**** BEGIN LOGGING AT Thu Mar 31 02:59:58 2016
**** BEGIN LOGGING AT Thu Mar 31 03:29:41 2016
Mar 31 04:45:18 <Herrie|2> morning
Mar 31 04:45:26 <DougReeder> Good morning
Mar 31 04:45:41 <Herrie|2> Tofe: Well it's always QtWebEngine that crashes :P
Mar 31 04:45:57 <Herrie|2> I guess you'd need to get deep into it again :P
Mar 31 04:46:07 <Herrie|2> That larger HD/SSD isn't a luxury I guess :p
Mar 31 04:46:21 <Herrie> DougReeder: morning ;)
Mar 31 04:46:26 <Herrie> Not sure it's a good one yet ;)
Mar 31 04:46:35 <DougReeder> :-S
Mar 31 04:46:40 <Herrie> Little one had a stomach bug since Saturday :S
Mar 31 04:46:58 <Herrie> Seemed better yesterday finally but we'll see when he gets up in a bit :P
Mar 31 04:47:52 <DougReeder> When a little one is sick, no one in the family has a good time.
Mar 31 04:49:47 <Herrie> DougReeder: Yeah and me and the Mrs were flu-ish too :S
Mar 31 04:49:51 <Herrie> So all good fun....
Mar 31 04:49:54 <DougReeder> Oi
Mar 31 04:50:20 <Herrie> But seems he's recovering now finally :) Just he barely ate anything for 4 days....
Mar 31 04:50:29 <Herrie> So well he's lacking a ton of energy
Mar 31 04:50:44 * DougReeder nods
Mar 31 05:07:02 <Herrie> But at least he had quite a bit of food yesterday and all still seems inside, so that's progress ;)
Mar 31 05:07:55 * DougReeder nods
Mar 31 05:44:20 <Tofe> Morning!
Mar 31 05:44:33 <Tofe> (I'm skipping "good" for now ;) )
Mar 31 05:46:39 <Herrie> Tofe: LOL ;)
Mar 31 05:46:56 <Herrie> Well I had a coffee and some time to play with an update to Fingerterm
Mar 31 05:47:02 <Herrie> Had worse mornings :P
Mar 31 05:47:13 <Tofe> :)
Mar 31 05:47:21 <Tofe> You sent the N4 to RMA?
Mar 31 05:50:08 <Herrie> Not yet
Mar 31 05:50:18 <Herrie> Making a service ticket on the website didn't work :S
Mar 31 05:50:40 <Herrie> So filled in their contact form which they said would take 1-2 days to get back to me and that was Tuesday
Mar 31 05:50:49 <Herrie> So I guess I'll hear today from them
Mar 31 05:52:21 <Tofe> ok
Mar 31 06:25:49 <EricBlade> i know this is a bit of an ancient history question, but would anyone be interested in helping advise me in a way to get the old LunaCE builder working?  It seems to have suffered from some bit rot, perhaps
Mar 31 06:27:12 <EricBlade> it utterly fails in it's attempts to get IsisCE, when I give it the correct location for that, it starts spewing shell errors, then i fix that, then it spews Python errors.. so I just removed Isis from the build to see if it would get any farther.. which it did.. but it completely broke down trying to build qt4
Mar 31 06:55:21 <Herrie> EricBlade: SOme logs would be good
Mar 31 07:10:42 <Herrie|Veer> morning
Mar 31 08:04:36 <Tofe> JaMa: you said that debug symbols are no longer copied to sysroots, right?
Mar 31 08:05:58 <Tofe> Is there some advice on how to do a remote debug using build sysroot? Copy needed .debug by hand?
