[07:50] <dholbach> good morning
[07:56] <Kivi> dholbach, morning
[07:57] <dholbach> hi Kivi
[09:30] <DanChapman> morning all o/
[09:39] <andve> asd
[14:19] <Kivi> Happiness... :)
[16:12] <jplatte> Hi, is anybody online who works on the Music App?
[16:12] <jplatte> (core app)
[16:12] <ahayzen> jplatte, o/ i do
[16:13] <jplatte> ahayzen: I guess you are aware that the App currently can't find any new files on the SD card?
[16:14] <jplatte> I thought the app broke for me, but then I saw some reviews mentioning this as well. But it seems to be since a system update, not an App update?
[16:14] <ogra_> most likely not the apps fault
[16:14] <ahayzen> jplatte, i don't have a device with any SD card support but we have bug 1508873
[16:15] <ogra_> mediascanner is acting up in various ways recently it seems
[16:15] <ahayzen> it seems mediascanner2 has broken something for *many* people... expect popey ...
[16:15] <ahayzen> *except
[16:15] <popey> uh
[16:15] <ogra_> he's just lying :P
[16:15] <davmor2> ahayzen: hasn't broken for me either
[16:15] <popey> bug 1513756
[16:15] <popey> filed today
[16:15] <ahayzen> popey, but can yours could read files on the SD card?
[16:15] <popey> on OTA-7?
[16:16] <ahayzen> ok its broken for everyone except for QA and any devs lol
[16:16] <ahayzen> popey, yeah
[16:16] <jplatte> Well I still have my old files in the Music App
[16:16] <popey> yup
[16:16] <jplatte> Fortunately.. Although I tried about everything to reset the app because I thought that might help :D
[16:16] <ahayzen> comment #22 in bug 1508873 looks potentially like the issue
[16:17] <ahayzen> jplatte, would you be able to pastebin your mediascanner2.log or send it to me ?
[16:17] <jplatte> Yeah it's the corrent bug. Thanks :)
[16:17] <jplatte> ahayzen: Wanted to ask if I could help next. I'll gladly collect some logs :)
[16:17] <popey> what do i need to do? nuke the music, wipe ms2 db and put it back?
[16:18] <ahayzen> popey, people are just stating that they update to OTA7 and suddenly the music-app won't see anything on the SD card
[16:18] <davmor2> ahayzen: http://people.canonical.com/~davmor2/phone-screenshots/working-sdcard-scan.png see I have no music on the device only on sdcard
[16:18] <popey> odd
[16:18] <popey> i updated like a normal person
[16:18] <ahayzen> davmor2, thats the scope not music-app
[16:18] <popey> this is my retail bq phone
[16:18] <ahayzen> davmor2, that'll be using the cpp not qml mediascanner interface
[16:18] <ahayzen> popey, like a "normal" person ;-) lol
[16:19] <popey> I can be normal, sometimes.
[16:19] <jplatte> ahayzen: No the music app does see everything on the SD card that was added to the library previously
[16:19] <ahayzen> jplatte, oh your saying stuff that has been *added* cannot be seen?
[16:20] <jplatte> And even erasing the music apps config, cache, local/share data, and .local/share/mediascanner doesn't change that the old stuff is still there :D
[16:20] <jplatte> Yeah exactly
[16:21] <davmor2> ahayzen: http://people.canonical.com/~davmor2/phone-screenshots/working-music-app-sdcard.png
[16:21] <davmor2> :P
[16:21] <ahayzen> jplatte, i guess that could still be the same issue, it'd still be in the mediascanner db and then if it is having trouble mounting it then it wouldn't be able to scan it. If you could retrieve the log in ~/.cache/upstart/mediascanner-2.0.log that could be useful
[16:21] <ahayzen> davmor2, :-P
[16:21] <popey> (and older logs in there too)
[16:21] <popey> they may have been rotated out
[16:21] <ahayzen> ah yeah and that ;-)
[16:21] <popey> in my bug report I concatenated all the gzipped logs
[16:21] <jplatte> Okay, how do I copy files over adb?
[16:22] <jplatte> Actually...
[16:22] <popey> i use nautilus
[16:22] <ahayzen> adb pull /path/on/device /path/locally
[16:22] <davmor2> ahayzen: I also have all my music in the music app and scope on my dogfood phone that has been ota'd from ota2
[16:22] <ahayzen> davmor2, ok but if you now add something to the SD card does it get picked up?
