is for manual community tests18:42
=== XenGi_ is now known as XenGi
sergiusensogra_: ack..18:43
sergiusensogra_: well I'll find it hard to test telephony et.al. in an automated fashion18:43
sergiusensogra_: best is to crowdsource18:43
mhall119not there's an interesting idea, I suppose you could connect a bit of testing hardware to a modem and have it listen for a call18:44
ogra_sergiusens, ++18:45
mhall119or just pick a number at random from the company directory, call it and say "press 1 if you can here this, otherwise pres 493829404817849e7"18:45
sergiusensmhall119: heh... how about "Network Busy", or "Missed Call"18:46
sergiusensmhall119: that's just telephony, then there's the bug I found yesterday, on manta, after a photo is taken, the view is inverted when in landscape18:46
plarssergiusens: don't forget 911 testing :)18:46
plarssergiusens: actually, from what I've read, that's not a *huge* ordeal to setup, provided you have a valid reason (and we do)18:47
mhall119sergiusens: camera isn't even working on grouper18:47
plarsmhall119: that's been the case for quite a while18:47
mhall119I know18:47
sergiusensmhall119: none of the core devs have a grouper device, that's why ;-)18:47
sergiusensplars: mhall119 ogra_ I guess I am biased to what I read in How Google Test Software when the talk about Android...18:49
plarssergiusens: which part exactly?18:50
mhall119sergiusens: I'm more than happy to help the core devs with testing on mine18:50
sergiusensplars: the part where everything is automated, except Android testing... it's all manual18:50
mhall119you can promote me to a core dev, if you think that'll speed things up :)18:50
plarssergiusens: not really *all* of their android is manual is it?18:52
sergiusensplars: let me pull up my kindle annotations18:52
sergiusensplars: http://ubuntuone.com/0Mo3ZXxiWryvphgCFUAa1o http://ubuntuone.com/25Q65nHJgM0hqmYRoG6eKT http://ubuntuone.com/557JZkkmb9Q5bE7vuY5J4I19:06
=== salem_ is now known as _salem
renatopmcgowan, jhodapp, rsalveti about the ffmepg discussion I can try replace it with others codes19:13
jhodapprenato, using Qt to do the still image generation?19:13
renatojhodapp, do you think that qtmultimedia is ready to do that?19:13
renatojhodapp, you can try use others gstreamer plugins19:14
jhodapprenato, I'm not sure, never tried such a thing19:14
renatojhodapp, we tried it in the spring in Boston, do you remember?19:14
jhodapprenato, yes19:15
renatojhodapp, but for some reason android api did not work19:15
jhodapprenato, oh that's right, I forgot about that19:15
mhall119fginther: is there a way to tell Jenkins not to auto-land code into a branch if the author hasn't signed the CLA?19:15
jhodapprenato, the map() function19:15
renatojhodapp, the quick way is try different gst codes19:16
jhodapprenato, yeah, I'm not going to have time to try that any time soon, you have any spare cycles?19:17
sergiusensmhall119: those don't get built by default...19:17
renatojhodapp, yes I can do that right know19:17
mhall119sergiusens: ?19:17
jhodapprenato, cool19:17
sergiusensmhall119: non CLA committers19:17
renatojhodapp, the problem is that we can not use multiverse packages. correct?19:17
fginthermhall119, not at this time, please see: https://bugs.launchpad.net/jenkins-launchpad-plugin/+bug/113442819:18
ubot5Error: launchpad bug 1134428 not found19:18
mhall119thanks fginther19:18
mhall119sergiusens: I was asking in reference to the Core Apps projects19:18
jhodapprenato, apparently universe19:18
jhodapprenato, has to come from main19:18
sergiusensmhall119: yeah, I guess you want to push for this: https://bugs.launchpad.net/jenkins-launchpad-plugin/+bug/113442819:18
renatook lets see which gst codecs we have on the main repository19:18
