Beirdobot, irc.freenode.net

Wednesday, July 2nd, 2014, 00:17 UTC

[00:17:28]

moparisthebest (moparisthebest!~quassel@cpe-74-140-228-202.swo.res.rr.com) has joined #mythtv

[00:20:21]

poptix (poptix!poptix@poptix.net) has quit (Ping timeout: 244 seconds)

[00:22:29]

poptix (poptix!poptix@poptix.net) has joined #mythtv

[00:24:25]

rsiebert (rsiebert!~quassel@f052137152.adsl.alicedsl.de) has joined #mythtv

[00:27:32]

rsiebert_ (rsiebert_!~quassel@f052000161.adsl.alicedsl.de) has quit (Ping timeout: 260 seconds)

[00:40:25]

t0mahawk (t0mahawk!60f4fcb6@gateway/web/freenode/ip.96.244.252.182) has quit (Ping timeout: 246 seconds)

[00:40:51]

arescorpio (arescorpio!~arescorpi@89-26-245-190.fibertel.com.ar) has joined #mythtv

[00:53:03]

jafa2 (jafa2!~jafa@50.242.92.249) has quit (Ping timeout: 240 seconds)

[01:03:52]

tris (tris!tristan@2001:1868:a00a::4) has quit (Quit: Leaving)

[01:21:25]

tris (tris!tristan@2001:1868:a00a::4) has joined #mythtv

[01:23:28]

poptix- (poptix-!poptix@poptix.net) has joined #mythtv

[01:24:47]

poptix (poptix!poptix@poptix.net) has quit (Read error: Connection reset by peer)

[01:25:54]

jafa2 (jafa2!~jafa@c-50-131-89-4.hsd1.ca.comcast.net) has joined #mythtv

[01:41:20]

andreaz (andreaz!~andre_000@p57922329.dip0.t-ipconnect.de) has quit (Read error: Connection reset by peer)

[01:49:10]

gigem (gigem!~david@mythtv/developer/gigem) has quit (Quit: WeeChat 0.4.3)

[01:50:24]

gigem (gigem!~david@mythtv/developer/gigem) has joined #mythtv

[02:23:33]

CrypticSquared (CrypticSquared!~CrypticSq@unaffiliated/crypticsquared) has quit (Ping timeout: 255 seconds)

[02:41:25]

CrypticSquared (CrypticSquared!~CrypticSq@unaffiliated/crypticsquared) has joined #mythtv

[02:49:42]

CrypticS_ (CrypticS_!~CrypticSq@unaffiliated/crypticsquared) has joined #mythtv

[02:49:51]

peper03_ (peper03_!~peper03@mythtv/developer/peper03) has joined #mythtv

[02:50:10]

CrypticSquared (CrypticSquared!~CrypticSq@unaffiliated/crypticsquared) has quit (Ping timeout: 264 seconds)

[02:53:08]

peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 260 seconds)

[02:53:08]

peper03_ is now known as peper03

[02:57:11]

CrypticS_ (CrypticS_!~CrypticSq@unaffiliated/crypticsquared) has quit (Quit: Leaving...)

[03:16:07]

CrypticSquared (CrypticSquared!~CrypticSq@unaffiliated/crypticsquared) has joined #mythtv

[03:17:24]

arescorpio (arescorpio!~arescorpi@89-26-245-190.fibertel.com.ar) has quit (Excess Flood)

[03:40:58]

fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 240 seconds)

[03:41:38]

aloril (aloril!~aloril@dsl-tkubrasgw2-54f80b-12.dhcp.inet.fi) has joined #mythtv

[03:42:25]

fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv

[03:54:57]

jafa2 (jafa2!~jafa@c-50-131-89-4.hsd1.ca.comcast.net) has quit (Ping timeout: 272 seconds)

[04:57:05]

jafa2 (jafa2!~jafa@c-50-131-89-4.hsd1.ca.comcast.net) has joined #mythtv

[05:06:33]

jafa2 (jafa2!~jafa@c-50-131-89-4.hsd1.ca.comcast.net) has quit (Ping timeout: 240 seconds)

[05:45:53]

jya (jya!~jyavenard@mythtv/developer/jya) has quit (Quit: jya)

[05:53:58]

SteveGoodey (SteveGoodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has joined #mythtv

[06:09:55]

dekarl:

jya, the observation from https://code.mythtv.org/trac/ticket/12166 appears to be correct

[06:10:00]

SteveGoodey (SteveGoodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has quit (Quit: Konversation terminated!)