Mar 31 08:10:54 <JaMa> Tofe: correct, sysroot files are stripped just like binaries for target packages
Mar 31 08:11:07 <JaMa> there is script to unpack all -dbg packages in debugfs directory
Mar 31 08:11:11 <JaMa> let me find it
Mar 31 08:12:53 <JaMa> $ rm -rf debugfs ; mkdir debugfs ; cd debugfs
Mar 31 08:12:53 <JaMa> $ for pkg in ../tmp/deploy/rpm/*/*-dbg* ; do
Mar 31 08:12:53 <JaMa> ../tmp/sysroots/x86_64-linux/usr/lib/rpm/rpm2cpio $pkg | cpio -i ; done
Mar 31 08:12:53 <JaMa> $ tar xvfj ../tmp/deploy/images/qemux86/core-image-base-qemux86.tar.bz2
Mar 31 08:12:55 <JaMa> $ cd ..
Mar 31 08:13:00 <JaMa> you'll need to adapt this for .ipk
Mar 31 08:13:05 <JaMa> and your image
Mar 31 08:13:30 <JaMa> $ ./tmp/sysroots/x86_64-linux/usr/bin/i586-poky-linux/i586-poky-linux-gd(gdb) set sysroot debugfs
Mar 31 08:13:33 <JaMa> (gdb) set substitute-path /usr/src/debug debugfs/usr/src/debug
Mar 31 08:14:08 <Herrie|Veer> Tofe: I found some "interesting" bits in journalctl. Touch.webkitRadiusX is deprecated. Same for webkitRadiusY, webkitRotationAngle and webkitForce. Not sure it's Enyo or something else that uses them.
Mar 31 08:18:08 <Tofe> JaMa: thanks!
Mar 31 08:20:00 <Tofe> Herrie|Veer: that would be Enyo, yes; especially the part about multitouch
Mar 31 08:23:03 <Herrie|Veer> Tofe: OK
Mar 31 08:23:20 <Herrie|Veer> I'll check to see if they addressed it in Enyo 2.7
Mar 31 08:31:31 <Tofe> JaMa: or I could also do things like linking the .debug of the -dbg directory in packages-split to /usr/lib/.debug
Mar 31 08:31:43 <Tofe> (well, or copying it, I don't know yet what would be the smartest)
Mar 31 08:35:22 <JaMa> Tofe: if you're using rm-work then packages-split will be already empty
Mar 31 08:35:48 <JaMa> Tofe: newer oe-core can generate this debugfs automatically, but that IIRC isn't yet in jethro
Mar 31 08:40:44 <Tofe> JaMa: I don't do rm-work
Mar 31 08:41:18 <Tofe> I'm often tinkering in the work directories to hack around :)
Mar 31 10:19:01 <Herrie|Veer> Tofe: I'm playing a bit with Fingerterm to see if we can bump to more recent since there were plenty of commits since we forked. Might solve the cursor position issue along the way :) We might need a patch or 2 in our recipe to bypass the Mer/Nemo stuff.
Mar 31 10:19:02 <Herrie|Veer> Hopefully will have something tonight to play with :)
Mar 31 10:23:02 <Herrie|Veer> Just my build environment was messed up from toying with it too much :P
Mar 31 10:58:51 <Tofe> Herrie: ok :)
Mar 31 11:34:44 <Herrie|Veer> Tofe: Would like to patch the font for example, because the default one is horrible :P
Mar 31 11:36:46 <Herrie|Veer> And a few other small things. Will have a go at it :) Might even send something upstream :)
Mar 31 11:36:47 <Herrie|Veer> Seems their current build does Meego and Mer so they could easily have a few LuneOS specific lines of code in there :P
Mar 31 12:19:30 <Herrie|Veer> Seems it uses very limited bits specific for Mer :)
Mar 31 12:25:11 <Tofe> Herrie|Veer: we'll see, I'll first try to see what is crashing on jethro+qt5.6
Mar 31 12:27:37 <Herrie|Veer> Tofe: Yeah :) Since I don't have my Mako working I cannot do that much currently anyway so will look into some smaller things. Still have a Maguro & TP though :)
Mar 31 12:40:27 <Herrie|Veer> Tofe: My guess it's QtWebEngine somehow :P
Mar 31 12:51:37 <Tofe> yes, pretty sure also
