| === phunyguy_ is now known as phunyguy | ||
| AskUbuntu | Ubuntu phone accessibility | http://askubuntu.com/q/329130 | 00:41 |
|---|---|---|
| siv | hello all | 00:56 |
| siv | i cant seem to figure this out.. i am trying to flash nexus 7 wifi, follow the instructions, when the evice reboot its always to "cwm-base recovery", and never unity shell. wwhat am i doing wrong ? it says autodeploy.zip not found, so i manually point it to the file on the nexus, but no success ? | 00:59 |
| Bonius | Are you doing phablet-flash? | 01:00 |
| siv | yes phablet-flash | 01:00 |
| siv | yes . | 01:00 |
| Bonius | I did my wifi nexus 7 for the first time yesterday, and I founf that you just have to wait and wait and wait | 01:00 |
| Bonius | it looks like its hung, but its actually copying stuff | 01:01 |
| siv | you mean at the cwm recovey screen? | 01:01 |
| Bonius | yeah, but the shell on your desktop where you issued that phablet=flash command will say that its copying one of the big image files (forget which one) | 01:02 |
| siv | got it ... you were right. thanks for your help . | 01:03 |
| Bonius | no problem | 01:03 |
| === chihchun_afk is now known as chihchun | ||
| === lilstevie_ is now known as lilstevie | ||
| bzoltan | mhall119: as sergiusens correctly said, the packaging is not a problem, but building the source package needs the information what is traditionally located in the debian/control and debian/rules files | 03:30 |
| DJJeff | Preparing to replace ofono 1.12phablet11 (using .../ofono_1.12phablet12_armhf.deb) ... | 03:33 |
| DJJeff | :( | 03:33 |
| DJJeff | I hope this fixes the damn 100% CPU issue on wifi only devices | 03:34 |
| DJJeff | nope it does not....... syslog full of this !@#$ http://puu.sh/3US1K.png | 03:36 |
| _lunarcold | trying to follow the porting guide, where is "breakfast"? | 03:42 |
| DJJeff | hahahahaha I ran into the same problem | 03:42 |
| DJJeff | but I ummmm didnt write down how I solved it | 03:42 |
| _lunarcold | do I need to build the huge repo that I downloaded? | 03:42 |
| DJJeff | yes | 03:42 |
| DJJeff | I think you do | 03:42 |
| _lunarcold | hm... that doesn't build because of missing kernel sources or something | 03:43 |
| _lunarcold | http://paste.kde.org/pdd565e06/ | 03:44 |
| _lunarcold | I'd prefer not ot have to do the "depreciated" steps | 03:47 |
| _lunarcold | think I found it. running "source build/envsetup.sh" makes breakfast work apparently | 03:49 |
| _lunarcold | that should really be added to the porting doc | 03:49 |
| dejello | Hello | 03:52 |
| === darek is now known as Guest4918 | ||
| _lunarcold | added information to the porting guide. there are some packages that you may need (ubuntu developer something, but the messages will walk through so I don't think I need to add) | 04:03 |
| _lunarcold | also, "bash" must be used, and you are warned if you use an alternative shell. I use zsh, but had to switch shells, is it worth adding that to the guide too? | 04:03 |
| === Namidairo`bnc is now known as Namidairo | ||
| * dejello blinks | 05:16 | |
| dholbach | good morning | 07:14 |
| === alan_g|EOD is now known as alan_g | ||
| swordfish | Hi everyone I just made some updates to the minesweeper application... Can someone update the package in the collection ppa please? | 08:10 |
| === schwuk_away is now known as schwuk | ||
| === mpt_ is now known as mpt | ||
| ev | tvoss_: is mir on grouper supposed to be working at this point? I've tried http://s-jenkins:8080/job/ubuntu-touch-phablet-image-saucy-mir/lastSuccessfulBuild/artifact/saucy-preinstalled-phablet-armhf.zip but unity crashes before clearing the bootloader image. | 08:30 |
| ev | trying to make some headway on building a test environment for hanging applications | 08:30 |
| tvoss_ | ev, help me, grouper is which nexus? | 08:30 |
| ev | 7 | 08:30 |
| tvoss_ | ev, ah, 7 is interesting in that the nvidia driver is a beast | 08:31 |
| tvoss_ | ev, best to check with kdub once he comes online | 08:31 |
| ev | tvoss_: would this work better with a 4? | 08:31 |
| ev | will do | 08:31 |
| tvoss_ | ev, @4: for sure | 08:31 |
| ev | excellent. That's the second use case I have for one now. I think that's new toy bingo. | 08:32 |
| tvoss_ | ev, yup :) | 08:32 |
| === alan_g is now known as alan_g|tea | ||
| psypher246 | hello all. Is there a limit to how many characters a mail to the mailinglist list can contain? I sent a rather large mail and it does not appear in the archive and no-one responded | 08:52 |
| === alan_g|tea is now known as alan_g | ||
| psypher246 | hello all, is the mailist list and launchpad broken at this time? | 09:04 |
| ogra_ | launchpad works for me | 09:05 |
| psypher246 | i can't see any bugs? | 09:05 |
| ogra_ | and the last mail came in 3min ago on the list | 09:05 |
| Wellark | hi guys! | 09:05 |
| psypher246 | is there a limit on the amout of characters you can have in a mail to the list? | 09:06 |
| ogra_ | a few MB i would guess | 09:06 |
| psypher246 | I have sent3 mails so far to the list and not coming through, definitely not even 1MB | 09:06 |
| ogra_ | well, did you join the team on LP ? | 09:07 |
| mfisch | tvoss_: why do we need to remove the hybris dep on non-ARM platforms? for testing? | 09:07 |
| ogra_ | i' not sure non members can send mail | 09:07 |
| psypher246 | YUP | 09:07 |
| ogra_ | (though you should have gotten an answer mail telling you that, check your spam folder) | 09:07 |
| psypher246 | nope | 09:08 |
| tvoss_ | mfisch, nope, pulling it in as a runtime dependency on amd64/i386 results in an egl/gl library being installed and set as default that is not working, shadowing the mesa egl/gl | 09:08 |
| tvoss_ | mfisch, it's less of an issue if powerd is limited to armhf/arm64 | 09:08 |
| ogra_ | tvoss_, how about intel/android based phones/tablets | 09:09 |
| sforshee | tvoss_: isn't there some better basis for this than CPU architecture? | 09:09 |
| ogra_ | we will support them some day | 09:09 |
| ogra_ | thats not gonna scale ... | 09:09 |
| psypher246 | ogra_, I am part of the team but when I go to bugs ubnder ubuntu phone team I see any and can;t search anything | 09:10 |
| tvoss_ | sforshee, well, the idea simply is: if we are not on arm, we shouldn't install hybris | 09:10 |
| ogra_ | psypher246, we dont manage bugs under that team ... | 09:10 |
| sforshee | tvoss_: but like ogra_ said, what about intel-based android targets? | 09:10 |
| ogra_ | see the channel topic :) | 09:10 |
| ogra_ | tvoss_, that will make the emulator impossible | 09:11 |
| tvoss_ | sforshee, let's solve that once we encounter it. We don't have such a target right now | 09:11 |
| ogra_ | we do | 09:11 |
| tvoss_ | ogra_, why is that? why would the emulator use hybris on i386/amd64? | 09:11 |
| ogra_ | and we have many requests from padfone users and razor users | 09:11 |
| ogra_ | tvoss_, because it is an android emulator and you want to emulate phone input etc | 09:12 |