[06:10:39]

dekarl:

hm, bad timing

[06:10:59]

FabriceMG (FabriceMG!~Thunderbi@LCaen-156-54-30-212.w80-11.abo.wanadoo.fr) has joined #mythtv

[06:38:00]

Tobbe5178 (Tobbe5178!~asdf@h104n2-sv-a13.ias.bredband.telia.com) has joined #mythtv

[06:46:48]

FabriceMG (FabriceMG!~Thunderbi@LCaen-156-54-30-212.w80-11.abo.wanadoo.fr) has quit (Ping timeout: 255 seconds)

[06:47:48]

FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has joined #mythtv

[07:16:34]

purserj (purserj!~purserj@hosting.collaborynth.com.au) has quit (Ping timeout: 264 seconds)

[07:17:05]

purserj (purserj!~purserj@hosting.collaborynth.com.au) has joined #mythtv

[07:17:44]

xris (xris!~xris@mythtv/developer/xris) has quit (Ping timeout: 260 seconds)

[07:28:00]

Chutt (Chutt!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has quit (Read error: Connection reset by peer)

[07:28:45]

Chutt (Chutt!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has joined #mythtv

[08:00:24]

stuarta:

unpicking the plugin build issue is a right pita

[08:08:12]

Merlin83b (Merlin83b!~Daniel@office.34sp.com) has joined #mythtv

[08:11:32]

paul-h (paul-h!~Paul@94.6.239.28) has joined #mythtv

[08:37:16]

xris (xris!~xris@xris.forevermore.net) has joined #mythtv

[09:15:17]

paul-h:

natanojl: do you want me to commit that widget border fix?

[10:04:48]

FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has quit (Quit: FabriceMG)

[10:30:42]

dekarl1 (dekarl1!~dekarl@p4FCEE936.dip0.t-ipconnect.de) has joined #mythtv

[10:33:16]

dekarl (dekarl!~dekarl@p4FE859B0.dip0.t-ipconnect.de) has quit (Ping timeout: 260 seconds)

[11:34:13]

jya (jya!~jyavenard@mythtv/developer/jya) has joined #mythtv

[11:44:46]

jya:

stuartm: 4813104b2bbc0861fe59a9c07952ca2798dbe35e isn’t required for me, i had checked tht mytharchive and mytharchive helper compiled just fine (both their .pro have that line already)

[11:46:00]

jya:

it seems that it’s a link to libmythhdhomerun that is required there…

[11:46:27]

stuarta:

jya: it gets quite messy, i was attempting to fix it last night but didn't get very far

[11:46:38]

jya:

maybe I should just move MythAVCopy and MythFrame to libmythbase

[11:46:59]

stuarta:

jya: we though libmythav would be a good idea

[11:47:13]

stuarta:

which is what i was trying.

[11:47:30]

stuarta:

hadn't resolved some link issues in programs tho

[11:47:42]

stuarta:

and you are right, stuartm's fix didn't make any difference

[11:48:01]

jya:

all those cross dependencies in lib is a nightmare to compile

[11:48:15]

jya:

add one lib, and you got to link 20 more

[11:48:23]

stuartm:

they really don't belong in libmythbase

[11:48:35]

jya:

i know they don't...

[11:49:06]

stuartm:

jya: which is one thing we're trying to eliminate, if the recorders aren't in libmythtv then you don't also have to link livemedia, hdhomerun etc etc

[11:49:08]

jya:

but i find it less ugly than linking stuff that should be unecessary

[11:50:07]

jya:

shouldn’t the linker be smart enough to realise that MythAVCopy has no dependency to any other code, and not attempt to link everything else?

[11:52:00]

stuartm:

I'd like to think it should, but at least on some platforms that's not the case, and those platforms it is building on probably only work because it's automatically linking all dependent libs rather than determining they aren't actually required

[11:55:26]

stuarta:

jya: there was cheating before, the relevant code was just copied into the plugin itself

[11:55:42]

jya:

i’m going to look into it

[11:56:16]

stuarta:

i've already got a .pro file for libmythav if you want it

[11:57:27]

jya:

ok

[11:57:36]

jya:

should we move the audio there too?