mhall119fginther: sergiusens: LP says there's no such bug19:19
fginthermhall119, oh, it's private19:19
sergiusensmhall119: try again19:19
sergiusensmhall119: it was private, no need for it to be19:19
mhall119thanks, is that being worked on?19:20
fginthermhall119, I don't see any branches for it yet19:21
fginthermhall119, we'll see if we can get someone working on this. I understand it's a big deal for the core apps19:22
mhall119thanks fginther19:22
mterrymzanetti, thanks for the lightdm branch review!  I've updated it now to fix issues19:34
mzanettimterry: ack19:34
mterrymzanetti, so why can't we launch env btw?19:35
mterrydoes launch_test_application inspect arg[0] and do something with it?19:35
mzanettimterry: maybe just env would have worked but not "env" as in that case it expects an executable called env. don't know exactly whats happening on the inside of autopilot19:38
mterrymzanetti, env should be a valid executable.  But whatever, I fixed it another way19:39
mzanettimterry: oh... right... didn't know that... but in that case autopilot expects env to publish the autopilot D-Bus interface19:49
mterry mzanetti, but I figured it would via its sub-executable...  ::shrug::19:50
mzanettimterry: probably it should, yeah... you might want to report a bug.19:50
mterrymzanetti, launch_test_application might very well be inspecting its arguments though19:52
mterrymzanetti, that comes from Qt, didn't dive into its exact behavior19:52
=== _salem is now known as salem_
moocow1452Hello everybody!20:37
sergiusensawe_: building the whole thing for the galaxy s and going to report back20:54
=== juicyjones is now known as juicyjones|away
=== salem_ is now known as _salem
FelipeDuarteHi guys.21:47
OrangerHi FelipeDuarte21:48
FelipeDuarteCan I install ubuntu touch in my motorola xoom verizon model? I can patiente for my english, I still learn21:48
OrangerApparently yes : https://wiki.ubuntu.com/Touch/Devices21:51
* genii-around looks around for smartboyhw21:52
FelipeDuarteThanks Oranger. :)21:57
OrangerFelipeDuarte ;)21:58
OrangerFelipeDuarte: Report all bugs you will find !21:58
FelipeDuartesure :)21:58
moocow1452Does this mean anything to anyone?22:05
moocow1452build/core/combo/select.mk:60: build/core/combo/TARGET_linux-armv7-a-neon.mk: No such file or directory22:05
=== davidm` is now known as davidm
SonikkuAmericaDoes Ubuntu Touch's calling feature only work with GSM phones thus far?22:20
rsalvetiSonikkuAmerica: yup22:20
=== SkavenXXI-[OFF] is now known as SkavenXXI
SonikkuAmericaDang... a friend of mine might've been interested... he has a Razr HD... but it's a CDMA phone. (Verizon)22:21
mhall119yeah, being on Verizon sucks for phones22:21
* mhall119 knows22:21
SonikkuAmericaAT&T is America's only reputable GSM carrier (T-Mobile is mainly HSPA+ now)22:23
SonikkuAmericaAnd America thinks they stink.22:24
rickspencer3mhall119, is there any kind of location service built into the SDK yet?22:29
=== XenGi is now known as XenGi_
moocow1452Any idea what the hangup is? http://paste.ubuntu.com/5714506/22:33
moocow1452Hello, anyone know how to fix a build in the room? http://paste.ubuntu.com/5714506/22:48
achiangsergiusens: hey, if i want the latest and greatest bits for a maguro, is it simply just calling phablet-flash?22:48
achiangsergiusens: would that be raring?22:48
achiangrsalveti: ^^22:48
rsalvetiachiang: raring is not yet the default22:49
rsalvetiit'll grab the latest, but still based on quantal22:49
rsalvetiwe're about to switch, but still need to fix a few issues22:49
achiangrsalveti: any benefit going to raring? or safer to just stick w/quantal for now?22:49
rsalvetiachiang: we want to be closer to the archive as we want to start pushing our stuff to s-series once it's open22:50
achiangrsalveti: makes sense, thanks22:50