| tvoss_ | ogra_, still, we shouldn't install hybris on non-arm systems | 09:12 |
| ogra_ | we should | 09:12 |
| ogra_ | we shouldnt install it on non android systems | 09:13 |
| psypher246 | ogra_, thx got it. do you know of an already open bug about the CPU leak, overheating issues on a nexus 4? | 09:13 |
| sforshee | tvoss_: if it's using androids hal we're going to need hybris, no? | 09:13 |
| tvoss_ | ogra_, true, but our packaging system does not allow for that right now | 09:13 |
| tvoss_ | sforshee, true, see my comment to ogra's remark | 09:13 |
| ogra_ | psypher246, there are a few ... unity8 running at 60% CPU usage constantly, some races with the android container etc | 09:13 |
| ogra_ | tvoss_, ?? | 09:14 |
| ogra_ | our packaging system clearly allows that | 09:14 |
| ogra_ | it is designed for this :) | 09:14 |
| sforshee | tvoss_: well how functional does powerd need to be on non-arm platforms? If I just stub out the stuff using hybris then we'll lose sensor and input support. | 09:14 |
| ogra_ | make hybris depends on lxc-android-config ... which is only installed on android based systems | 09:14 |
| tvoss_ | ogra_, hmmm, how can I then express that I only want to install a build dependency on something having an android core | 09:15 |
| ogra_ | our image/install system doesnt allow it atm ... the package system clearly does | 09:15 |
| tvoss_ | ogra_, then I phrased it wrongly, but the problem is still there | 09:15 |
| ogra_ | tvoss_, we will need to build separate images with the current image design .... (unless we decide to have an actual installer that could judge what needs to be installed and what not) | 09:16 |
| ogra_ | the problem is different than you anticipate :) ... and it isnt arch bound | 09:17 |
| tvoss_ | ogra_, I know it is not arch dependent, but it is the only marker we can leverage right now | 09:17 |
| tmoenicke | mzanetti: ping | 09:17 |
| ogra_ | it is image bound ... once we build x86 images it is the job of the image creator to make sure we have the right bits in the right images | 09:17 |
| ogra_ | since we currently dont build x86 images at all its a moot point | 09:18 |
| ogra_ | and the first x86 ubuntu touch image we will build wont be for PCs | 09:18 |
| tvoss_ | ogra_, but we have issues right now, installing anything from the rdepends here http://pastebin.ubuntu.com/5954397/ breaks gl acceleration | 09:18 |
| ogra_ | (but for the android emu) | 09:18 |
| tvoss_ | ogra_, as it pulls in hybris, which in turn installs alternative gl/egl libs as defaults | 09:18 |
| ogra_ | well, dont install hybris then ... but on proper basis ... not just blantly based on arches | 09:19 |
| psypher246 | ogra_, do you know if there is a bug for the broken keyboard in the terminal? | 09:20 |
| ogra_ | (amke it depend on lxc-android-config, make sure lxc-android-config isnt a hard dep on x86) | 09:20 |
| ogra_ | psypher246, whats broken there ? works for me on all devices ... | 09:21 |
| ogra_ | you can check the maliit bugs on launchpad though | 09:21 |
| psypher246 | ogra_, KB is about 10 pixels too low so i only see half of the last line, thats if the kb works at all when changing to apps and back again | 09:22 |
| tvoss_ | ogra_, are our images switched to the lxc-android-config, yet? | 09:22 |
| ogra_ | psypher246, well, i'd say the terminal "underlaps" (is that a word) the kbd | 09:23 |
| psypher246 | ogra_, no idea what that means :) | 09:23 |
| ogra_ | psypher246, since the kbd works fine with all other apps , the widget drawing the terminal has a bug | 09:23 |
| tvoss_ | ogra_, but hold on: how would I make lxc-android-config an optional build or runtime dep in debian/control? | 09:23 |
| ogra_ | tvoss_, all arm images use it atm ... ubuntu-touch doesnt depend on it on x86 currently | 09:24 |
| psypher246 | ogra_, no sorry i just checked it's all apps | 09:24 |
| ogra_ | (if it does, *that* would be tthe bug) | 09:24 |
| ogra_ | psypher246, file it then ... | 09:24 |
| ogra_ | i dont see it on all apps | 09:25 |
| ogra_ | psypher246, which device is that ? probably the UI grid value is wrong | 09:25 |
| psypher246 | N4 | 09:25 |
| ogra_ | (thats the value that makes sure there are no overlapping UI elements) | 09:25 |
| ogra_ | hmm, no that should be fine | 09:25 |
| SuperMatt | hey all, has anyone expressed any interest in creating a touch IRC app? | 09:31 |
| psypher246 | ogra_, should I be flashing the latest image everyday nor is an apt-get update/upgrade good enough? | 09:32 |
| ogra_ | SuperMatt, i would express high interest in using one :) | 09:33 |
| SuperMatt | awesome :D | 09:34 |
| SuperMatt | I've been playing around with writing an irc bot, so I would certainly be interested in making a client too | 09:34 |
| ogra_ | psypher246, apt-get update/dist-upgrade is fine as long as you dont see platform-api or libhybris in the updates ... if either of them is there you should update the android side too (i,.e. flash) | 09:34 |
| psypher246 | ok thx | 09:36 |
| ogra_ | psypher246, note that we'll be switching to image based upgrades within the next two/three weeks though (apt will be disabled, rootfs will be readonly by default and updates will happen through image diffs) | 09:36 |
| psypher246 | ah ok, so then just hook it up to my pc and do phablet flash | 09:37 |
| dholbach | sergiusens, did you find anything for MaxWallstedt's adb problem with the new phablet-flash? | 10:08 |
| sergiusens | dholbach: yes, we fixed it | 10:13 |
| sergiusens | dholbach: there's a new entry in the manifest.json for the i9xx | 10:13 |
| dholbach | is that something all ports will require? | 10:13 |
| sergiusens | dholbach: no, it's optional | 10:14 |
| dholbach | ah cool | 10:14 |
| sergiusens | dholbach: after some thought, he seemed to not have an external sdcard, so /sdcard is not mountable. We added an option to set where to land the files | 10:14 |
| dholbach | and now it works just fine? | 10:15 |
| sergiusens | dholbach: yes | 10:15 |
| dholbach | you make me a happy man! | 10:15 |
| sergiusens | dholbach: here's the install log if you care about it: http://paste.ubuntu.com/5952762/ :-) | 10:16 |
| dholbach | sergiusens, I was wondering: should we have a way to approve ports once they're working fine? like right now everybody could break their device by just trying to flash one of these: https://code.launchpad.net/phablet-image-info | 10:16 |
| dholbach | on the other hand are we going to rely on the ACK of the port/image maintainer or testers of the device as a piece of feedback anyway | 10:17 |
| dholbach | it might be nice to have some indication of "somebody tested this and it worked" :) | 10:18 |
| smartboyhw | dholbach, you should | 10:18 |
| sergiusens | dholbach: add a 'stable' or 'tested' entry in manifest.json | 10:19 |
| dholbach | sergiusens, what would "tested: yes/no" do? would we want to have a --override option in phablet-flash or something? | 10:20 |
| sergiusens | dholbach: --unstable ? | 10:21 |