[11:58:22]

jya:

or at least the few utility classes provided by the audio class that are FFmpeg helper

[11:58:23]

stuarta:

http://fpaste.org/114951/30227014/

[11:59:09]

jya:

there will be a required addition to mythframe.cpp and mythframe.h

[11:59:20]

stuarta:

i think that makes sense

[11:59:28]

jya:

mythavutil make use of it

[12:00:25]

jya:

will also certainly requires link to the ffmpeg libraries

[12:00:48]

stuarta:

biab, fetching lunch

[12:01:19]

stuartm:

jya: I was thinking a libmythaudio, but since you can't have audio without something doing the decoding first, having it in libmythav instead and keeps the number of new libs low

[12:02:08]

stuarta:

you also have to watch what libmyth* are provided by our ffmpeg copy, that cuts down the number of options

[12:02:17]

jya:

and then all the decoder related classes then

[12:02:40]

jya:

(going to have libmythtv renamed into libmythav soon :) )

[12:03:22]

stuartm:

jya: yeah, that was my thinking, basically libmythtv would disappear and at a minimum we'd have libmythav and libmythrecorder(s)

[12:04:39]

stuartm:

possibly another lib if there's still code left in libmythtv that's not properly part of either of those two, only think that comes to mind atm is metadata related and would be at home in the existing libmythmetadata

[12:04:49]

stuartm:

s/think/thing/

[12:07:54]

stuartm:

libmythrecorders would have to link libmythav, but not the other way around

[12:09:10]

stuartm:

if we get it right mythfrontend and plugins wouldn't ever need to link libmythrecorders and all it's dependencies

[12:16:53]

** stuarta is back **

[12:17:48]

jya:

can leave the recorder in libmmythtv and move everything else out...

[12:18:50]

paul-h:

Couldn't we just move that function out of mytharchivehelper and put it into one of the libs that we already link to or put it in mythutil?

[12:19:25]

paul-h:

Or mythpreviewgen wou;d probably be a better fit

[12:21:43]

jya:

paul-h: you would still need to link to mythtv regardless

[12:34:45]

wagnerrp (wagnerrp!~wagnerrp_@mythtv/developer/wagnerrp) has quit (Read error: Connection reset by peer)

[12:35:28]

wagnerrp (wagnerrp!~wagnerrp_@mythtv/developer/wagnerrp) has joined #mythtv

[12:40:06]

paul-h:

jya: mythutil and mythpreviewgen already link to libmythtv so there is no problem there

[12:40:25]

jya:

oh, i’m referring to the plugins

[12:41:14]

jya:

mytharchive has a lot of duplicated code there to decode and extract a frame

[12:42:11]

jya:

actually, the inclusion of using_live: LIBS += -lmythlivemedia-$$LIBVERSION

[12:42:12]

jya:

using_mheg: LIBS += -lmythfreemheg-$$LIBVERSION

[12:42:13]

jya:

in the top mytharchive.pro serves no purpose as this .pro isn’t used to build the actual plugin

[12:44:07]

paul-h:

Not sure what you are referring to buth there is the plugin and mytharchivehelper it's the latter with the problem

[12:45:59]

paul-h:

The grabbing of the thumbnails needs to be changed to use the new filter stuff to remove the depreciated code anyway

[12:48:50]

jya:

paul-h: both mytharvice and mytharchive helper now make use of mythavutil (they both use MythAVCopy). mythutil and mythpreviewgen have always linked to libmythtv in the past. i added the link to libmythtv for mytharchive only a couple of days ago

[12:50:18]

jya:

it’s weird it fails for mytharchivehelper but not mytharchive/mytharchive

[12:50:36]

jya:

as both use the same

[12:51:02]

stuartm:

the issue being that in linking libmythtv, they also need to link all the libs that are linked into libmythtv – libmythhdhomerun etc

[12:51:28]

jya:

yes, i’ve just addressed that in my commit

[12:51:49]

stuartm:

that was more for paul-h

[12:52:17]

jya:

what i’m saying is that mytharchive/mytharchive like mytharchive/mytharchivehelper, both link to libmythtv

[12:52:24]

jya:

yet only one fail to do so

[12:53:24]

jya:

it must be used something else from libmythtv that is causing the linkage issue, but considering the only thing changed lately with the use of MythAVCopy, why it worked in one and not the other is a mystery

[12:53:25]

stuartm:

something to do with one building as an app and the other a lib?