Mar 31 12:53:42 <JaMa> Herrie|Veer: we should enable fontconfig support in qtbase if we don't do it already, qtbase fonts are gone from 5.7 Qt so we can switch now
Mar 31 13:12:54 <Herrie|Veer> JaMa: not sure??? Tofe: ^
Mar 31 13:19:25 <Tofe> I... don't remember if we already enable it
Mar 31 13:20:48 <JaMa> we don't have it enabled yet
Mar 31 13:21:05 <JaMa> lets switch separately after jetro/5.6 upgrade
Mar 31 13:22:08 <Tofe> yes, one thing at a time, otherwise we'll never congerge
Mar 31 13:22:12 <Tofe> converge
Mar 31 13:22:38 <Herrie|Veer> Tofe/JaMa: My idea :)
Mar 31 13:22:57 <Herrie|Veer> We'll have a bit of time before 5.7 hits anyway. QT is always late :P
Mar 31 16:57:33 <Herrie> Tofe: ping
Mar 31 16:58:47 <Tofe> Herrie: pong
Mar 31 16:59:06 <Herrie> Tofe: Do we use .spec files in QT? I noticed the Mer guys have some: https://git.merproject.org/mer-core/fingerterm/blob/ec43d3661edf2f44fce6853e95a9419b7c67d0dc/rpm/fingerterm.spec
Mar 31 16:59:33 <Herrie> Seems it's for their build system, but not sure, so thought to ask you :P
Mar 31 17:02:42 <Tofe> .spec is specific to rpm packaging, afaik
Mar 31 17:02:52 <Herrie> OK :)
Mar 31 17:03:35 <Herrie> Good to know :) Seems I have an up to date Fingerterm to test on TP shortly :)
Mar 31 18:10:17 <Tofe> Herrie: looks like we get a crash in wayland when closing a window
Mar 31 18:15:08 <Herrie> Tofe: OK well that's slightly easier to debug then WebEngine right?
Mar 31 18:15:14 <Herrie> I mean wrt amount of code :P
Mar 31 18:15:19 <Tofe> weeell
Mar 31 18:15:40 <Tofe> very, very slightly better
Mar 31 18:15:45 <Herrie> Might be related to the commit you did recently to fix it from not booting at all?
Mar 31 18:17:39 <Herrie> I mean this one: https://github.com/webOS-ports/qtwayland/commit/ef6a77a3be965c5e5ab1bee1cd9fbba68bac52d9
Mar 31 18:18:30 <Herrie> Tofe: We still miss quite some commits from upstream as well
Mar 31 18:18:32 <Herrie> For 5.6
Mar 31 18:18:57 <Herrie> Everything after 12 Feb we miss it seems
Mar 31 18:20:22 <Tofe> not related, but might very well be another bug yes
Mar 31 18:21:06 <Herrie> Tofe: Seems you cherry picked that commit from dev but it's not in 5.6 branch
Mar 31 18:21:14 <Herrie> There's quite a few in 5.6 branch that might be of help
Mar 31 18:22:06 <Herrie> This one is a bit more up to date: https://github.com/webOS-ports/qtwayland/commits/webOS-ports/master-next-newer
Mar 31 18:22:41 <Herrie> Still misses a few commits though
Mar 31 18:24:18 <Tofe> The problem is I don't want this one: https://github.com/webOS-ports/qtwayland/commit/2a0a3f8055754962bf3969229c265f810e3aed1f
Mar 31 18:26:10 <Herrie> Tofe: Ah OK
Mar 31 18:26:13 <Herrie> How about the newer ones?
Mar 31 18:26:17 <Herrie> You're OK with those?
Mar 31 18:26:25 <Herrie> I can update the branch to revert this commit I guess
Mar 31 18:26:53 <Tofe> the others are ok I think
Mar 31 18:29:08 <Herrie> Tofe: OK that's gone: https://github.com/webOS-ports/qtwayland/commit/e411d1acbcfb514857d9339cd169ff9bc1f81838
Mar 31 18:29:17 <Herrie> Let me cherry pick the other few commits
Mar 31 18:34:03 <Tofe> ok
Mar 31 18:34:06 <Tofe> thanks
Mar 31 18:40:54 <Herrie> Tofe: OK done
Mar 31 18:41:02 <Herrie> Not very clean but well... Might clean it up later :)
Mar 31 18:41:08 <Herrie> Let's first see if it solves stuff :P
Mar 31 18:42:23 <Tofe> you included the commit from -next ?