| dholbach | sure, that'd work for me | 10:21 |
| dholbach | would we want to print out something like "port/image exists, but has not been tested yet, use --unstable to override."? | 10:21 |
| dholbach | sergiusens, so it wouldn't be the phablet-port-admins who bless ports/images? | 10:36 |
| sergiusens | dholbach: ah, I was hoping you would take care of the process part :-) | 10:37 |
| sergiusens | dholbach: another option is to have a specific branch with just a list of devices that are 'stable' or 'approved' | 10:37 |
| dholbach | sergiusens, well, it's simple - if it's just a matter of adding "tested: yes" into the manifest somewhere and everybody on the team can do it, then we don't need to do anything | 10:38 |
| dholbach | sergiusens, as we rely on the feedback of others anyway, I'm not sure how much sense this makes | 10:38 |
| sergiusens | dholbach: the only question is if you want a 3rd party to say it's good or just the same person | 10:39 |
| dholbach | I could imagine I'd be fine with the image maintainers to say "WFM" | 10:41 |
| === VDVsx_ is now known as VDVsx | ||
| === chihchun is now known as chihchun_afk | ||
| === MacSlow is now known as MacSlow|lunch | ||
| xnox | ogra_: phablet-flash --legacy -b is giving me a traceback. I guess I should be the one fixing it if I need it?! | 11:03 |
| ogra_ | i thought we dropped that | 11:03 |
| ogra_ | sincew we dont have the images anymore .... at least no daily ones | 11:04 |
| xnox | ogra_: well, i'm fine with the last available image. will try manually flashing it, it's just my tablet was bootstrapped for image-updates.... | 11:05 |
| === gusch is now known as gusch|lunch | ||
| ogra_ | xnox, well, i'm not sure legacy is supposed to work | 11:06 |
| sergiusens | xnox: don't fix it | 11:08 |
| sergiusens | xnox: can you paste the trace? | 11:08 |
| xnox | sergiusens: http://paste.ubuntu.com/5954689/ | 11:10 |
| sergiusens | xnox: I'll hurry in my refactor branch which works. I have a suspicion for this one though... | 11:11 |
| === nik90|Office is now known as nik90|Lunch | ||
| sergiusens | xnox: so the unmerged mega branch works ok... I'll see if I can get it in today... | 11:20 |
| xnox | sergiusens: which one is that? i'll use it locally here =) | 11:21 |
| sergiusens | xnox: https://code.launchpad.net/~sergiusens/phablet-tools/flash_change/+merge/177927 | 11:21 |
| sergiusens | xnox: ./phablet-flash cdimage-legacy -b | 11:21 |
| ogra_ | we really shoould drop legacy imho | 11:22 |
| sergiusens | ogra_: rsalveti wants me to keep it for demo's ... but I've isolated it completely | 11:23 |
| xnox | ogra_: well i'm using legacy at the moment to try bootstrapping emulator for one last time =) | 11:23 |
| ogra_ | sergiusens, phablet-flash is like windows ... backwards compatible to day one | 11:23 |
| sergiusens | ogra_: it's my enduring pain! | 11:23 |
| ogra_ | yeah | 11:23 |
| ogra_ | i dont see the benefit of letting people use buggy old and unmaintained stuff | 11:24 |
| sergiusens | ogra_: I tried it the other day, rock solid! | 11:25 |
| sergiusens | full of hacks | 11:25 |
| sergiusens | but stable :-) | 11:25 |
| ogra_ | heh | 11:25 |
| ogra_ | but not reflecting reality :) | 11:25 |
| sergiusens | xnox: hold that branch a bit, landing a fix | 11:25 |
| * ogra_ hopes on release day we will drop all this completely | 11:26 | |
| xnox | ogra_: he =) | 11:27 |
| rickspencer3 | ogra_, hmm, what to do with phablet-flash is an interesting topic | 11:27 |
| rickspencer3 | ogra_, maybe a UDS session? | 11:27 |
| sergiusens | the pain of supporting a gazillion options would make things easier to maintain | 11:27 |
| ogra_ | rickspencer3, well, if we have a release we consider stable we should definitely not support old half finished cruft with it | 11:27 |
| ogra_ | phablet-flash after release should have --current and --devel | 11:28 |
| ogra_ | nothing more | 11:28 |
| sergiusens | ogra_: and community ... | 11:28 |
| sergiusens | in some form | 11:28 |
| ogra_ | (and --current being the default makes an option obsolete) | 11:28 |
| rickspencer3 | ogra_, well, I was thinking phablet flash would only be necessary for initial phone set up | 11:28 |
| ogra_ | --community would base on --current | 11:29 |
| rickspencer3 | then image based updates going forward | 11:29 |
| ogra_ | yeah | 11:29 |
| ogra_ | well | 11:29 |
| ogra_ | depends how we support community ports | 11:29 |
| sergiusens | xnox: you want revno 168 from that branch | 11:29 |
| sergiusens | ogra_: I gave some thought for image based upgrades and for zips | 11:29 |
| ogra_ | sergiusens, i think the only issue with them is the signing | 11:30 |
| rickspencer3 | ogra_, good point, anyway, just saying, seems like a good UDS topic, invite the port authors, etc... | 11:30 |
| ogra_ | we should have an "unsafe" key for community builds | 11:30 |
| ogra_ | yeah | 11:30 |
| sergiusens | ogra_: if they flash an unprotected recovery, they're good | 11:30 |
| ogra_ | sergiusens, right, so we need to make sure they get one :) | 11:31 |
| sergiusens | ogra_: well, they are building it, we just need instructions so people can add their own keys or skip it completely | 11:31 |
| ogra_ | i would go for skip | 11:32 |
| sergiusens | and we need instructions on how to setup such server | 11:32 |
| ogra_ | and just have that as the default in the code | 11:32 |
| ogra_ | with an override for us | 11:32 |
| === schwuk is now known as schwuk_away | ||
| === alan_g is now known as alan_g|lunch | ||
| === alan_g|lunch is now known as alan_g | ||
| === gusch|lunch is now known as gusch | ||
| === jpetersen_ is now known as jpetersen | ||
| vincentbosch | Hi, I have got a question; Is it already possible to run Ubuntu Touch on the Nexus 7 2nd Gen? | 11:51 |
| === nik90|Lunch is now known as nik90|Office | ||
| === alan_g is now known as alan_g|lunch | ||
| === dholbach_ is now known as dholbach | ||
| Mirv | alex-abreu: any progress with http://pad.lv/1208355 , it's still failing&blocking? | 12:05 |
| ubot5 | Launchpad bug 1208355 in unity-webapps-qml "HUD autopilot tests failing" [Undecided,New] | 12:05 |
| === tvoss_ is now known as tvoss|lunch | ||
| === MacSlow|lunch is now known as MacSlow | ||
| === hikiko is now known as hikiko|lunch | ||
| jasbir | hi | 12:30 |
| jasbir | can any body help me out here?? | 12:30 |
| jasbir | i am trying to restore my nexus to android | 12:32 |
| jasbir | but i get an error instead | 12:32 |
| jasbir | jasbir@jasbir-Singhs-dabba:~/occam-jdq39$ adb reboot-bootloader error: insufficient permissions for device jasbir@jasbir-Singhs-dabba:~/occam-jdq39$ sudo adb reboot-bootloader error: insufficient permissions for device jasbir@jasbir-Singhs-dabba:~/occam-jdq39$ | 12:32 |
| === ritz_ is now known as ritz|zzz | ||
| alex-abreu | Mirv, yes afaik, no progress, rvr is supposed to help w/ that along w/ fginther | 12:38 |
| === dholbach_ is now known as dholbach | ||