[12:53:39]

jya:

ah yes...

[12:53:41]

jya:

make sense

[12:53:43]

stuarta:

wtf

[12:54:46]

stuartm:

stuarta: mytharchivehelper is an application that's run by the plugin (library) mytharchive

[12:55:11]

stuarta:

ah, and we have to build the plugins as libraries?

[12:55:47]

jya:

will know in 7 minutes if that was enough...

[12:55:51]

stuarta:

:)

[12:56:33]

jya:

aren’t clang and icc buildbot not using the same linker as the others?

[12:57:32]

stuartm:

stuarta: current plugin architecture has plugins as libs that are dynamically loaded at runtime with dl_open

[12:57:54]

stuarta:

stuartm: makes sense

[12:58:12]

stuarta:

jya: i thought they used a whole different toolchain (ld included)

[12:58:41]

jya:

ah could be… surprising

[12:58:47]

jya:

obviously doing a smarter job

[12:59:15]

stuarta:

sheesh, you can't see the compiler output for all the warnings the clang builder throws building the plugins

[12:59:49]

stuartm:

fwiw, I believe the reason mytharchivehelper was written as a lib instead of being internal was because of historical instability in the decoder that meant it would crash occassionally – that's unlikely to still be the case

[13:00:16]

stuarta:

seems like a candidate for merging

[13:02:32]

paul-h:

mytharchivehelper is also used by the mythburn.py python script so it's just easier if it a separate app

[13:03:00]

stuartm:

wonder if we can catch SIGSEGV in plugin libs and simply unload then without bringing down the whole frontend

[13:03:27]

stuartm:

paul-h: noted

[13:06:28]

** jya checking the open tickets over the past month to see if i’ve missed patches **

[13:08:48]

stuarta:

paul-h: i was meaning mytharchive plugin

[13:11:03]

stuarta:

jya: that didn't work by the looks of it. question i have is does mytharchive even need libmythmheg or libhdhomerun?

[13:11:19]

jya:

no… but it needs libmythtv

[13:11:22]

stuartm:

stuarta: might be wise to wait until I've fixed the huge list of bugs in the last merged plugin ...

[13:11:27]

jya:

and libmythtv needs them

[13:11:36]

stuarta:

doh

[13:11:56]

stuarta:

different question then, what does mytharchive need from libmythtv?

[13:12:05]

jya:

if the build failed, it means with_freemheg etc aren’t read properly

[13:12:12]

jya:

MythAVCopy

[13:12:29]

stuarta:

doh! i was looking at that last night me--

[13:12:34]

stuartm:

stuarta: my fault really, I was so keen to see it happen that I didn't really give the patch (mythgallery v.2) enough attention

[13:12:44]

jya:

its the new class I added that takes care of convertion for frame format

[13:13:30]

jya:

the code was duplicated everywhere, using a single static lock, causing in effect all the format conversion to be serialised

[13:13:46]

jya:

on my machines, that cause playback to stutter when i have a pip opened

[13:13:48]

stuarta:

yes i should have remembered that

[13:14:23]

jya:

plus now, once I made a single adjustment to speed up frame conversion, it will be enabled everywhere

[13:15:15]

jya:

there’s still things I’m unahppy about … like when using VDA, the decoder output YUVYV frame, it gets converted to YV12 internally, to get converted to YUVYV again for display

[13:15:52]

stuarta:

is that an api deficiency or just a side effect of the format we use?

[13:15:59]

jya:

our VideoBuffers class requires that all internal frame to be YV12, so on most systems with hardware decoding, we do the conversion twice

[13:16:10]

** stuarta facepalms **

[13:16:48]

jya:

if we got the decoding to be made in the same format we will require for display, that already will save a fair bit.

[13:17:24]

jya:

on a h264 here with hardware decoding, 30% of the mythfrontend usage is purely spent one that double conversion

[13:18:02]

jya:

now that I have the frames 32 bytes aligned, ffmpeg uses SSE accelerated code, that sped things quite significantly though

[13:18:19]

stuartm:

that YV12 conversion is purely so we can composite the OSD?

[13:18:36]

jya:

no.. it’s just what we use internally

[13:18:56]

jya:

on windows that’s converted to RGB for display

[13:19:05]

stuartm:

oh, thought we already supported pipelining in a variety of formats

[13:19:06]

jya:

on opengl that depends on the driver etc.