[16:22] <jplatte> Heh
[16:22] <jplatte> remote object '/home/phablet/.cache/upstart/mediascanner-2.0.log' does not exist
[16:22] <ahayzen> i'm wondering if ms2 is infact looping and 'stuck' like it was on popey's so then its not finding anything new
[16:23] <jplatte> And no results for find -name "mediascanner*.log"
[16:23] <ahayzen> jplatte, you may have to $ ls | grep mediascanner  that directory as popey said the logs may have been rotated
[16:23] <davmor2> ahayzen: let me try on my dogfood phone
[16:23] <ahayzen> davmor2, thanks
[16:23] <jplatte> Yeah okay, there are .log.*.gz files
[16:24] <ahayzen> yeah bring those over :-)
[16:25] <jplatte> Heh
[16:25] <jplatte> One error
[16:25] <jplatte> A lot of times
[16:26] <jplatte> terminate called after throwing an instance of 'std::runtime_error'
[16:26] <jplatte>   what():  UNIQUE constraint failed: media_attic.id
[16:26] <ahayzen> ah interesting
[16:26] <ahayzen> that's similar to what comment #22 said
[16:26] <ahayzen> jplatte, could you get a copy of the actual db itself as well? .. let me find the path
[16:27] <ahayzen> jplatte, /home/phablet/.cache/mediascanner-2.0/mediastore.db .. as that may be useful for the ms2 guys to debug with
[16:28] <davmor2> ahayzen: so I just copied across 5 folders of music and they all show in the music app
[16:28] <ahayzen> davmor2, interesting ... if you look at the end of this log https://launchpadlibrarian.net/223103287/mediascanner-2.0.log i see.. Volume /media/phablet/UBUNTU-SD was mounted.
[16:28] <ahayzen> terminate called without an active exception
[16:28] <popey> how are people copying the music?
[16:28] <popey> permissions?
[16:28] <jplatte> ahayzen: Posted in on the bug
[16:28] <ahayzen> the last thing on that log was .. Extracting metadata from /home/phablet/Videos/Gerald Troost - Ons moment.webm. .. popey when ms2 looped on your desktop was it scanning a webm ?
[16:29] <popey> haha
[16:29] <popey> yes
[16:29] <ahayzen> \o/
[16:29] <ahayzen> i blame that then :-P lol
[16:29] <popey> don't put webm on the phone, that'll kill mediascanner
[16:29] <popey> yes, that will do it
[16:29] <ahayzen> jplatte, do you have webm on your phone?
[16:29] <popey> bug 1513756
[16:29] <jplatte> Pretty certainly not
[16:29] <jplatte> But I'll check
[16:29] <davmor2> I used mtp
[16:30] <jplatte> argh
[16:30] <jplatte> mtp doesn't wanna work
[16:30] <jplatte> actually, a shell and find will serve me better for this anyway ^
[16:30] <jplatte> ^^
[16:31] <jplatte> FUUUUUU adb shell, FUUUUU
[16:31] <ahayzen> jplatte, can you add you latest ms2 log as well to the bug
[16:31] <jplatte> Okay, so I just checked, no .webm in /home/phablet or /media/phablet/SD\ 32GB
[16:31] <jplatte> Where do I find that log?ß
[16:31] <ahayzen> If you are copying via adb you should check the permissions are correct but i thought that worked now that adb goes to the phablet user
[16:32] <ahayzen> jplatte, that was the .cache/upstart/mediascanner*.log*.gz thing ?
[16:32] <jplatte> I'm confused
[16:32] <jplatte> Ooh ms2 = mediascanner 2 :D
[16:32] <ahayzen> yah
[16:33] <jplatte> so I should upload all the logs?
[16:33] <ahayzen> jplatte, you could but the bug already has like 10 logs .. maybe just extract the last one and post to pastebin.ubuntu.com then i can check it quickly
[16:35] <jplatte> what's the xzcat for gzip called?
[16:35] <jplatte> ugh, whatever will just copy the file
[16:36] <jplatte> http://pastie.org/10534469
[16:36] <ahayzen> hah ok :-)
[16:37] <ahayzen> jplatte, can go go through the files and find what the one where it first started erroring with that "what():  UNIQUE constraint failed: media_attic.id" ?
[16:38] <jplatte> what do you mean, going through the files?
[16:39] <ahayzen> jplatte, i assume you have .log.1.gz .log.2.gz .log.3.gz etc ?
[16:39] <jplatte> *sigh*
[16:39] <ahayzen> hah
[16:39] <ahayzen> which one was that log you posted from ?
[16:40] <jplatte> which what?
[16:40] <jplatte> which log file?