| xnox | ogra_: i think i need to be reconstructing unflipped on emulator by hand. The stuff that makes unflipped image from android is: /data/ubuntu/ chroot + android init tweaks to kick off the container. Anything else? additional init tweaks to prevent android shell from starting? | 12:46 |
| ogra_ | xnox, i think you need to manually replicate what the zip installation does | 12:46 |
| ogra_ | during flashing | 12:47 |
| xnox | ogra_: i can't install zips, but sure, i'll follow the steps that the zip does. i guess i should diff the normal android zip vs our zip, for the unflipped system.zip (android/armel). | 12:48 |
| xnox | (in the updater scripts & our script) | 12:48 |
| ogra_ | the unzipping is identical | 12:49 |
| ogra_ | they only differ by initrd | 12:49 |
| ogra_ | (and by dropped hacks in the rootfs) | 12:49 |
| * xnox ponders if emulator runs initrd, it looks like it does direct kernel boot. | 12:50 | |
| xnox | it does have initrd. | 12:51 |
| oSoMoN | fginther: ping | 12:53 |
| === _salem is now known as salem_ | ||
| fginther | oSoMoN, pong | 12:56 |
| oSoMoN | fginther: hey, I’ve got some autopilot tests failing on the community apps jenkins instance with a "OSError: [Errno 12] Cannot allocate memory" error: https://code.launchpad.net/~osomon/ubuntu-calendar-app/refactor-autopilot-tests/+merge/178524 | 12:57 |
| === chihchun_afk is now known as chihchun | ||
| Mirv | alex-abreu: thanks, let's hope the fix is found | 12:58 |
| fginther | oSoMoN, that's interesting. I haven't seen that one before. I'll dig into and see if I can find anything | 12:58 |
| oSoMoN | fginther: thanks! | 12:58 |
| oSoMoN | fginther: it might have been a temporary issue now resolved, in which case could you please re-trigger a CI run on this MR? | 12:59 |
| fginther | oSoMoN, I'll give it a try | 12:59 |
| oSoMoN | thanks | 12:59 |
| === hikiko|lunch is now known as hikiko | ||
| davmor2 | Yeah my phone lights up and pings when I get a message woohoo! | 13:10 |
| dholbach | janimo, it looks like flashing worked for MaxWallstedt and his i9100 port now - how are things working for you? :) | 13:10 |
| janimo | dholbach, works too after device hiccups with corrupted partitions | 13:11 |
| davmor2 | dholbach, mhall119: We need a competition to create ringtones and notification pings......go make it happen ;) | 13:11 |
| AskUbuntu | Switching from Ubuntu Desktop to Ubuntu Tablet | http://askubuntu.com/q/329304 | 13:11 |
| dholbach | davmor2, you go and make it happen ;-) | 13:12 |
| dholbach | janimo, so users of the device can now just use sergiusens' phablet-flash and flash their u9200 devices? | 13:12 |
| === alan_g|lunch is now known as alan_g | ||
| davmor2 | dholbach: Okay you cover the apps queue for a couple of weeks and I'll see what I can do :D We'll see who cries first :) | 13:13 |
| dholbach | davmor2, you know what - let's do it the next time I'm twiddling thumbs ;-) | 13:14 |
| janimo | dholbach, they should indeed, but in order for it to work the latest not-yet automatically QAd ubuntu image is needed | 13:14 |
| janimo | since only those post Aug 1 have the udev rules | 13:14 |
| dholbach | ok | 13:14 |
| janimo | but according to ogra yesterday the newer images have regressions still | 13:14 |
| davmor2 | dholbach: hey that not on you have time to twiddle your thumbs, I'm complaining to Jono ;) | 13:14 |
| dholbach | great | 13:14 |
| ogra_ | dholbach, janimo, i expect it to at least take til end of the week until people have fixed their app tests | 13:16 |
| ogra_ | and before the dashboard finishes with 100% successfull tests i'm not alllowed to release an image | 13:16 |
| davmor2 | dholbach: on a more serious note, the messaging ping is really boring so having some sample sounds to flick between might be a fun way to contribute like the photos for the desktops, just an idea at any rate :) | 13:16 |
| dholbach | davmor2, the u-phone list might be a good target | 13:18 |
| davmor2 | dholbach: indeed good plan | 13:19 |
| oSoMoN | fginther: so the CI job re-ran, but the situation is not any better apparently | 13:20 |
| fginther | oSoMoN, I'm looking at some other issues on that setup now. They could be related | 13:21 |
| === schwuk_away is now known as schwuk | ||
| === alan_g_ is now known as alan_g | ||
| fginther | oSoMoN, https://code.launchpad.net/~osomon/ubuntu-calendar-app/refactor-autopilot-tests/+merge/178524 SUCCESS now | 13:45 |
| oSoMoN | fginther: awesome, thanks! | 13:46 |
| oSoMoN | fginther: there are other pending MRs that probably need a re-run too | 13:46 |
| fginther | oSoMoN, ack. I will try to run through them | 13:47 |
| uphone | hello, recently started using ubuntu touch mako on a nexus 4, question : to update the os, do i only need to push a new phablet -b ? | 13:47 |
| oSoMoN | fginther: thanks | 13:47 |
| pmcgowan | hey asac the mako results seem wacky today, need any help? | 13:48 |
| uphone | maybe not the right forum for asking questions, where should i go to do so ? ubuntu forums ? | 13:48 |
| pmcgowan | uphone, to update just do phablet-flash, the -b option will do a full wipe of data etc | 13:49 |
| pmcgowan | if you want the very latest use --pending, which is not yet QA blessed | 13:49 |
| asac | pmcgowan: talking to oSoMoN :) | 13:50 |
| pmcgowan | asac, maguro looks great, mako not so much | 13:50 |
| pmcgowan | somewhat suspicious | 13:50 |
| uphone | great ! thanks alot, I'll do it daily, wher can i report findings regarding issues and bugs, lauchpad ? | 13:50 |
| pmcgowan | uphone, yes, all the projects are in launchpad | 13:50 |
| === alan_g is now known as alan_g|afk | ||
| === tvoss|lunch is now known as tvoss|hungry | ||
| === tvoss|hungry is now known as tvoss_ | ||
| uphone | thanks again, good job, i hope to contribute once i get the hang of what info is required. | 13:52 |
| asac | pmcgowan: indeed. | 13:52 |
| pmcgowan | uphone, good to have you aboard | 13:53 |
| asac | i was rehoping that infra knows more about bug 1208494 | 13:53 |
| ubot5 | bug 1208494 in Ubuntu File Manager App " utah: YAML parsing error during filemanager-app tests" [Undecided,New] https://launchpad.net/bugs/1208494 | 13:53 |
| asac | maybe that can cause weird behaviour | 13:53 |
| pmcgowan | ah | 13:53 |
| pmcgowan | looks like the phone failures need a new release of the app | 13:53 |
| asac | doanac: plars: when awake, can you check maguro vs. mako? | 13:53 |
| asac | pmcgowan: where did you spot that? | 13:54 |
| === chihchun is now known as chihchun_afk | ||
| === alan_g|afk is now known as alan_g | ||
| asac | doanac: gema: i think we should invest in visualising the package list and the diff of versions soonish. would help blaming and explaining more efficiently it feels | 13:55 |
| pmcgowan | asac, other channel, UITK change needs a new release of the app | 13:56 |
| asac | right | 13:56 |
| asac | pmcgowan: other apps as well? | 13:56 |
| pmcgowan | asac, not sure, the others all seemed to run clean | 13:56 |