[13:19:13]

jya:

oh we do!

[13:19:22]

jya:

everything is converted to YV12 :)

[13:19:29]

stuartm:

heh

[13:19:44]

jya:

then no format compatibility issue :)

[13:20:11]

stuarta:

we have cpu cycles to burn, dammit lets burn them!

[13:20:12]

stuartm:

are we at least using shaders for the format conversions?

[13:20:16]

jya:

and our opengl render wants YV12, packed

[13:20:32]

jya:

well, there is support for shader and format conversion

[13:20:39]

jya:

but I’m yet to see it happening

[13:20:56]

jya:

everytime the code in opengl.cpp that uses software conversion is used

[13:20:59]

stuartm:

not that shaders are the answer to all the world's problems, but at least we get to hand off the conversion to the GPU

[13:21:16]

jya:

on both my machine with intel, nvidia or amd when i use the opengl render

[13:22:25]

jya:

stuartm: in all my tests: https://github.com/MythTV/mythtv/blob/master/ . . . deo.cpp#L812

[13:22:29]

jya:

is always called

[13:22:54]

jya:

i haven’t checked why that is the case, i’ve concentrated on speeding up the software conversion first.

[13:23:08]

stuartm:

jya: will have to take a look later, I've got to run

[13:23:22]

jya:

if we want myth to ever run on the Pi and other such hardware, we must address that issue

[13:23:28]

stuartm:

yep

[13:40:06]

jya:

dblain: can you give me more details about what the error is in compiling videoout_d3d.cpp ?

[13:46:16]

jya:

stuartm: it seems the what is causing an error for win build is in c:\buildbot\slaves\master-win8-msvc-2010–32bit\build\source\mythtv\externa l\minilzo\lzodefs.h(2085) : error C2118: negative subscript

[13:47:00]

MythBuild_:

build #1822 of master-ubuntu-12_04-lts-64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/1822

[13:47:17]

jya:

yeahh!

[13:48:20]

stuarta:

\o/

[13:55:00]

MythBuild_:

build #5034 of master-freebsd-64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/5034

[13:56:45]

MythBuild_:

build #929 of master-fedora-32bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . t/builds/929

[13:56:50]

MythBuild_:

build #455 of master-f20–64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . t/builds/455

[13:58:53]

jafa2 (jafa2!~jafa@c-50-131-89-4.hsd1.ca.comcast.net) has joined #mythtv

[13:59:29]

jya:

paul-h: I’m going to change audio so it saves the volume in the destructor

[13:59:40]

jya:

player shouldn’t have to worry about that

[13:59:44]

MythBuild_:

build #1025 of master-f19–64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/1025

[14:00:38]

** stuarta demands better io bandwidth on his server **

[14:02:17]

MythBuild_:

build #2133 of master-debian-wheezy-64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/2133

[14:02:37]

poolejj (poolejj!~poolejj@adsl-074-167-126-214.sip.mco.bellsouth.net) has quit (Quit: Leaving)

[14:06:54]

MythBuild_:

build #1799 of master-ubuntu-current-64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/1799

[14:07:58]

MythBuild_:

build #333 of master-f20-qt5–64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . t/builds/333

[14:08:15]

stuarta:

jya: do you need the devel/iptv branch on the builders any longer?

[14:08:19]

MythBuild_:

build #438 of master-freebsd10–64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . t/builds/438

[14:08:32]

jya:

dblain: this seems to be the first commit that caused the msvc to fail: https://code.mythtv.org/buildbot/builders/mas . . . t/builds/792

[14:08:47]

jya:

stuarta: I will at some stage.. just put that change on the backburner

[14:09:05]

stuarta:

no worries. i won't clean it up then

[14:09:21]

jya:

but i think i got most of the build stuff i was worried about

[14:09:29]

jya:

that was to add a new library to external

[14:09:41]

jya:

now that I know that it builds, probably don’t need it anymore

[14:11:44]

stuarta:

lemme know when you are sure

[14:12:10]

poolejj (poolejj!~poolejj@adsl-074-167-126-214.sip.mco.bellsouth.net) has joined #mythtv

[14:16:46]

jya:

i think it’s safe to remove it

[14:17:03]

jya:

maybe I should just have a generic branch just to test compilation

[14:17:31]

stuarta:

devel/testbuild?