[16:40] <jplatte> .log.1.gz
[16:40] <ahayzen> ah cool
[16:40] <ahayzen> so what does the oldest one say ?
[16:40] <jplatte> wait I need to find out what xzcat for gzip is, copying via adb then unpacking with a graphical program is barbaric
[16:41] <ahayzen> yeah it is :-/
[16:41] <jplatte> Okay, it's z*
[16:41] <jplatte> so zcat, zmore, zless, ...
[16:41] <ahayzen> \o/
[16:42] <jplatte> the oldest one doesn't look any different
[16:43] <jplatte> but now I can tell you SINCE HOW LONG I HAD TO DEAL WITH THIS SHIT:
[16:43] <ahayzen> ok it most have been rotated out
[16:43] <jplatte> 2015-10-20 11:56:50.
[16:43] <jplatte> :D
[16:43] <ahayzen> :-)
[16:43] <ahayzen> we need a mediascanner person to look at this bug badly  :-/
[16:44] <ogra_> and you think if he looks badly the bug gets scared and goes away ?
[16:45] <ahayzen> popey, anyway we can raise the priority of bug 1508873 its currently on the "image backlog", or should i include it in my mass email rant to the ms2 guys ;-) ?
[16:47] <popey> its in progress, right?
[16:47] <popey> and high, so, they know
[16:47] <ahayzen> "confirmed" and "high"
[16:47] <ahayzen> popey, but ideally we would want this in OTA8 otherwise it could be ages before this is fixed?
[16:48] <popey> I agree. pmcgowan ^
[16:48] <popey> pmcgowan is the magic wand that can raise priority on things IME
[16:48] <ahayzen> :-)
[16:49] <pmcgowan> popey, will look in a bit
[16:51] <jplatte> Somebody want to give me a bzr clone-or-whatever-it's-called URL for mediascanner? I'd like to look at the code but I think I'd need an hour to find it myself on launchpad :>
[16:51] <ahayzen> jplatte, https://code.launchpad.net/mediascanner2 and the bzr path lp:mediascanner2
[16:52] <jplatte> thanks
[16:52] <jplatte> heh, don't even have bzr installed
[16:52] <ahayzen> jplatte, even if you can find where/why it is exploding i'm sure that would help alot
[16:53] <jplatte> I'll see what I can do
[16:53] <jplatte> ^^°
[16:53] <ahayzen> thanks :-)
[16:53] <jplatte> What's clone called in bzr-land?
[16:53] <ahayzen> bzr branch lp:mediascanner2
[16:54] <jplatte> thanks
[16:59] <jplatte> Uhhh
[16:59] <jplatte> media_attic.id..
[17:00] <jplatte> INTEGER PRIMARY KEY
[17:00] <jplatte> are the mediascanner guys manually creating primary key IDs or does sqlite have auto-increment / serial without specifiying it explicitly??
[17:01] <ahayzen> jplatte, sqlite should auto-increment IIRC from the table it looks like it is auto done
[17:02] <jplatte> well the error message tells me that sqlite doesn't do that.. what part of it looks like it's auto-done?
[17:03] <jplatte> oh god, what is this sql table?
[17:03] <ahayzen> idk i haven't really looked
[17:03] <jplatte> it has fields for images, fields for audio files and potentially other stuff
[17:04] <jplatte> and everything that doesn't apply is just set to default
[17:04] <jplatte> or actually, maybe not default but 0
[17:04] <jplatte> because I don't see any `NOT NULL`s, so NULL should be default
[17:06] <jplatte> I'll backup the db and see what happens when mediascanner tries to create ir from scratch
[17:06] <jplatte> s/ir/it
[17:06] <ahayzen> jplatte, if jamesh is ever around on IRC he is the ms2 guy (note his timezone is Australia and he usually hangs around in #ubuntu-touch)
[17:06] <vthompson> jplatte, hey there. I work with ahayzen. What do you want to know?
[17:06] <ahayzen> jplatte, and vthompson maybe able to help ;-)
[17:07] <jplatte> can I force mediascanner to rerun?
[17:08] <vthompson> Yep "restart mediascanner-2.0" should still work
[17:09] <jplatte> what?
[17:09] <jplatte> /sbin/restart ?
[17:10] <jplatte> That says `restart: Unknown instance: `
[17:10] <jplatte> I also tried sudo service mediascanner-2.0 restart
[17:10] <jplatte> which didn't work either
[17:10] <jplatte> (unrecognized service)
[17:10] <vthompson> hm, let me connect to my device
[17:11] <ahayzen> phablet@ubuntu-phablet:~$ restart mediascanner-2.0
[17:11] <ahayzen> mediascanner-2.0 start/running, process 9321
[17:11] <ahayzen> worked for me
[17:11] <ahayzen> oh maybe your mediascanner2 crashed and wasn't running?