| pmcgowan | just looking at maguro results | 13:57 |
| _lunarcold | Hi, trying to build for my m7spr, and I keep getting this error | 13:59 |
| _lunarcold | make: *** No rule to make target `frameworks/base/nfc-extras/com.android.nfc_extras.xml', needed by `/home/jeff/m7spr/out/target/product/m7spr/system/etc/permissions/com.android.nfc_extras.xml'. Stop. | 13:59 |
| _lunarcold | make: *** Waiting for unfinished jobs.... | 13:59 |
| _lunarcold | this error occurs when I build with the "brunch m7spr" command | 14:03 |
| _lunarcold | I'm trying to build from the makefile and so far it hasn't thrown the error yet :/ | 14:04 |
| === _salem is now known as salem_ | ||
| sergiusens | janimo: good to know it was just a hiccup | 14:18 |
| brunoobr | Hello | 14:21 |
| brunoobr | Hello | 14:22 |
| brunoobr | anybody there? | 14:22 |
| TioBorracho | Hi everyone | 14:25 |
| doanac | asac: ack. we have that data in our .yaml files. We'd need to start storing it in the qa-dashboard and showing the diff. Not too hard, but we need to find time. I'll add it as a work-item for the dashboard | 14:26 |
| TioBorracho | Just a little question: I was testing ubuntu touch some weeks ago and the network manager didn' t have the option to connecto to EAP wifi network | 14:26 |
| TioBorracho | is there a way to do so? | 14:27 |
| mhall119 | cjwatson: bzoltan: are the click tools packages available in pre-saucy releases? | 14:27 |
| plars | asac: I've been looking at it, a few jobs have been restarted, same bug as yesterday for sure on filemanager tests | 14:30 |
| sergiusens | jdstrand: you should talk to diwic regarding his plans for PA->alsa and no binder | 14:30 |
| cjwatson | mhall119: bzoltan did some backports in the SDK PPA | 14:31 |
| === Namidairo is now known as Namidairo`bnc | ||
| jdstrand | sergiusens: ack | 14:39 |
| oSoMoN | mhall119: hey, I commented on https://code.launchpad.net/~mhall119/ubuntu-calendar-app/add-attribution/+merge/176809 some time ago | 14:41 |
| mhall119 | thanks cjwatson | 14:44 |
| mhall119 | oSoMoN: you are correct, I only removed people with @canonical.com addresses | 14:48 |
| oSoMoN | mhall119: I guess om26er should be using his @canonical.com address for contributions made during working hours | 14:49 |
| mhall119 | oSoMoN: it doesn't hurt to have him listed there separately though, so you can approve the branch if you don't want to wait for me to update it | 14:49 |
| mhall119 | oSoMoN: eh, that's not really that important, all of my contributions are with my @ubuntu.com address | 14:49 |
| mhall119 | that's just my default bzr info | 14:49 |
| oSoMoN | mhall119: yeah, I don’t really mind, was just pointing it out, I’ll approve the branch | 14:50 |
| mhall119 | and besides, I'm not sure he was making that contribution as part of his job | 14:50 |
| om26er | mhall119, oSoMoN right, core-apps generally don't fall as part of my job, but I have a certain interest in autopilot tests so I contribute a bit.. but in some cases I am asked by my superiors to do the work, like the recent autopilot tests fixes for touch. | 14:52 |
| mhall119 | the line between work and community contribtuions is always blurry for Canonical folks | 14:53 |
| oSoMoN | om26er, mhall119: yup | 14:54 |
| _lunarcold | any help with getting this build to work? I can't seem to find where this nfc step is failing | 14:55 |
| davmor2 | ogra_: I flashed my phone it seems to have no backdrop just black I'll reboot it and see what is what | 15:26 |
| harris | when will it be realesed to download | 15:26 |
| harris | hi | 15:27 |
| ogra_ | harris, it is downloadable since februar | 15:28 |
| ogra_ | davmor2, odd | 15:28 |
| harris | no not the developer | 15:28 |
| harris | the full thing | 15:28 |
| harris | like out for the public | 15:28 |
| ogra_ | on release day | 15:28 |
| ogra_ | with the rest of 13.10 | 15:28 |
| davmor2 | ogra_: reboot and I have the correct background no idea what caused that | 15:29 |
| davmor2 | ogra_: do you see these http://ubuntuone.com/1BONFEbcI2skarKPOfDKgR ? | 15:30 |
| davmor2 | ogra_: I'm assuming it's just all the scopes | 15:31 |
| ogra_ | davmor2, i see some of them (my image is two days old though) | 15:31 |
| davmor2 | ogra_: for me today there are about 2 pages of those :) | 15:32 |
| ogra_ | lovely | 15:32 |
| davmor2 | ogra_: I wouldn't mind but obviously most of the scopes only work on search from the home screen and does search work on the home screen, does it boat. ;) and if you click on them you get nothing where as in the desktop you at least can enable disable them on clicking them :0 | 15:35 |
| davmor2 | :) even | 15:35 |
| rsalveti | ogra_: having --legacy might still be useful for people tracking bugs and doing demos (I know, old stuff, but we know what is broken in there) | 15:49 |
| rsalveti | sergiusens: dude, you're starting your day at 7am, I wish I could do such thing hahah | 15:49 |
| ogra_ | rsalveti, sure, but we really need to drop it on release day ... once we have a stable release it should only be release or dev image | 15:49 |
| balloons | m-b-o, ping | 15:49 |
| rsalveti | ogra_: right, indeed | 15:50 |
| sergiusens | rsalveti: ogra_ in the new branch it's very isolated from the rest of the stuff so I wouldn't care anymore | 15:50 |
| awafaa | is the latest release really a week old? | 15:53 |
| m-b-o | balloons: pong | 15:53 |
| === jasoncwarner__ is now known as jasoncwarner | ||
| awafaa | according to phablet-flash 20130731.3 | 15:53 |
| balloons | m-b-o, I wanted to chat with you and elopio about your tab troubles in the weather app. elopio just finished some changes to the tab emulator, but I don't think they would affect your issue. maybe :-) | 15:54 |
| balloons | awafaa, that's the last release that passed all tests, hence it's pushed to you by default | 15:54 |
| balloons | daily builds are still occuring | 15:54 |
| awafaa | balloons: thanks, would i be getting that release because i used -b for a new device? | 15:55 |
| m-b-o | balloons: would it be possible in 3h? I have saddled my bicycle just a moment ago... :) | 15:56 |
| balloons | m-b-o, of course :-) you enjoy you ride! | 15:56 |
| m-b-o | I will, thanks :) | 15:56 |
| balloons | awafaa, tes | 15:56 |
| awafaa | ok so without the -b i should get the daily, thanks | 15:57 |
| balloons | awafaa, it doesn't matter which setting you used.. phablet flash will pull the stable by default | 15:57 |
| balloons | sorry to confuse, I misread :-0 | 15:57 |
| awafaa | oh, so how do i get the stuff that slices my eyes when i look at it? | 15:57 |
| balloons | --pending should pull the latest | 15:58 |
| balloons | I believe :-) | 15:58 |
| awafaa | great, tvm | 15:58 |
| === greyback_ is now known as greyback|food | ||
| jram | hi | 16:32 |
| jram | im trying to port ubuntu touch and it wont build a zip that i can flash | 16:32 |
| jram | please help | 16:33 |
| jram | hello | 16:35 |
