[00:00] asac: yeah, camera is very crashy... just taking a picture crashed it here [00:02] asac: phone/sms works though [00:08] plars: so camera is a real regression over what is in RO daily? [00:09] asac: appears to be, yes [00:09] asac: both maguro and mako failed twice in a row on it [00:09] asac: previous images didn't [00:11] asac, plars: Problem is that the directory has wrong permissions now [00:11] it's read-write, but not usermetrics:usermetrics [00:11] stgraber: ^ [00:14] sergiusens, lool: Debian bug #722045 for dh-python :any support; it's a rather big change which I haven't had a chance to test very broadly yet, so I'm hoping to get some feedback on it from Debian before pushing into saucy [00:14] Debian bug 722045 in dh-python "Please support python:any dependencies for multiarch compatibility" [Normal,Open] http://bugs.debian.org/722045 [00:15] lool: which one? [00:15] the camera? [00:15] lool: did that regress from daily? [00:15] * asac had crashes on that as well in retrospect [00:15] lool: note: in automation we use RW [00:17] asac: it's the same thing I discussed earlier [00:17] slangasek: thanks! [00:17] asac, plars, sergiusens, stgraber: I've confirmed that "just" fixing the permissions of /var/lib/usermetrics to usermetrics:usermetrics works [00:18] but I don't know how to do that properly in the initramfs [00:18] lool: but when did that sneak in? [00:18] asac: it's in the r/o images [00:18] so the camera app is not a regression over latest daily? [00:19] asac: It couldn't work with read-only images IMO [00:19] I couldn't test on grouper since it doens't work there with frontcam for some reason [00:19] but on mako, the directory permissions prevented it from working [00:19] can i run ubuntu on a firefoxOS phone? [00:20] lool: since THIS build [00:20] we see crashes in automation for camera [00:20] i think it never worked on RO [00:20] how do they look like? [00:20] so i guess it doesnt matter [00:20] red [00:20] lool: http://reports.qa.ubuntu.com/smokeng/saucy/image/4041/camera-app-autopilot/ [00:21] mediascanner crashed [00:21] and of coruse that was let in [00:21] so sigh [00:21] thats the explain [00:21] in what phone are you running ubuntu on? [00:23] Gsport: http://wiki.ubuntu.com/Touch/Devices [00:23] plars: lool: so i guess we can pack our things :) [00:24] mediascanner would need a backout [00:24] would take another half day [00:24] does know why it crashed? [00:24] lool: sure there was an update [00:24] small/safe patch [00:25] we let that in for thsi build [00:25] 100% sure its that [00:25] https://launchpad.net/ubuntu/+source/mediascanner [00:25] is that libmediascanner-1.0-1:armhf or grilo-plugins-0.2-mediascanner:armhf? or both? [00:25] plars: try both [00:25] plars: but it wont fix it locally [00:26] plars: in RO mode [00:26] only in automation [00:26] plars: RO is said to never have worked [00:26] not sure if I want to go back [00:26] we could then argue we ignore because it doesnt matter [00:26] but you said that camear worked for you [00:26] in previous build [00:26] :) [00:26] it didnt work for me though [00:27] * asac installs the previoyus image again [00:27] folks were claiming that it worked for them [00:27] but i never saw it working [00:28] lool: plars: so i am tempted to publish this if the camera app never worked [00:28] and sergiusens confirms that he uploads the fix above [00:29] (e.g. fix RO) [00:29] and we backout mediaservice somehow [00:29] asac: I can try the previous image locally [00:29] try mako [00:29] i will try maguro [00:29] I saw the camera working, but half the time I had my rootfs in read-write mode, and I've also seen my camera *not* working, wiw [00:29] asac: I'm flashing it now [00:29] f [00:29] I won't put it in rw mode [00:30] asac: getting a bt [00:31] how can we backoutt he mediascanner change? [00:31] https://code.launchpad.net/~mediascanner-team/mediascanner/trunk [00:31] revert commit? [00:31] :) [00:31] inserve MP [00:31] inverse [00:35] asac: https://bugs.launchpad.net/mediascanner/+bug/1221984 [00:35] Launchpad bug 1221984 in mediascanner "Crashes under automation" [Undecided,New] [00:36] asac: Yes, bzr merge -r latestrev:previousrev [00:36] lool: i somehow feel we just decide to let camera regreess in the RW images [00:36] i am sure it worked before and now it will crash :) [00:36] not good [00:36] asac: video never worked for me if it's any confort [00:37] err [00:37] i am talking about photos :) [00:37] I