Mar 31 18:43:14 <Tofe> nevermind, I've done the rebase
Mar 31 18:43:42 <Herrie> :D
Mar 31 18:50:58 <Tofe> nope, it's not that
Mar 31 18:51:48 <Tofe> The issue comes from a quite specific case, where a fake window is created by Qt just to flush the remaining painting of a dying window
Mar 31 19:07:40 <EricBlade> Herrie: first error occurs here: http://pastebin.com/eqmT0nxS
Mar 31 19:09:27 <EricBlade> so then i delete everything in the Makefile relevant to isis so it won't attempt to build that .. and i get... (waiting for it to hit that point...)
Mar 31 19:10:44 <Herrie> I didn't even know that https://github.com/openwebos-ports existed :P
Mar 31 19:11:15 <EricBlade> right, correcting that to github.com/woce/ still results in the same syntax error
Mar 31 19:11:23 <EricBlade> and fixing that syntax error results in more errors
Mar 31 19:11:49 <EricBlade> .... and somehow or other, it has decided it's going to build Qt this time
Mar 31 19:11:53 <Herrie> That URL looks like dkirkers private playpen
Mar 31 19:12:04 <EricBlade> it had been failing horrifically on qmake .. but now it's working
Mar 31 19:13:39 <Herrie> Let me see what it does here
Mar 31 19:20:10 <Herrie> EricBlade: This might be easier? http://webos-ports.org/wiki/WOCE_Build_Environment
Mar 31 19:33:41 <Tofe> Herrie: it seems to be either a libhybris bug, or a wayland bug
Mar 31 19:34:56 <EricBlade> Herrie: that's what i followed
Mar 31 19:35:00 <EricBlade> so.. i guess it built now.
Mar 31 19:35:52 <Herrie> Tofe: https://github.com/mer-hybris/droid-config-mako/commit/d8fc0d3ff7b2ea6190c32fec428b34516ab6ef73 That looks interesting for future ;)
Mar 31 19:37:16 <Tofe> yup :)
Mar 31 19:40:31 <Herrie> Tofe: you know if I can run QtCreator 3.5.1 on Ubuntu 14.04?
Mar 31 19:40:44 <Herrie> Seems I got old 3.0.1 now :S
Mar 31 19:41:11 <Tofe> I guess so
Mar 31 19:43:45 <Tofe> morphis: ping ?
Mar 31 19:44:20 <Herrie> nizovn: PR looked good, merged :) Will do some more thorough testing now :)
Mar 31 19:46:55 <nizovn> ok, thanks
Mar 31 19:47:43 <Herrie> I think we'll be pretty OK now :)
Mar 31 19:48:12 <Herrie> Might be a small thing here and there still, but having the files properly is key for the rest to work well :)
Mar 31 19:48:52 <nizovn> i realized that mediaindexer doesn't support unicode tags for now..
Mar 31 19:59:03 <Tofe> Commenting out https://github.com/libhybris/libhybris/blob/master/hybris/egl/platforms/wayland/wayland_window.cpp#L218 does avoid the crash
Mar 31 19:59:15 <Tofe> I'm just not sure it's the right thing to do :p
Mar 31 19:59:20 <Tofe> morphis: ^
Mar 31 20:08:14 <Tofe> Herrie: also we have this issue (not a serious one hopefully): https://bugreports.qt.io/browse/QTBUG-51156
Mar 31 20:13:26 <Herrie|Laptop> nizovn: So Russian doesn't work for example?
Mar 31 20:14:01 <nizovn> yes
Mar 31 20:15:01 <Herrie|Laptop> That's for the tag contents or the tag name?
Mar 31 20:15:24 <nizovn> tag contents
Mar 31 20:15:54 <Herrie|Laptop> We might want to try to bump taglib to at least 1.10 which has this in changelog: Fixed backwards-incompatible change in TagLib::String when constructing UTF16 strings
Mar 31 20:16:27 <Herrie|Laptop> 1.11 Beta 1 and Beta 2 are also out.