[14:17:47]

jya:

it gets a bit tiring to find that you’ve broken everything after a merge

[14:17:51]

paul-h:

jya: do you know if the the internal video player can play an mp3? Just looking at #11597 and it just fails with 'Video frame buffering too many times'

[14:17:51]

** MythLogBot http://code.mythtv.org/trac/ticket/11597 **

[14:18:07]

jya:

paul-h: normally yes it should

[14:18:50]

jya:

mp3 file that is

[14:18:58]

jya:

never tried with a url

[14:19:11]

jya:

its a different ringbuffer that is created then

[14:19:50]

jya:

when would mythfrontend directly attempt to play such file?

[14:20:18]

paul-h:

That's trying to play a podcast in MythNews

[14:21:21]

jya:

why would it think there’s a video stream in there? i wonder if that’s another of those too small initial read that cause it to fail

[14:21:57]

jya:

ahhh it’s that mjpeg stuff.

[14:22:11]

jya:

problem is with those file that contains some artwork

[14:22:25]

jya:

it’s seen as mjpeg and it doesn’t realise that it’s a single frame only

[14:23:06]

jya:

mythmusic handles those just fine cause it ignores it completely

[14:25:30]

paul-h:

OK thanks for looking

[14:26:11]

jya:

i need to find a way to ignore those video streams that really just artwork

[14:26:36]

jya:

maybe could only do so with mp3 file (looking at the extension) and completely ignore that video then

[14:26:50]

jya:

(though woud be nice to show a static image of the artwork)

[14:27:15]

dblain:

jya: That commit fixed the problem that FFmpeg wasn't compiling but was reported as successful (mythtv source never got built prior to commit).

[14:27:26]

jya:

wagnerrp: when is ping() ever called?

[14:27:58]

jya:

dblain: the link to the build shows it as error, with you as last commit. the build just before that was successful

[14:28:00]

dblain:

jya: The current problem is due to the new ffmpeg binaries causing a link issue in libmythtv. I'm slowly working on it

[14:28:25]

jya:

ok..

[14:28:36]

dblain:

if you looked at the logs for the last successful build, it really failed but reported success.

[14:29:05]

dblain:

So we haven't had a good msvc build since the ffmeg merge :(

[14:30:12]

jya:

is that the call to avpriv_strtod ?

[14:30:31]

dblain:

yes, and avpriv_snprintf

[14:30:50]

jya:

let me check

[14:31:35]

dblain:

I've tried a few things, but each one requires 45+ minutes of building before the error shows :(

[14:32:43]

jya:

i see something about those symbols in FFmpeg/configure

[14:32:48]

jya:

but not in our configure

[14:32:56]

jya:

that’s not good

[14:33:19]

jya:

but you’re build FFmpeg separately right?

[14:34:01]

dblain:

yes, I use FFmpeg/configure.

[14:34:43]

jya:

and red herring, was just not looking where it needed

[14:35:04]

jya:

can you build FFmpeg using myth configure? and let it generate the right makefiles?

[14:36:39]

dblain:

The approach I took with the msvc build was to use just the pro files to generate the makefiles. But FFmpeg needs configure to be run in mingw to allow it to generate appropriate msvc make files.

[14:36:44]

jya:

i wonder why we would need to use ffmpeg strtod anyway.. that’s standard stdlib stuff

[14:37:10]

jya:

my guess is that FFmpeg strtod isn’t used anyway

[14:37:15]

dblain:

I can try to run myth's configure in mingw, but I don't believe it will produce usable results for msvc

[14:37:55]

jya:

it should produce identical makefile than ffmpeg one

[14:37:56]

dblain:

I think it's due to a bug in the way the msvc handles NaN in the stdlib impl

[14:38:04]

jya:

except with all the flags we need

[14:38:22]

dblain:

but it also produces gcc make files for all mythtv source and I need nmake makefiles.

[14:38:56]

dblain:

qmake generates proper nmake makefiles from the pro files.

[14:39:04]

jya:

what I’m saying is that everyting in FFmpeg/configure is also in our configure

[14:39:20]

jya:

so if you can generate it with FFmpeg/configure, then so should ours

[14:39:21]

dblain:

understood.