| === greyback|food is now known as greyback | ||
| === sporkeee is now known as wilee-nilee | ||
| balloons | iBelieve, back from vacation? :-) | 17:26 |
| balloons | iBelieve, I'm sorry to have dropped such huges changes to file manager tests while you were away! | 17:26 |
| iBelieve | balloons, yep, and working on more cool stuff for the file manager :) | 17:28 |
| iBelieve | balloons, no problem about the huge changes, I was looking over the new autopilot tests and they look really cool how they work! | 17:28 |
| interloperX | who is working on porting to nexus 7 2013? (flo) | 17:34 |
| === schwuk is now known as schwuk_away | ||
| plars | click packages not installable right now? Or am I doing something wrong? I'm getting: | 17:48 |
| plars | ValueError: Framework "ubuntu-sdk-13.10" not present on system | 17:48 |
| xnox | if only emulator would not hang! | 18:10 |
| === Ursinha is now known as Ursinha-afk | ||
| rc-e66 | hi everyone | 18:44 |
| rc-e66 | I have a question about submitting an image | 18:45 |
| rc-e66 | Or publishing an image to be more appropriate | 18:47 |
| AskUbuntu | deleted ubuntu touch from galaxy nexus, now there is no os, how can i sideload os from windows | http://askubuntu.com/q/329426 | 18:49 |
| sergiusens | rc-e66: did you create a port? | 18:50 |
| m-b-o | balloons: ping | 19:02 |
| balloons | m-b-o, pong | 19:02 |
| m-b-o | what's up? :) | 19:02 |
| balloons | so let's talk tabs, and as luck would have it, I'm messing with your ap tests to fix the issues with the device | 19:03 |
| m-b-o | you want to branch this one https://code.launchpad.net/~martin-borho/ubuntu-weather-app/sdk_emulator | 19:03 |
| m-b-o | pushed some fixes | 19:03 |
| balloons | m-b-o, so what about the tab emulator failed you? and does this fix it? https://code.launchpad.net/~elopio/ubuntu-ui-toolkit/fix1206251-tabs/+merge/178410 | 19:04 |
| d53220 | so what's the best tablet to get for this? | 19:04 |
| balloons | elopio made changes to not be so hacky with how we get and switch tabs ;-) | 19:04 |
| d53220 | I understand that the google nexus is underpowered because of the RAM.. | 19:04 |
| === Ursinha-afk is now known as Ursinha | ||
| m-b-o | balloons: haven't seen this, will check | 19:05 |
| balloons | m-b-o, ohh you made a branch that converted.. nice | 19:05 |
| m-b-o | balloons: for sure! But what do you mean with converted? ;) | 19:06 |
| balloons | sorry, I mean you made a branch with the new toolkit.. I've been playing with the branch as published | 19:07 |
| rc-e66 | sergiusens: I followed the instructions on the wiki. I had my old droid incredible that I was interested in using as a PMP. | 19:07 |
| sergiusens | rc-e66: PMP? | 19:07 |
| * sergiusens is lost in acronyms today | 19:07 | |
| m-b-o | balloons: yes, I have overwritten the Tab emulator. See in emulator.py | 19:08 |
| rc-e66 | sergiusens: Portable Media Player | 19:08 |
| rc-e66 | sergiusens: It's not the fastest and I can't technically verify the modem pieces due to lack of service, but I only care for wifi and sd card functionality. | 19:08 |
| m-b-o | balloons: the one test that crashes is date related and should be okay tomorrow O:-) | 19:08 |
| sergiusens | rc-e66: so you want to install an image, not create one? | 19:09 |
| balloons | m-b-o, ok, I think there might be a need for more asserts, as the tests are failing on the device.. but I'll take care of that.. | 19:09 |
| rc-e66 | sergiusens: I'd like to have it added to the regular builds, like those for the Nexus devices. | 19:09 |
| sergiusens | rc-e66: oh, we don't do the regular building (yet) | 19:10 |
| m-b-o | balloons: which device do you use? | 19:10 |
| balloons | m-b-o, manta | 19:10 |
| sergiusens | rc-e66: but we do have a way to make it easily available | 19:10 |
| rc-e66 | sergiusens: I thought I saw a repo for the images on quantal? | 19:10 |
| balloons | but have a look: http://reports.qa.ubuntu.com/smokeng/saucy/image/3382/ubuntu-weather-app-autopilot/ | 19:10 |
| balloons | m-b-o, ^^ | 19:10 |
| sergiusens | rc-e66: https://wiki.ubuntu.com/Touch/MakingPortsAvailable | 19:10 |
| m-b-o | balloons: tried on maguro and test_refresh_tabs is the only one left that is failing (date related) | 19:11 |
| rc-e66 | sergiusens: Step one complete under prereqs. :) Guess I should get started on the rest... | 19:11 |
| balloons | m-b-o, :-) good so the changes in your branch have fixed things.. I'll stop messing with this old branch then | 19:12 |
| m-b-o | I've runned autopilot direct on the device, not with phablet-test-run. should try that one | 19:13 |
| balloons | m-b-o, in debian/control under the -autopilot package, you need to include ubuntu-ui-toolkit-autopilot | 19:13 |
| rc-e66 | sergiusens: Thanks for the link. Any idea on what I should do with questionable files/links? HTC Dev and CM 7 files were used, but I believe have an open(ish) license. | 19:14 |
| balloons | m-b-o, yea I do my runs now with phablet-test-run as that's how they are done in the lab | 19:14 |
| m-b-o | right, forgot that to ask :) | 19:14 |
| sergiusens | rc-e66: the license is more related to the binary blobs... you can probably get away with an _unkown_ for now | 19:14 |
| m-b-o | ballooons: what's the correct command to test the weather app? | 19:14 |
| balloons | m-b-o, the thing with phablet-test-run is it will run the installed version on the device | 19:15 |
| balloons | so I've been using adb push to overwrite the files so I can keep using it, hah ;p | 19:16 |
| balloons | m-b-o, however phablet-test-run -n ubuntu_weather_app | 19:16 |
| m-b-o | balloons :) | 19:17 |
| m-b-o | I've added the dependency | 19:17 |
| rc-e66 | sergiusens: Thanks again! | 19:19 |
| rc-e66 | Good day folks! | 19:19 |
| sergiusens | rc-e66: if you log in early | 19:19 |
| sergiusens | meh, too late | 19:19 |
| balloons | m-b-o, do you think you'll land that branch today? | 19:27 |
| m-b-o | balloons: yes. do you want to review? :) | 19:27 |
| balloons | m-b-o, I'd be happy to in order to get it in :-) | 19:27 |
| m-b-o | balloons, fine! :) currently the data api is down or to slow... so when it's up to normal speed I will do a MP | 19:28 |
| m-b-o | balloons: is there a way to mock http requests? | 19:29 |
| balloons | yea, those requests concern me a bit too.. umm, I don't know of a simple way to do it persay. my first thought would be to emulate the server response, and to play some funny games with how you route it, looping back to your box | 19:30 |
| balloons | someone else would have a better idea perhaps | 19:31 |
| === interloperX is now known as interloper | ||
| m-b-o | balloons: yes, that bites me sometimes too. Nonetheless, MP is waiting for you! :) | 19:41 |
| balloons | m-b-o, ok :-) | 19:42 |
| === Ursinha is now known as Ursinha-afk | ||
| plars | slangasek: so how would I go about starting apps with upstart? | 20:10 |
| === Ursinha-afk is now known as Ursinha | ||