Mar 31 20:17:22 <nizovn> i think tablib supports utf8, that should be enough
Mar 31 20:18:07 <Herrie|Laptop> nizovn: Yeah it should be... COuld be they fixed a bug in newer release that's not mentioned explicitly
Mar 31 20:18:20 <Herrie|Laptop> 1.11 Beta 1: "Many smaller bug fixes and performance improvements."
Mar 31 20:18:39 <nizovn> i'm not sure that the problem is in tablib
Mar 31 20:18:45 <nizovn> taglib
Mar 31 20:19:38 <Herrie|Laptop> db8 should be fine I "think"
Mar 31 20:19:47 <Herrie|Laptop> But that's fairly easy to test
Mar 31 20:20:26 <Herrie|Laptop> Looking at the recipe sqlite3 might be an obvious culprit: https://github.com/webOS-ports/meta-webos-ports/blob/fido/meta-luneos/recipes-luneos/mediaindexer/mediaindexer.bb
Mar 31 20:22:56 <Herrie|Laptop> But I see nothing too obvious in sqlite3 recipe
Mar 31 20:23:23 <nizovn> yes, seems db8 should be ok
Mar 31 20:33:20 <Herrie|Laptop> sqlite3 should do utf8 by default as well :S
Mar 31 20:47:44 <nizovn> Herrie|Laptop: seems the problem was due to incorrect tags, and mediaindexer is ok too
Mar 31 20:47:55 <Herrie|Laptop> :)
Mar 31 20:48:17 <Herrie|Laptop> Anyway I'll test a bit. Maybe I'll have a look into upgrading taglib as well :)
Mar 31 20:48:25 <Herrie|Laptop> See if it works and send the recipe upstream :)
Mar 31 20:48:33 <nizovn> ok
Mar 31 21:03:41 <EricBlade> so.. would it be interesting to have Qt5 for webOS 3.0.5 ?
Mar 31 21:09:04 <EricBlade> ... although i suspect that my configure line may be quite wrong, i do seem to have the thing actually building.
Mar 31 21:46:44 <EricBlade> .... and then i run into a "this is fixed in 5.5.1" after building 5.5.0.
Mar 31 22:28:21 <EricBlade> hmm.. so i reset the virtual machine, and now i can't login to it. argh
Mar 31 22:39:25 <Herrie> EricBlade: 5.5.x or 5.6 would be nice. We could bring a new browser :)
Mar 31 22:48:47 <EricBlade> well, this is horse-crap. i halted the vm, restarted it, and now i can't login to it anymore :-S
Mar 31 23:08:35 <EricBlade> brb
Mar 31 23:08:37 <EricBlade> rebooting host
Mar 31 23:14:06 <EricBlade> ... omfg. anyone know how to login to the vagrant box in woce-build, from console, so i can copy the files i changed to the shared disk, and save them, so i don't have to re-figure this out again? :-D
Mar 31 23:43:08 <EricBlade> aha.. had to change the network adapter to Bridged from NAT
Mar 31 23:43:22 <EricBlade> oh, wait, nope, that didn't work
Apr 01 00:14:55 <EricBlade> aha, username/pass is vagrant/vagrant
Apr 01 00:39:30 <EricBlade> this is weird. i've built qt4 about 5 times now, and it's failed 4 times in different places, and one time it worked. despite making no changes.  now i'm stuck in "OpenGL ES 2.0 functionality test failed!" .. grr
Apr 01 01:17:16 <EricBlade> ok, so, i seem to be missing some piece that puts the libGLES* and libEGL* into staging/ from rootfs/ .. interesting.
Apr 01 02:53:50 <EricBlade> no clue what's going on, but i've managed to rope myself into fixing the rest of it too lol
Apr 01 02:54:47 <EricBlade> ... unless i find myself in a situation where i have to download isis multiple times, in which case.. screw that
**** ENDING LOGGING AT Fri Apr 01 02:59:58 2016