know :) [00:37] ok, really heading off now [00:37] yeah i think we wont push this [00:37] bad luck [00:38] very depressing, but sometimes you just don't get lucky [00:39] plars: sergiusens: so no go i guess. too late, too many problems. better not bust RW for those that were fortunate and didnt go to RO yet [00:39] and more testing is just no time :) [00:41] will the ubuntu touch core apps become desktop core apps as well? [00:41] asac: I am not quite to a booted ro image, but at the very least it got worse in rw mode [00:47] plars: yeah the RO daily doesnt work camera hera [00:47] so given that we risk regression what works its not a good deal [00:47] RW will be a safe habour at least :) [00:47] so good night:) [00:54] camera seems to work on yesterdays ro daily [00:55] for me on mako [00:59] nexus 4 bootloader only | http://askubuntu.com/q/342391 === jhodapp is now known as jhodapp|afk === Namidairo`bnc is now known as Namidairo [02:51] phablet-flash: error: too few arguments | http://askubuntu.com/q/342412 [07:51] Ubuntu Touch not waking after standby and not booting after reboot | http://askubuntu.com/q/342483 [08:01] yo [09:42] Hi there, can maybe someone help me with a flash command, i want to test the ubuntu touch for the first time on my nexus 7 WIFI only (2012), and i followed the wiki, everything goes well but one command that is specified on the wiki fails, maybe i running it wrong ? [09:48] jppiiroinen: see mail [09:48] jppiiroinen: i asked for a backout of mediascanner [09:49] the last commit causes crashes and blocked us from releasing a critical fix last night [09:49] :) [09:49] thanks [09:50] asac: hmm not 100% it's the last commit [09:50] lool: it was only one commit that landed [09:50] so this morning I tried to use the app scope on n4, and it was completely broken [09:50] like search field at wrong place, no results [09:50] lool: http://bazaar.launchpad.net/~mediascanner-team/mediascanner/trunk/revision/380 [09:50] then app scope didn't show any apps anymore [09:50] thats the commit that landed [09:51] so I pressed power button to shutdown [09:51] but the device doesn't come back up... [09:51] I might have mounted it rw, in which case it's my bad I guess [09:51] would like to know how to check this though [09:51] like, seeing boot log [09:51] hmm. not sure :) [09:51] i know the click/app scope was always a bit problematic [09:52] throughout the whole day [09:52] asac: yeah, I had seen the commit from the launchpad diff of the last upload, and the backtrace I extracted from the .crash didn't look the same [09:52] https://bugs.launchpad.net/mediascanner/+bug/1221984 [09:52] Launchpad bug 1221984 in mediascanner "Crashes under automation" [Undecided,New] [09:52] lool: but tahts the only change and its now reproducible [09:52] it must be that [09:52] asac: yeah, I had the same impression in the last weeks (click scope) [09:52] so even if unsure i would throw it out [09:55] unity-webapps-qml:armhf from 0.1+13.10.20130904.2-0ubuntu1 to 0.1+13.10.20130907-0ubuntu1 [09:55] that landed in 07 imge [09:55] i am not sure [09:55] how that lands [09:55] but i thought it wouldnt [09:55] as we turned off all publishing [09:55] http://people.canonical.com/~ogra/touch-image-stats/20130907.changes [09:55] lool: so [09:55] the whole boost stack got landed yesterday [09:55] hell :) [09:55] http://people.canonical.com/~ogra/touch-image-stats/20130906.1.changes [09:56] so most likely it snot mediascanner alone [09:56] but this stuff [09:56] i thought we were in beta freeze [09:59] asac: the boost changes don't look like they would cause bugs, but a bunch of uploads for small errors https://launchpad.net/ubuntu/+source/boost1.53/+changelog [09:59] still worrying that folks did mass uiploads like this [09:59] xnox: from now on please dont upload without talking here :) [09:59] thx [10:00] Hello! I'm the developer of the RSS Feed Reader "RamSamSam Reader" (Ubuntu App showdown). I wanted to ask for help: I want to know if the application adheres to the user interface guidelines and how I could improve the user interface. [10:00] lool: i guess the experiemnt "lets see if we can get stuff under control" failed :) [10:00] asac: the latest upload seems to be really minor; bdep fix, and prior to that control fix; no change in functionality of the provided packages [10:00] so lets open floodgateds again on monday and cross our fingers until we have our engine improved so we can be in control about _all_ landings [10:00] lool: i dont know [10:00] everything seems minor [10:00] asac: we have to root cause the regression before assigning blame IMO [10:01] sure [10:01] thats why i now looked at the whole change set [10:01] lool: its not blaming... just told xnox to not upload anymore without coordinating with image producers [10:01] i blamed the otherse - maybe falsly - because it was mediascanner we promoted [10:01] as a "safe" landing [10:01] and it crashes [10:02] just tried to figure if they were not the cause [10:02] and found the big boost update [10:02] :) [10:02] it's not big though [10:02] lots of packages [10:02] it's a lot of packages in the list [10:02] but it's a minor change [10:02] which we should be confortable doing [10:03] can't have people come here for obvious minor fixes, wont scale [10:03] maybe if you had a pager it could scale [10:03] ;-) [10:04] in other news, I cant find the branch of ubuntu-download-manager anymore [10:04] it says Vcs-Bzr: https://code.launchpad.net/~ubuntu-system-image/ubuntu-system-image/downloader [10:06] yeah, i knwo its ultimately all about not having any testing upstream :) [10:07] lool: https://launchpad.net/ubuntu-download-manager probably here now? [10:09] lool: oh 28:20130906.1:20130906.1 [10:09] thats the last proposed he had on ro? [10:09] its weird because the last in touch is 20130906.3 [10:09] anyway :) [10:10] lool: because here the camear is working: http://reports.qa.ubuntu.com/smokeng/saucy/image/4038/ [10:11] lool: guess we really didint test the latest yesterday night :)? [10:13] ok so installed paending RW and camera clearly crashes [10:13] now backing out media stuff [10:15] lool: odd i installed just cdimage-touch and its not RW anymore [10:15] W: Not using locking for read only lock file /var/lib/dpkg/lock [10:15] E: Unable to write to /var/cache/apt/ [10:15] E: The package lists or status file could not be parsed or opened. [10:16] * asac installs one more time [10:30] lool: so for sure using phablet-flash cdimage-touch doesnt give me a rw install anymore :) [10:32] asac: do a --bootstrap [10:32] asac: or --wipe [10:32] ok [10:32] will try :) [10:33] err the latest RW pending one [10:33] has the old mediascanner packages still [10:33] oddish [10:35] sergiusens: both options will delete $HOME? [10:37] lool: check all the crashes we see since yesterdays image [10:37] http://reports.qa.ubuntu.com/smokeng/saucy/image/4038/ [10:37] like every test has something crashing [10:37] you see the increase here: http://reports.qa.ubuntu.com/smokeng/saucy/touch/ [10:37] it started with 20130906.3 [10:37] but I CANT FIND a .3 :) [10:37] at all [10:37] neither in ogras changes [10:37] http://people.canonical.com/~ogra/touch-image-stats/ [10:38] nor in the RO image version [10:38] which was always .1 [10:38] how confusing [10:38] how confusing :) [10:39] ok its the unity click store [10:39] its ALWAYS crashing :) [10:40] lool: ^^ [10:40] that came in through unity-scope-click from 0.1+13.10.20130903-0ubuntu1 to 0.1+13.10.20130906.1-0ubuntu1 [10:41] i dfont know how that came in [10:41] thoguht it would have gone through daily-release [10:44] asac: yes, both... sorry, was catching up with backlog [10:46] hey its SAT :) [10:47] it is, and I have to get ready for travel, full 16 hours on a plane or in between airports, somehow this flight plan is 6 hours shorter than usual [10:47] so guess unity-scope was this change http://bazaar.launchpad.net/~ubuntuone-hackers/unity-scope-click/trunk/revision/43 [10:48] sergiusens: hehe.. yeah. see you in boston [10:48] dont worry :) [10:48] asac: lol [10:48] i wanted to write slides [10:48] asac: there was also a download manager update [10:48] asac: where are you? [10:48] lool: at home [10:48] tomorrow flying [10:48] (i hope it was not today :)) [10:49] asac: I always get that feeling too [10:49] asac: I have to be at the airport at 2AM, so I still have time btw [10:49] just seeing if I'm flying all ARM or taking an x86 too [10:50] lol [10:50] sergiusens: take an x86 [10:50] :) [10:51] its odd... i thought nothing would go into the archive, but unity-click-scope made it [10:51] asac: you don't like arm? A chromebook is what I was talking about, not just tablets btw ;-) [10:51] sergiusens: i like armrests in the plane ... if i can conquer them :) [10:52] asac: lol [11:05] * asac goes off to barber [11:43] sergiusens: https://code.launchpad.net/~lool/phablet-tools/static-checkers/+merge/184432 [11:45] "you'll need the Ubuntu preinstalled image besides the one you're going to build with the instructions below." [11:45] https://wiki.ubuntu.com/Touch/PortingFlippedInProgress [11:45] for what? [11:45] what am I supposed to do with them? [11:45] they aren't mentioned in the porting guide [11:49] I have 'apparmor switch' line removed from click-applications.conf, and it doesn't seem to allow me to execute a click app.... Is that right? [12:02] I will just go for an upstart session that: sed "s/aa-exec.