[17:12] <jplatte> ooooh I have gdb on the UT device :)
[17:12] <jplatte> Of course it crashed, numerous times. Like the log says
[17:13] <ahayzen> IIRC you can run $ mediascanner-service-2.0  and that will run it in your console so you can see the output
[17:13] <jplatte> HEy I can start it directly :)
[17:13] <jplatte> Yeah that's what I meant
[17:13] <jplatte> So I can also run it in gdb, catch throw, and see  where the error comes from! :)
[17:13] <ahayzen> \o/
[17:13] <jplatte> It takes aaages
[17:14] <jplatte> and I haven't even run it inside gdb yet
[17:14] <popey> yeah, ms2 runs as you, not root.
[17:14] <popey> so no need for sudo
[17:14] <jplatte> ooor I can't
[17:14] <jplatte> apparently it just throws constantly
[17:15] <vthompson> It worked for me. I'm on rc-proposed... but the service name shouldn't have changed
[17:15] <jplatte> using exceptions for control flow hooray
[17:15] <vthompson> http://paste.ubuntu.com/13126319/
[17:15] <jplatte> Yeah start would probably work for me
[17:15] <jplatte> But now i wanna debug
[17:15] <vthompson> jplatte, cool
[17:16] <jplatte> Which is harder than it has to be because the things just constantly throws exceptions before getting to the actual problematic part
[17:16] <jplatte> And I'll have to go soon
[17:16] <jplatte> I'll see if I can limit the scope of catch throw somehow
[17:18] <vthompson> jplatte, so ms2 is repeated throwing exceptions? You see that in their log I assume?
[17:18] <jplatte> Nope
[17:18] <jplatte> Throwing exceptions doesn't mean they're logged
[17:18] <jplatte> WHAT
[17:19] <jplatte> I just looked at the function throwing exceptions for each file
[17:19] <jplatte> The first line in there... is a label for goto
[17:20] <jplatte> (I'm talking about Scanner::next())
[17:20] <jplatte> Yeah, control flow with exceptions: throw StopIteration();
[17:29] <jplatte> *sigh*
[17:29] <jplatte> I can't debug this properly
[17:30] <jplatte> No debugging symbols, exceptions are being thrown constantly...
[17:31] <jplatte> Hey wait, maybe I can. In a really awful way..
[17:39] <jplatte> If I would change to a -devel image, would that get me debugging symbols in mediascanner?
[17:40] <ahayzen> jplatte, probably not, IIRC you have to make your image writeable and then install the packages. But i'm no expert on that
[17:40] <jplatte> then install the packages? which packages?
[17:41] <jplatte> I have no problem with making the image writable. Are you saying I could build mediascanner myself, on the phone?
[17:41] <ahayzen> jplatte, possibly libmediascanner-2.0-dev
[17:41] <jplatte> Huh
[17:41] <jplatte> I'll see if that helps me, but I have a feeling it won't
[17:42] <ahayzen> yeah, hmm who would know how to get debug symbols on the device popey ?
[17:43] <jplatte> well I know, you get debugging symbols by building the library yourself, with the right compiler flags :D
[17:43] <jplatte> But I don't have a cross-compiler toolchain, have never used one with CMake and don't really want to anyway
[17:43] <jplatte> Though it would be interesting if I could build it on the device directly, I'll see if I can find gcc and cmake in the phone's repos
[17:44] <jplatte> Huh, yeah apparently they're available in the UT repos
[17:44] <jplatte> Let's see...
[17:44] <popey> we strip debug symbols unless specified otherwise
[17:44] <popey> I could probably build you a version of ms2 in a ppa which has dbgsym not stripped
[17:45] <jplatte> nah, I'll try to build it myself :)
[17:45] <popey> okay, ping if you need anything
[17:45] <popey> happy to help0
[17:45] <popey> <- note: not a developer
[17:45] <jplatte> thanks! :)
[17:45] <popey> np
[17:45] <popey> I personally build arm stuff in a chroot on my desktop
[17:47] <jplatte> good to know that there are people with a working cross-compiling setup :D
[17:48] <jplatte> do you know how to build apps using the normal framework with CMake for armhf too?
[17:49] <jplatte> I have been burned enough by the SDKs failure to even build for my smartphone to not use the SDK before there's a really big update, but maybe I could use the app skeleton I have and just work on it without the SKD?