sergiusensachiang: phablet-flash if you have the raring package... the daily releases sort of messed up my releaser for the other series and need to do it manually (haven't done it still)22:51
achiangsergiusens: np, i'll just grab quantal22:53
mhall119rickspencer3: there's QtLocation from upstream, I'm not sure how it works though, or if the current images support it23:02
rickspencer3hey mhall23:03
rickspencer3are their QML bindings for it?23:03
mhall119rickspencer3: I see qtdeclarative5-qtlocation-plugin in raring, but can't find the docs on http://qt-project.org/doc/qt-5.0/qtdoc/modules-qml.html23:05
rickspencer3mhall119, yeah23:06
rickspencer3let me poke at it23:06
Orangermhall119: I made a launcher in C++ for docviewer, it's temporary.23:07
mhall119rickspencer3: bzoltan1 might know where docs are23:07
mhall119Oranger: awesome!  Not too difficult to get it passing file paths to the QML I hope23:07
rickspencer3mhall119, well, I'm mostly just skeptical that it's supported yet23:07
Orangermhall119: Oh and I found a way to open plain text direclty in QML with XMLHttpRequest, but for other files we will need to use C++ to open the files23:08
Orangermhall119: Hum actually I can't achieve this, I don't know why but it don't work. Maybe i'm doing something bad I don't know23:08
Orangermhall119: Me idea was to create a new property called commandLineArgument from C++ to QML23:09
OrangerBut this property seem to be undeclared for QML... so I have to work on it23:10
mhall119Oranger: sounds like a good approach23:15
mhall119XMLHttpRequest probably won't work for local files23:15
Orangermhall119: XMLHttpRequest It is working for me ^^23:16
mhall119oh, I misunderstood then23:18
=== SkavenXXI is now known as SkavenXXI-[OFF]
Orangermhall119: But in the futur we will not use it :/ Only c++ can open file, not QML+JS23:19
rickspencer3Oranger, mhall119 I think the plan is for the SDK to provide a safe file browser23:24
rickspencer3like a service that lets applications open files23:25
mhall119rickspencer3: yeah, I believe jdstrand and his team were going to do something along those lines23:25
mhall119I didn't think it was going to be built into the SDK though23:25
mhall119jdstrand: thoughts?23:25
rickspencer3it's part of the app isolation specification23:25
mhall119right, though when it was initially discussed we were still thinking in terms of python and Gtk23:26
rickspencer3indeed, but I am pretty certain they have moved on23:26
mhall119this actually sounds like a better opportunity for us, since we don't have to get people to un-learn how to access files in QML23:26
Orangerrickspencer3: Oh good23:27
mhall119Oranger: I'll have a chat with jdstrand about long-term solutions, in the mean time we can put this in place for the doc viewer23:27
mhall119and change it once something better is available23:27
Orangermhall119: Ok, sound good23:28
OrangerSo, i'm sorry but I have to go23:28
mhall119Oranger: I'm wondering if it would be easier to sub-class MainView in Qt, add the command line properties and methods for opening/reading files onto that, and use the sub-classed component in the QML23:28
Oranger(1:28 am here ouch)23:29
mhall119Oranger: no problem, we'll chat later23:29
=== moocow1452_ is now known as moocow1452
OrangerHum yes, you have a good idea mhall11923:29
moocow1452Anyone good with mkimage?23:30
Orangermhall119: But why not create a class only made for IO ?23:30
Oranger-IO +opening23:30
mhall119Oranger: it just sounds like more work than is necessary, if we're going to replace it with something else later23:31
Orangermhall119: Yes, I understand23:31
OrangerSo bye mhall119 , we'll talk later !23:32
mhall119g'night Oranger23:34

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!