| sergiusens | slangasek: stgraber lool shouldn't we rename developer mode to image developer mode? I should necessarily need to go into this current form of developer mode if developing apps, would I? | 20:20 |
| mariusko | Hi, does anyone know if the flipped image stuff broke support for MultiROM? | 20:21 |
| rickspencer3 | sergiusens, ah, the classic Ubuntu ambiguity of "developer" | 20:22 |
| sergiusens | rickspencer3: I think it gains importance here because this developer mode would prevent the updates from rolling in and if I were just an app developer I wouldn't want that to happen | 20:23 |
| rickspencer3 | indeed | 20:24 |
| sergiusens | don't want to get people confused into making them think they need this either, the image should be enough for building and testing | 20:24 |
| sergiusens | ...building and testing apps that is | 20:25 |
| Tassadar_ | mariusko: yeah, initialy, but it should be okay now | 20:26 |
| mariusko | Tassadar_: ah, is there any version released that support them? | 20:27 |
| Tassadar_ | mariusko: sorry, AMD drivers crashed my kernel, did you get the respons about MultiROM and flipped images? | 20:31 |
| slangasek | plars: hmm, I'm not really in the loop wrt best practices for using upstart to manage apps; that's tedg's baby | 20:37 |
| slangasek | plars: I can only tell you that it involves the 'start' command | 20:38 |
| slangasek | sergiusens: yes... we discussed last week the need for such a separation | 20:38 |
| tedg | plars, You need the upstart-app-launch package. Then you can do "start application APP_ID=inkscape" for instance. | 20:38 |
| slangasek | sergiusens: I think ChickenCutlass has the big picture on this | 20:38 |
| tedg | plars, There'll be more utilities to make it a bit easier once I get these branches cleaned up (actually working on that now) | 20:39 |
| plars | tedg: ah, I just had heard about that, where can I find it? does it do more than just running /sbin/start application APP_ID="foo"? | 20:39 |
| tedg | plars, Uhm, not really. We have the library just to make the interface consistent for Unity. Incase we need to change the name or something. | 20:39 |
| tedg | plars, The lib also does things like set up watches, etc. | 20:40 |
| tedg | plars, Stuff that Unity needs. | 20:40 |
| tedg | plars, What are you thinking of using it for? | 20:40 |
| plars | slangasek, tedg: so one thing we're realizing, is that all the application test owners are going to need to change their tests to launch apps differently in autopilot tests, right now they seem to be calling autopilot's launch_test_application() with the hard path under /usr | 20:41 |
| * slangasek nods | 20:41 | |
| slangasek | so I chatted with thomi about this briefly last week too | 20:41 |
| plars | tedg: I'm hoping something like this can be used by them instead, but will it handle the transition from .deb installed apps to click without anything extra? | 20:41 |
| tedg | Yup. Would probably be best to install the click packages. | 20:41 |
| tedg | Yes, it'll work with both. | 20:42 |
| tedg | The big difference is that legacy apps aren't confined by default. | 20:42 |
| plars | tedg: so is there a better way to call this from python? | 20:42 |
| slangasek | you just want to invoke the 'start' command, whether from python or otherwise | 20:43 |
| tedg | We'll have a little utility in upstart-app-launch-tools called "upstart-app-launch" which would probably be better. | 20:43 |
| slangasek | there's no (supported) python interface for upstart's RPC | 20:43 |
| plars | tedg: I think some also need to be launched with arguments like --testability, or --use-test-data, etc | 20:43 |
| slangasek | ah, or ted's wrapper is fine too :P | 20:43 |
| plars | tedg: is that possible here? | 20:43 |
| tedg | plars, Nope | 20:43 |
| slangasek | tedg: no way to pass commandline args to the invoked app? | 20:43 |
| tedg | They'll need to install a different application. | 20:43 |
| tedg | slangasek, No, only URIs | 20:43 |
| tedg | Which are kinda params. | 20:44 |
| slangasek | different application> that sounds distressingly incompatible with integration testing | 20:44 |
| tedg | But we're not really looking at command line apps. Things with desktop files. | 20:44 |
| slangasek | we're not talking about commandline apps | 20:44 |
| mariusko | Tassadar_: sure, I don't get the flipped image to work with the Jul 27 release. Should it? | 20:44 |
| slangasek | we're talking about how to hook up the autopilot test harness for the running app | 20:44 |
| slangasek | I did say that this ought to be an env var rather than a commandline arg | 20:44 |
| tedg | Sure, but I'm saying what upstart-app-launch is designed for. | 20:44 |
| mariusko | Tassadar_: and the Ubuntu touch option when adding image does not work at all, so I do it manually as before | 20:45 |
| tedg | Yes, you could inject the env var into upstart. | 20:45 |
| Tassadar_ | mariusko: yeah. Where are you installing it, USB drive or internal memory? | 20:45 |
| Tassadar_ | how "does not work"? | 20:45 |
| slangasek | plars: if you could set an env var, could autopilot be fixed to work with this? | 20:45 |
| slangasek | instead of relying on commandline args | 20:45 |
| plars | slangasek: no idea, I suppose that would depend on all the app developers supporting taking those args from the environment rather than on the command line | 20:46 |
| slangasek | plars: I thought this was all being handled at a much deeper level than the app | 20:46 |
| slangasek | like, autopilot hooking into qt | 20:46 |
| thomi | right, it is. We could patch Qt | 20:47 |
| mariusko | Tassadar_: internal memory | 20:47 |
| thomi | autopilot can easily be changed to export an env var instead of add command line args (in fact, that's what we already do for Gtk apps) | 20:47 |
| slangasek | I think given where it's hooked in, it makes more sense for it to be an env var instead of a commandline arg anyway | 20:47 |
| slangasek | and then the env var can be exported as part of the upstart session itself | 20:48 |
| thomi | but we'd need to patch Qt to read that env var instead of reading the command line args | 20:48 |
| mariusko | Tassadar_: it does not boot at all. I just see the Google logo forever | 20:48 |
| slangasek | thomi: right | 20:48 |
| Tassadar_ | mariusko: hmm, let's go to channel #multirom, so that we don't spam this one | 20:48 |
| mariusko | I'm not sure if it could be caused by me having older versions of Multirom/TWRP installed before | 20:48 |
| thomi | so... who do we bug to patch Qt? | 20:48 |
| slangasek | thomi: would that be acceptable? I guess it needs to be discussed with Qt upstream (so that the Kubuntu team don't balk) | 20:48 |
| tedg | plars, thomi, you guys will still need to install the apps as click packages so all the apparmor hooks run. | 20:48 |
| tedg | We'll fail if we can't setup confinement on click stuff. | 20:49 |
| slangasek | thomi: ah, "who", good question. I think the SDK team owns the Qt5 packages | 20:49 |