[14:40:09]

dblain:

I can give it a try, but it will effect the build environment for the rest of the msvc build (which I could always work around if need be)

[14:40:34]

jya:

it’s H264Parser.cpp that cause the problem, probably with one of the #include, and it’s missing a define somewhere

[14:41:01]

dblain:

It was just much cleaner to use only the pro files and the ffmpeg specific configure... (I treat ffmpeg as an external library and only compile it when missing/forced)

[14:41:05]

jya:

because H264Parser certainly don’t make a call to avpriv_strtod

[14:41:17]

dblain:

saw that.

[14:42:07]

jya:

but right now, the error I’m seeing on the msvc build are related with minilzo

[14:42:32]

jya:

that used to compile, so it’s obviously missing some fancy flag goodies you added in the new .pro

[14:42:39]

dblain:

minilzo? haven't see that one yet.

[14:42:50]

jya:

that’s in all the failure

[14:43:03]

jya:

c:\buildbot\slaves\master-win8-msvc-2010–32bit\build\source\mythtv\externa l\minilzo\lzodefs.h(2085) : error C2118: negative subscript

[14:43:13]

jya:

that fails early on

[14:44:20]

dblain:

That must be a new external lib... I don't even have it in my local source tree (last pulled 1 week ago)

[14:45:46]

joki- (joki-!~joki@p54860337.dip0.t-ipconnect.de) has quit (Ping timeout: 244 seconds)

[14:47:00]

dblain:

I see it now.

[14:47:02]

jya:

yes.. minilzo was moved to external/

[14:47:18]

dblain:

It looks like a path prefix issue.

[14:48:19]

dblain:

guess I need to do a pull and start over with my build.

[14:50:38]

joki- (joki-!~joki@p54861F9D.dip0.t-ipconnect.de) has joined #mythtv

[14:51:54]

dblain:

jya: any help is appreciated, but FWIW I am working on fixing the build every chance I get... Just need to do it between other tasks with higher priority.

[14:52:14]

poolejj (poolejj!~poolejj@adsl-074-167-126-214.sip.mco.bellsouth.net) has quit (Quit: Leaving)

[14:52:14]

jya:

i need to get myself a build environment

[14:52:21]

jya:

VC2010 is a bugger though

[14:52:29]

jya:

I have 2008 and 2013

[14:52:44]

jya:

(not the free version either)

[14:53:27]

dblain:

2013 might be easier to get working (more C99 compliant). I only focused on 2010 due to requests from others since most 3rd party tools only worked with 2010 at the time I implemented this

[14:54:09]

jya:

the minilzo issue is going to obfuscate all other build issue

[14:54:12]

dblain:

I did start to set the pro/ps1 files up to work with 2010, 2013

[14:54:36]

dblain:

I just did a pull so I could look at that issue.

[14:55:22]

dblain:

I can work on getting visual studio 2013 working as well (I have purchased version as well)

[14:56:18]

dblain:

I may take some time to get my main machine setup to build as well... currently working in a VM which takes forever to compile

[15:02:43]

stuartm:

dblain: it was in libmythtv, got moved to external and updated to the latest version

[15:03:08]

dblain:

jya: looks like qt is available for vs2013 with OpenGL now... (that was the other issue, had to compile qt from trunk to work with 2013)

[15:03:18]

jya:

cool

[15:03:21]

stuartm:

I'm not sure whether we even need it on Windows since it's only used internally by the old analogue recorder which doesn't work on windows anyway

[15:04:00]

stuartm:

oh and mythtranscode for transcoding to nuv

[15:04:08]

stuartm:

but who wants to do that?

[15:05:59]

** dblain need to go out and fix the pool... be back later to check current compile status **

[15:08:23]

stuartm:

"LZO 2.08 has been released; just a small update that corrects some reported build problems." might be worth checking whether that has any effect

[15:12:56]

stuartm:

nope, doesn't even touch lzodefs.h

[15:13:36]

stuartm:

just fixes some endian issues

[15:23:37]

jya:

stuartm: stichnot still uses the nuv transcode

[15:30:59]

stuarta:

crazy person

[15:35:50]

wagnerrp (wagnerrp!~wagnerrp_@mythtv/developer/wagnerrp) has quit (Read error: Connection reset by peer)

[15:36:31]

wagnerrp (wagnerrp!~wagnerrp_@mythtv/developer/wagnerrp) has joined #mythtv

[15:46:12]

jya:

trying to reproduce #12153, mythtranscode gives me an error: E No video information found! and E Please ensure that recording profiles for the transcoder are set

[15:46:12]

** MythLogBot http://code.mythtv.org/trac/ticket/12153 **

[15:46:35]

jya:

any ideas what that means? do i must create a recording profile? don’t one exist by default?