*-- //g" /home/phablet/.local/share/applications/*.desktop [12:14] OrokuSaki: look at the override file [12:17] lool: do you want to add phablet-flash to the files to check? [12:17] et.al. [12:20] slangasek: manually install python:any (if not python:armhf was picked up by default), changing the glib2.0-dev dep to python:any allowed me to fully install libnotify-dev:armhf [12:22] what should I do with the Ubuntu preinstalled images? [12:23] I have a system.img, boot.img, etc [12:47] hello [12:47] i am trying to install touch on a galaxy 10.1 [12:48] i have been through the tutorial and everything installed properly [12:48] only problem is that android boots and not ubuntu [12:49] i have used the pre-installed image under daily build touch (which is Saucy by now) [12:49] should i have used the phablet-touch-preview ( which is old ) === wil is now known as Guest7522 [13:00] Русские есть? [13:01] hello [13:05] I am from Russia. I would like to find out firmware for Acer A700/7001? [13:19] zeta: https://wiki.ubuntu.com/Touch/Devices [13:20] https://wiki.ubuntu.com/Touch/Devices/a700 [13:24] getting the following error: E: Package 'python-lzma' has no installation candidate ubuntu@ubuntu:~$ sudo add-apt-repository ppa:phablet-team/tools You are about to add the following PPA to your system: [13:45] where to get Python? The following packages have unmet dependencies: phablet-tools : Depends: python-lzma but it is not installable Recommends: ubuntu-dev-tools but it is not installable [13:46] sergiusens: so --wipe made my system now boot to busybox only [13:46] i assume that means i need to start from scratch? [14:05] lool: i am still super confused about the versioning:) [14:05] lool: actually not [14:05] i think... its just the dashboard that tricks me it seems [14:21] so for me the camera works well on maguro [14:21] rw latest [14:21] interesting that we are so crashy then on the dashboard [14:21] hehe [14:22] plars: any idea why the 07 build isnt showing up? [14:22] the touch_ro? [14:22] sergiusens: Pushed r196, with love [14:25] asac: system-image crontab got disabled [14:25] asac: I think it was in manual mode friday evening, and we forgot to reenable it [14:29] asac: I'm not confortable enough ensuring the right flags are set there, I'll mail stgraber [15:03] lool: oh ... but the other images continue? [15:03] thats interesting :) [15:20] Do any of you guys know how to access the value of an enum variable? [15:20] I am referring to http://paste.ubuntu.com/6075061/ [15:21] This is the alarms API provided by the SDK [15:23] http://qt-project.org/doc/qt-5.0/qtqml/qml-enumeration.html [15:24] asac: I think it was put into manual mode to generate the new daily-proposed channel [15:27] netcurli: reading now. thnx [15:31] netcurli: I understand there is a and . I was able to assign text to a variable. So that variable got that respective enum value. [15:31] netcurli: However I want to do the reverse. Which is from the enum value, I want to get the type [15:32] netcurli: so in the code I pasted, I set the day to be Alarm.Monday which has enum value 0x01. When reading it now, I get the result 1. How do I translate that back to Alarm.Monday? [15:34] nik90_: I don't know if you can do that without an explicit function for that [15:35] netcurli: ah so I will have to create a function with switch cases to get back to Alarm.Monday? that's a pity [15:38] hello, I need some help working on the N80xx series tablet :) [15:39] after the first boot, the following won't lead to a working envirorment.. [15:40] nik90_: does something like Alarm.DayOfWeek(1) work? [15:41] ogra_, hi, got a moment? [15:44] sergiusens, can you help me a little? [15:45] netcurli: will try now [15:48] what should I do with the Ubuntu preinstalled images? I have system.img, boot.img, etc. [15:53] netcurli: nope that doesnt work [15:53] netcurli: I tried Alarm.DayOfWeek(1), Alarm.DayOfWeek(daysOfWeek) and daysOfWeek(Alarm.DayOfWeek) [15:53] none