| thomi | slangasek: I doubt upstream would take that patch, but I'm only guessing | 20:49 |
| plars | tedg: we aren't installing the apps, they are all in the image | 20:49 |
| plars | tedg: that's why I'm looking for something that will work this week (while they are coming from .debs) and next week forward | 20:49 |
| slangasek | thomi: I don't see any good reason for them not to. An envvar is a more appropriate interface for this kind of thing. | 20:49 |
| plars | tedg: because I'd like to get people fixing these tests right away | 20:49 |
| thomi | slangasek: I am pleased by your optimism :) | 20:50 |
| slangasek | thomi: sorry if that came across as optimism, it was intended to be technically arrogant ;) | 20:50 |
| slangasek | "they'll take this patch if they have an ounce of sanity" | 20:50 |
| slangasek | ;) | 20:50 |
| plars | conditional optimism, good | 20:51 |
| tedg | plars, It will work for apps installed as desktop. They should eventually be put on the image as "installed" so that should be fine. | 20:51 |
| tedg | plars, I was thinking that you were testing additional versions, and those should be installed as well. | 20:52 |
| plars | tedg: I'm just on the image testing bits, fginther may know more about the pre-image stuff | 20:52 |
| thomi | slangasek: heh well, yes | 21:00 |
| fginther | plars, tedg, the pre-image testing would install the app as a click package, and then pull in everything else that is needed under developer mode | 21:01 |
| thomi | so, just to confirm, we're saying that autopilot should esentially run "QT_ENV_VAR=1 start app-name" (where QT_ENV_VAR is whatever Qt is patched to look for) | 21:05 |
| thomi | slangasek: tedg ^^ | 21:06 |
| tedg | thomi, No | 21:06 |
| tedg | thomi, You guys will need to set the variable for upstart. | 21:06 |
| tedg | thomi, I'd recommend having an upstart job do it for you so the whole desktop gets it. | 21:06 |
| thomi | tedg: pretend I don't know anything about upstart... how do I do that? | 21:07 |
| slangasek | thomi: no, because 'start' is an RPC call to upstart; its own environment is discarded. As tedg says, the variable needs to be set in upstart, which is basically 'initctl set-env foo=1' | 21:07 |
| tedg | thomi, I'd look at "unity-gtk-module" | 21:07 |
| tedg | thomi, That's a job that we have that sets up the environment for the unity-gtk-module | 21:07 |
| tedg | thomi, The start condition there makes sure that every application that is started or dbus activated gets the environment. | 21:08 |
| thomi | tedg: ok, I have the source branched, where am I looking? | 21:08 |
| tedg | thomi, /usr/share/upstart/sessions/unity-gtk-module.conf | 21:08 |
| thomi | oh :) | 21:08 |
| tedg | You're already seeing the future. Today! | 21:09 |
| tedg | ;-) | 21:09 |
| thomi | tedg: so.. how and when is that script called? | 21:09 |
| tedg | thomi, It is called before the session dbus is started. | 21:09 |
| tedg | thomi, initctl2dot will generate a graph for you | 21:10 |
| thomi | tedg: that's because the file has "start on starting dbus" at the top? | 21:10 |
| tedg | thomi, correct | 21:10 |
| tedg | thomi, Handy link for when I'm asleep ;-) http://upstart.ubuntu.com/cookbook/ | 21:11 |
| thomi | tedg: OK, so we don't want to export this variable all the time, only when autopilot is running | 21:11 |
| tedg | thomi, Sure, I'd only install the job on your test system. You can do it in the user's home directory if you wish. | 21:11 |
| thomi | hmmm | 21:12 |
| thomi | so perhaps autopilot would, at test runtime, install the file, then launch the app.. when the test ends, remove the file again - would that work? | 21:12 |
| tedg | You should in theory be able to just do "initctl set-env" but you can't guarantee everything sees it. | 21:12 |
| thomi | or would we just install it whenever autopilot package is installed? | 21:12 |
| tedg | If you only want it for a single app... but, don't you want it for the system? | 21:12 |
| tedg | You wouldn't need a job if it was just for one call. | 21:13 |
| thomi | I guess it depends on what we're testing | 21:13 |
| tedg | For instance, if you want Unity to get it, you can't do it after unity starts. | 21:13 |
| thomi | right' | 21:13 |
| thomi | ugh | 21:13 |
| thomi | tedg: OK, so lets ignore the case where we want it for the whole system for now - how do I set the env for a single app only? | 21:14 |
| tedg | Just: initctl set-env --global MYVAR=foo | 21:14 |
| tedg | That'll make everything from then on get it. | 21:14 |
| stgraber | sergiusens: I don't particularly care about the name especially as it's not exposed in the UI at this point. I went with developer mode in the same sense that it's used on Android, which is really meant to be for OS developers and not app developers. | 21:14 |
| tedg | You can unset it as well if you'd like | 21:14 |
| thomi | tedg: awesome. that sounds like the path of least resistance for now | 21:15 |
| stgraber | sergiusens: I have no idea what our plans are for the settings UI in that regard, but I guess it'd make sense to have a clear "Make the filesystem writable" option in whatever advanced options UI we'd get. | 21:15 |
| tedg | thomi, Yeah, that'll work for apps. I thought you needed something bigger. | 21:15 |
| AskUbuntu | Help with installing ubuntu on a tablet | http://askubuntu.com/q/329492 | 21:15 |
| thomi | tedg: we probably will, but I'm hoping that by then someone who knows upstart can take over ;) | 21:15 |
| tedg | thomi, We can do it in the hallway track at the next UDS... oh, wait... | 21:16 |
| thomi | heh | 21:16 |
| sergiusens | slangasek: thanks | 21:24 |
| sergiusens | tedg: we will be migrating to click packages soon for all the apps, shouldn't be a problem | 21:25 |
| sergiusens | for installing packages on the fly, we already sinked up with fginther and rolling out an initial implementation to get this going | 21:25 |
| zakker | i am a new bi | 21:29 |
| tedg | sergiusens, Cool | 21:40 |
| slangasek | tedg, thomi: "can't guarantee everything sees it" - that only applies for jobs started before you run the command, AFAIK. So for an app you know you're launching yourself, initctl set-env should suffice | 22:00 |
| tedg | slangasek, Yes, I was thinking for testing Unity. | 22:00 |
| thomi | well, we re-launch unity anyway | 22:01 |
| tedg | thomi, "restart unity" ? | 22:01 |
| tedg | You should really use the upstart job to do it. | 22:01 |
| thomi | tedg: right now we kill unity and launch from the command line. Obviously that will need to change | 22:01 |
| tedg | Actually, in the Mir world I think killing it might kill your session as well. | 22:02 |
| tedg | So, yes, yes you will :-) | 22:02 |
| === wilee-nilee is now known as curlee | ||
| steven_ | I have a question if the Samsung Exhibit II will work with this? | 23:06 |
| slangasek | steven_: https://wiki.ubuntu.com/Touch/Devices | 23:22 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