[17:49] <jplatte> s/SKD/SDK
[17:49] <jplatte> Would be really nice, I don't like qtcreator anyway, and even had to set up a VM for all that stuff
[17:50] <jplatte> (I should probably mention: I'm now talking about my attempt to create a UT app, has nothing to do with mediascanner :D)
[17:51] <popey> i dont use qtcreator
[17:51] <popey> I use my own script for building stuff in the schroot
[17:51] <jplatte> But have you created / built UT apps?
[17:51] <popey> yeah
[17:51] <jplatte> cool
[17:52] <jplatte> well I think I'll have enough other stuff to do now and the next days, but I might ask you about that again :)
[17:52] <popey> click chroot -a armhf -f ubuntu-sdk-15.04 create
[17:52] <popey> I think that is enough (if you have the sdk (ppa) installed) to create an armhf chroot in which to build stuff
[17:52] <popey> but yeah, ping me if you want help, happy to
[17:53] <mcphail> ahayzen: you don't need the debug symbols on the device
[17:53] <ahayzen> mcphail, jplatte is the one looking for help :-)
[17:53] <mcphail> aah
[17:53] <ahayzen> mcphail, jplatte is trying to get the debug symbols for mediascanner2
[17:54] <jplatte> popey: Oh okay, you're using click.
[17:54] <jplatte> Well then I probably still need to do everything in the VM, but whatever :)
[17:54] <popey> heh
[17:54] <popey> I also sometimes sacrifice a nexus 7 to build on
[17:54] <popey> but I rarely do that these days because it's s o  s l o w
[17:54] <mcphail> jplatte: if there is a .ddeb available for mediascanner2, you can simply download and unpack that on your build machine. You can the point gdb at the symbols
[17:55] <jplatte> mcphail: Oh, well that sounds useful
[17:55] <jplatte> Buut I have no clue if there is a .ddeb available for mediascanner2, or where to search for it
[17:57] <mcphail> https://wiki.ubuntu.com/DebuggingProgramCrash gives some pointers
[17:58] <popey> if ms2 is in the overlay ppa, bet there is no ddeb
[17:58] <popey> because ppa
[17:58] <mcphail> aah - of course
[17:58] <jplatte> *sigh*
[17:58] <jplatte> TLS 1.0
[17:58] <jplatte> why does everyone use outdated TLS? :D
[17:59] <jplatte> anyway, apparently that doesn't help?
[17:59] <jplatte> because ppa?
[17:59] <jplatte> but building on the device also is running into problems at the moment
[17:59] <popey> yeah, the phone is on 15.04 of ubuntu with extras added via a ppa
[17:59] <jplatte> installing -dev packages doesn't help cmake find the header files apparently
[18:00] <mcphail> jplatte: you make have to hack the pkg-config search path
[18:01] <jplatte> oh, okay
[18:01] <jplatte> where are the pkgconfig files installed to=
[18:03] <mcphail> /usr/lib/arm-linux-gnueabihf/pkgconfig/
[18:21] <vthompson> ahayzen, you want to review this? lp:~vthompson/music-app/fix-1511585-part-2
[18:21] <vthompson> AP is still broken
[18:21] <ahayzen> vthompson, ah yes :-)
[18:21] <vthompson> :)
[18:22] <ahayzen> vthompson, is that the new ms2 package that appeared in rc-proposed :')
[18:22] <ahayzen> vthompson, mediascanner2.0/vivid 0.108+15.04.20151102-0ubuntu1 armhf [upgradable from: 0.107+15.04.20150922.1-0ubuntu1] right ?
[18:22] <vthompson> It should be
[18:22] <ahayzen> as my lappy is still vivid so have to use device to test
[18:23] <vthompson> Yep, the 20151102 update
[18:23] <ahayzen> coolio i'll update and run AP
[18:23] <vthompson> Godspeed ;)
[18:47] <jplatte> *sigh*
[18:47] <jplatte> libdbus-cpp-dev is installed and the .pc file is there, cmake doesn't find it (but finds all other packages)
[18:47] <jplatte> actually, it finds all other pkg-config packages
[18:48] <jplatte> doesn't find qt5core, but haven't tried installing something there, don't know what to install
[18:48] <jplatte> gonna give up for today
[23:54] <jplatte> > popey: I could probably build you a version of ms2 in a ppa which has dbgsym not stripped
[23:54] <jplatte> Was just remembering this. I don't really know how I'd continue trying to build it myself, so would be nice if I didn't have to ^^
[23:55] <jplatte> I'll go to bed soon and won't be online until tomorrow late evening (CET) though.