of them work [15:53] ok [15:54] was worth a shot [15:54] yes [15:54] hmm I am seeing no benefit of using enum for the api [15:54] should ask the dev [15:54] other than being able to assign by name like Alarm.Tuesday and so on [15:55] why exactly do you need the value of the enum? [15:55] to save it? [15:56] no I need it to read when the alarm is set to ring [15:56] so by getting the value of the enum which may be 1,2,4,8,10,20,40 and 80, I will be able to correspond it to whether the alarms rings on Monday, Tuesday or wednesday etc. [15:57] and the reverse is needed for assigning the day [15:57] for assigning the day, I do Alarm.Monday etc [15:58] oh, there can be multiple days for the alarm?! [15:59] netcurli: yup [15:59] you can make comparisons against enums just fine [15:59] but if you have to do some math stuff [15:59] .. [15:59] right now I did switch(enumvalue), case 1: return "Monday" and so on [16:00] to get a string you would need something like that anyway [16:01] the best idea however is probably to talk to the guys who designed the api [16:01] yeah [16:01] to ask how they imagined it to be used [16:02] yeah I feel like I am coding the long way :D [16:03] if anyone could show me how to get apparmor working on clickapps by default I would.. really appreciate it.. my kernel doesn't have apparmor, and the .desktop files seem to get regenerated on every boot.. [16:04] I know it is /usr/share/upstart/sessions/applications-click.conf and override related [16:07] hi OrokuSaki, does your device have wake up problems? [16:07] no [16:07] Ubuntu Touch: Screen color tone | http://askubuntu.com/q/342615 [16:08] OrokuSaki, any idea why all of mine has? :D [16:08] got me on that one [16:09] guess it's a bug [16:18] Kaleo, you around? [16:29] what should I do with the Ubuntu preinstalled images? I have system.img, boot.img, etc. [17:00] asac: looks like there was no new image? [17:01] plars: the cronjob to import to system-image was disabled until 30min ago, an import must be running pretty much now [17:01] stgraber: ok [17:21] asac: version 29 is up [17:21] plars: ^ [17:22] lool: yep, testing has already kicked off on it [17:25] stgraber: it's cool to see the small deltas [17:25] like 13M today for me [17:33] lool: http://paste.ubuntu.com/6075720/ [17:33] lool: that should fix the other bug, testing now === Namidairo is now known as Namidairo`bnc [17:34] stgraber: Do I understand correctly that we have to be careful to preserve the ordering of uids in images? [17:34] stgraber: that is, if a rootfs has a new system user taking over the current number of usermetrics, it will take over ownership of its files? [17:35] stgraber: nice that you got rid of the hardcoded phablet uid/gid though! [17:35] lool: yeah, we unfortunately will have problems if uids start changing [17:36] lool: once we have the boot time hooks, we'll be able to fix any problem in there though. So those packages that have non-root owned files divert to persistent storage should ship a hook to check and fix them [17:36] lool, stgraber: I guess 20130907 won't have a different mediascanner? [17:38] plars: if that's part of android, then no. The 20130906.1 => 20130907 is rootfs only [17:39] stgraber: no, it's not... but when lool and asac and I were talking last night, I thought the working theory was that the new failures were likely related to mediascanner getting updated in 6.1 [17:40] the new camera test failures to be specific [17:41] plars: http://people.canonical.com/~ogra/touch-image-stats/20130907.changes so no [17:44] stgraber: ack; we should eventually add some kind of image tests to avoid getting hurt by this uid thing [17:44] yeah, that'd be useful [17:44] we might generally need to review other things, but hard to tell in advance [17:45] AFAIK usermetrics is the first non-root directory (outside of /home obviously), but I guess probably not the last [17:45] (e.g. maintainer scripts doing stuff that we need to do too) [17:45] lool: plars: nice! [17:45] lool: guess the chown for camera didnt land :)? [17:47] root@ubuntu-phablet:/# ls -lh /var/lib/ | grep usermetrics [17:47] drwxr-xr-x 2 usermetrics usermetrics 4.0K Sep 7 17:47 usermetrics [17:47] root@ubuntu-phablet:/# touch /var/lib//usermetrics/abc [17:47] root@ubuntu-phablet:/# rm /var/lib//usermetrics/abc [17:47] root@ubuntu-phablet:/# [17:47] lool: ^ [17:48] hmm, the mode is wrong, it should have been 750... [17:49] ah of course if I forget the chmod that won't help... [17:51] root@ubuntu-phablet:/# ls -lh /var/lib/ | grep usermetrics [17:51] drwxr-x--- 2 usermetrics usermetrics 4.0K Sep 7 17:50 usermetrics [17:51] now that's good! Uploading [18:04] stgraber: cool [18:04] hello [18:04] ? [18:05] hi [18:17] does anyone know where I can find information on inverse mouse area? I tried searching on google and upstream qt docs but couldnt find them [18:17] while I see the calculator app using it [18:19] what is the newest way to install? anything special like flipped/nonflipped i need to know for phablet flash? [18:20] if anyone could show me how to get apparmor working on clickapps by default I would.. really appreciate it.. my kernel doesn't have apparmor, and the .desktop files seem to get regenerated on every boot.. [18:20] I know it is /usr/share/upstart/sessions/applications-click.conf and override related [18:42] boiko: ping [18:42] nik90: pong [18:43] boiko: hi, I am not sure if you remember, but almost 2 months ago, you provided instructions for bug https://bugs.launchpad.net/ubuntu-clock-app/+bug/1179486 [18:43] Launchpad bug 1179486 in Ubuntu Clock App "[Clock app] Ubuntu Touch keyboard hides the name preset text box" [Medium,Triaged] [18:44] boiko: your suggestion was to add the KeyBoardRectangle.qml file to the clock-app and anchors elements to it so that they are not hidden by the osk [18:44] nik90: ah yes, but that's not required anymore [18:44] boiko: really? [18:44] boiko: damn I just started working on that and copied that file :P [18:45] nik90: if your app has a MainView, you can just set anchorToKeyboard: true [18:45] boiko: that's it? [18:45] nik90: should be it [18:47] boiko: hmm, that doesn't seem to do it [18:47] boiko: my textfield is still beneath the osk [18:47] boiko: since you are here, do you also happen to know where I can find information about inverse mouse area? [18:48] nik90: well, it is an internal component, so obviously there is no documentation about it [18:49] nik90: but using it is pretty simple: you fill an area with that component, and then it will report a clicked signal that happens anywhere on screen but the area that is filled [18:50] boiko: ah okay. Do you know what the propagateComposedEvents property does? [18:50] boiko: I saw this being used in the calculator to remove focus from the textfield when clicking outside [18:50] boiko: I need that same behavior for clock app as well [18:51] boiko: so I copied the code and it seems to work. However wasnt sure what the propogatecomposedevents property was for [18:52] nik90: hmm, I would say it does the same as it does on MouseArea [18:52] nik90: http://qt-project.org/doc/qt-5.1/qtquick/qml-qtquick2-mousearea.html#propagateComposedEvents-prop [18:52] boiko: ah thnx. did not know that it had that property :) [18:52] nik90: :) [18:52] boiko: so regarding the textfield hidden beneath osk, does the textfield need to be anchored to bottom of the app? [18:53] nik90: either that, or if it is in a flickable, you need to scroll for it to be in view [18:53] boiko: it is in a flickable [18:53] boiko: but I was hoping it would scroll automatically [18:54] nik90: I used to have this in phone-app, let me see if I can find the code, just a sec [18:54] boiko: http://bazaar.launchpad.net/~phablet-team/phone-app/trunk/view/head:/src/qml/DetailViewCommunication/CommunicationView.qml#L235 [18:54] boiko: I took this from our previous conversation 2 months ago :D [18:54] nik90: ah yes, that's it :) [18:55] boiko: so in this file, you say anchor.bottom: keyboard.top [18:55] nik90: ah wait, well, I was looking for a different piece of code, let me find it here [18:58] nik90: so, that code (from the URL you pasted) is for the case of when your textfield is outside a flickable [18:58] nik90: but when it is inside a flickable, you need to get it on screen. I had a code for something similar here: http://bazaar.launchpad.net/~phablet-team/phone-app/trunk/view/head:/src/qml/DetailViewContact/ContactDetails.qml#L233 [18:59] boiko: which line should I be looking at? [18:59] nik90: well, that function I wrote is to jump the view to a position inside a repeater or something like that [18:59] scrolltosectionposition() ? [18:59] nik90: yep [19:00] nik90: but I can help you with that on monday if you need [19:00] nik90: but right now I am running late for an appointment :/ [19:00] boiko: okay..go ahead [19:00] nik90: have a nice weekend! :) [19:00] boiko: that's fine. thnx for the link [19:00] u 2 boiko [19:01] if surfaceflinger is never loading, just "waiting..." on my new port, is that a bug on my porting side or not? (im using flipped images now) [19:03] lool: stgraber: so 30 might contain the chown? great :) [19:04] lool: stgraber: what is odd that the camera-app autopilot consistently succeeds on RW [19:04] asac: I'm uploading android now to pick up the chown/chmod change, so any build done in 3+ hours should have it [19:04] (even with crashes) [19:04] asac: but I doubt it'll make it for the next tick (in an hour) [19:04] right [19:05] android takes 3 hours to build and propose? [19:05] good to know [19:06] lool: stgraber: camera app is crashy on RW as well, but the one test always fails on RO and always succeeds on RW [19:06] compare: http://reports.qa.ubuntu.com/smokeng/saucy/image/4050/camera-app-autopilot/ [19:06] (RW) [19:06] vs. http://reports.qa.ubuntu.com/smokeng/saucy/image/4058/camera-app-autopilot/ [19:06] RO [19:06] really oddish [19:07] probably more like 2 hours, it'll take me quite a few minutes to upload (the source is 150MB large), then the build takes around 50min and probably around 30min for the processing (copying files around, britney, ...) [19:07] yeah. all good. guess tomorrows build is then pretty good [19:07] I hope so [19:07] i just cant see why a random test would fail on RO :) [19:07] disable_shoot_button [19:07] maybe that tries to save a pref? [19:08] :) [19:08] I just tried the camera app here on mako and it works fine [19:08] both front and back camera, shooting cameras, switching flash on/off and video mode [19:08] right. i did it too [19:08] not sure what "disable_shoot_button" is though [19:09] really weird that it fails 100% on RO and 0% on RW :) [19:09] on both devices [19:10] asac: what's the full name of the test? [19:13] sergiusens: test_capture.TestCapture.test_shoot_button_disable [19:14] sergiusens: see http://reports.qa.ubuntu.com/smokeng/saucy/image/4058/camera-app-autopilot/ [19:14] vs http://reports.qa.ubuntu.com/smokeng/saucy/image/4050/camera-app-autopilot/ [19:23] asac: that test is very fragile [19:24] asac: still in process of flashing latest to see why this happens, but it could be that taking photos is faster on touch_ro [19:35] sergiusens: shoot_button_disable is about taking pictures? [19:35] also i assume you wanted to say that touch is faster than touch_ro? [19:35] or do yuou think touch_ro is fastre in this regards? [19:36] sed -i [19:41] asac: takes a picture, checks if the shutter button is disabled mid flight and enabled again [19:42] asac: but there is no guarantee an autopilot inspect would capture that at the right moment [19:42] asac: I think this test adds little value IMO [19:42] sergiusens: testing here, the button is greyed out for <1s so seems like a potential race indeed [19:43] stgraber: yeah, very likely [19:48] stgraber: are you running daily-proposed with no mods? [19:52] sergiusens: I'm ony daily-proposed with my initrd change applied locally (writable /var/lib/usermetrics) [19:53] stgraber: ah, let me do that [19:58] Anyone have the link so I can download this on my sprint samsung galaxy s3? [19:58] sergiusens, asac: So mediascanner crash is gone? [19:59] hmm apparently not [19:59] What really [20:03] I'm going to keep on looking for it [20:20] I wonder whether it's the removal of demo medias that causes the mediascanner to fail to start/init correctly [20:21] lool: demo content is still there [20:23] I've posted a bt with symbols in https://bugs.launchpad.net/mediascanner/+bug/1221984 [20:23] Launchpad bug 1221984 in mediascanner "Crashes under automation" [Undecided,New] [20:23] and some debug logs [20:23] I see gst_discoverer is NULL, but not sure why it isn't set [20:31] lool: reliable on RO ... not on RW [20:31] even though camera doesnt crash [20:31] err works [20:34] asac: but should work with stgraber's fix, right? [20:42] sergiusens: the chown? i dont think so [20:42] the autopilots run in RW mode [20:42] and without that fix it doesnt work at all her [20:42] and autopilot succeeds, just for one test not [20:45] asac: thought you were talking about your device [20:48] sergiusens: hey, is there any way to just wipe the userdata? [21:02] cwayne: not really, since the image builder constructs the users home too [21:03] cwayne: we can certainly write one for touch_ro [21:07] sergiusens: no ... locally all is fine... just that test :_) === Fyodorovna is now known as wilee-nilee [21:07] asac: I get to crash mediaservice locally [21:08] could be it happens here too [21:08] but ... i can take photos :) [21:08] etc. [21:08] seems that the RW tests also produce crashes [21:08] but still they all succeed [21:08] right maybe the mediascanner thing is unrelated [21:08] hence: wondere whats going on with that test :) [21:08] feels its unrelated to the test [21:08] asac: that test http://bazaar.launchpad.net/~phablet-team/camera-app/trunk/view/head:/tests/autopilot/camera_app/tests/test_capture.py#L139 [21:08] asac: is not good [21:08] but since the day before the RO images passe [21:08] d [21:09] it is still one of the candidates i feel [21:09] asac: does something need to rerun? [21:09] asac: two asserts on the same thing at the end there [21:09] cjohnston: dont think so for now [21:09] ack [21:09] asac: expecting two different results and nothing triggering the change [21:09] sergiusens: hell [21:09] how can that suceeed at all? [21:09] asac: not a good candidate for autopilot testing [21:09] asac: might of been slower before [21:10] asac: very race prone [21:10] ah eventually waits [21:10] i see [21:10] so they try to figure if the button goes disabled [21:10] and then comes back [21:11] otoh i get a stall right after button is disabled if camear app crashes :) [21:11] asac: yup, but between the time you get to the point where you check it could have already happened [21:12] sure? doesnt it just return if its true already? [21:13] asac: yeah but what happens with self.assertThat(exposure_button.enabled, Eventually(Equals(False))) if it's already True? [21:13] it times out [21:13] oh ic [21:13] so we fail in the False line? [21:13] but why would RO be faster? [21:13] maybe it fails fast to save? [21:14] asac: that's a callback, the shutter disable has to do with the sensor being ready or not [21:14] so IIUC, asac is looking after the failing autopilot test which is broken by design and needs to be fixed to not be racy, and I was looking at a mediascanner crash that is not a blocker for anything; in any case camera-app and gallery-app work, right? :-) [21:14] sergiusens: thatd be very helpful for me :D [21:14] asac: anyways, looking at the test, the return to True is what fails on utah [21:15] lool: latest install of daily-proposed doesn't have the camera-app working for me out of the box [21:16] sergiusens: well still the chown missing, but that's uploaded, right? [21:16] lool: not sure, haven't been keeping track of that [21:16] 20130905-0ubuntu4 [21:16] is still in proposed [21:17] in fact still building! [21:17] but near the end [21:17] (android pkg) [21:25] actually seems to be something wrong with this build [21:36] build was stuck; doubt we will have time to sort this + build an image tonight [21:36] next image tomorrow morning should be good though [22:07] sergiusens: so the fact that you had to manually install python:any indicates a bug in apt; :any is definitely supposed to take the native version when possible [22:07] sergiusens: and yeah, I saw the same behavior here [22:39] hmm odd, on mako if I shutdown -h now from adb shell with full battery, it will restart after some secs into the "battery charging" screen, then again into the "full battery" screen, then after a longer while it automatically boots again [22:47] Hey! I started ubuntu touch on my samsung galaxy tab 2 but after flashing only a blank screen shows up. Any Ideas how to solve it? [22:48] Seems like Quantal Release is working but saucy dont: http://forum.xda-developers.com/showpost.php?p=45207701&postcount=358 [23:15] Any ideas? [23:18] asac, lool, stgraber: looks like camera app is still failing on 7.1 [23:19] plars: yeah, I don't see what would cause it to suddenly work (except if you get lucky, assuming it's indeed a race) [23:20] stgraber: ok, I misunderstood, I thought some sort of fix was landing for it [23:21] plars: I pushed a fix to the initrd and android to make /var/lib/usermetrics properly writable [23:21] but it's not in any image yet as we had some problems with Launchpad that prevented anything from publishing for a few hours [23:21] ok, I thought lool was saying that just changing the permissions made it at least somewhat better for him yesterday [23:21] ah, ok [23:22] ah, if that's the case, then the next image will work better. Though I fail to see how the camera app would be related to that. [23:22] I'll kick a rebuild once android is ready, so we should have a new image in ~1h30