[15:47:36]

jya:

and where are those set?

[15:47:42]

jafa2 (jafa2!~jafa@c-50-131-89-4.hsd1.ca.comcast.net) has quit (Ping timeout: 255 seconds)

[15:47:54]

** jya can never get away around our settings, that says a lot :( ) **

[15:52:32]

stuartm:

mythtv-setup

[15:52:47]

stuartm:

it's one of the main menu items

[15:53:30]

stuartm:

but a set of default recording profiles should already be set up

[15:55:45]

stuartm:

jya: a few people do use nuv transcodes, mainly because we don't offer any alternative codecs/containers, if we did then we could drop it from mythtranscode entirely

[15:56:20]

stuartm:

however of those few people who use it, almost none will be on windows because you can't record on windows so there's nothing to transcode

[15:56:30]

jya:

stuartm: i needed to create a lossless profile. now i have mythtranscode.

[15:57:01]

jya:

iv’e fixed the crash, but why it gets there I’m not sure… looks like an audio stream is dropped mid-recording

[15:57:11]

stuartm:

jya: odd, a default lossless profile _should_ be created, if not that's a bug or regression

[15:57:22]

jya:

stuartm: well, you can certainly record with the hls recorder

[15:57:49]

jya:

that’s what i use on mac for all my testing

[15:58:02]

stuartm:

jya: that can happen ... several broadcasters provide more than one audio stream – either an alternate language or audio description, when they aren't being used some broadcasters turn them off

[15:58:23]

jya:

not sure the best way to handle that, especially if the stream index changes halfway

[15:58:54]

jya:

it’s going to screw our transcode as we use the ffmpeg stream index for storing into our transcoding file

[15:59:11]

jya:

so if stream3 becomes streams2, the transcode will be rubbish

[16:00:15]

jya:

sure enough, i have no audio in that transcoded file

[16:01:45]

stuarta:

hmmm, current cppcheck run, eta 16m, next run starts in 13m time...

[16:02:29]

jya:

should we really drop 0 channel audio stream?

[16:02:39]

jya:

after all, seeing that it’s a lossless transcode

[16:03:03]

jya:

better keep the channels as they are, espeically if it’s likely it’s a ffmpeg mis-detection

[16:03:19]

stuarta:

the "lossless" transcodes drop non audio and video streams

[16:03:51]

jya:

yes.. but it also drops 0 channels audio

[16:03:59]

jya:

and the logic in there seems broken

[16:04:28]

jya:

we use the avstream index as reported by ffmpeg, and store that in aFrame[x].

[16:04:57]

jya:

but if an audio stream got dropped, the index of the stream isn’t going to match the index in aFrame

[16:05:44]

jya:

say you have stream 1,2,3 and we drop 2. aFrame is created for 2 streams, we then attempt to store in aFrame[3] for all stream 3

[16:07:43]

jya:

we do store it in aud_map, but it’s not used later in GetFrame

[16:15:46]

stuarta:

do you ever sleep?

[16:16:40]

jya:

i’m in the US

[16:17:01]

jya:

and no.. with jetlag, i’ve been up since 3

[16:17:03]

stuarta:

ah, that explains it

[16:17:13]

jya:

so i have to keep myself busy

[16:17:25]

jya:

not the best way to get back to sleep for sure

[16:17:36]

jya:

time for breakfast anyway

[16:20:42]

jarle (jarle!~jarle@70.84-234-133.customer.lyse.net) has quit (Quit: Leaving)

[16:24:12]

jarle (jarle!~jarle@70.84-234-133.customer.lyse.net) has joined #mythtv

[16:28:52]

jafa2 (jafa2!~jafa@50.242.92.249) has joined #mythtv

[16:38:18]

jya:

what does the lossless trnscode drop?

[16:38:40]

jya:

subtitles, mhi stuff like that?

[16:39:01]

jya:

i wonder if we could always use the ffmpeg mpegts demuxer then

[16:44:29]

jya (jya!~jyavenard@mythtv/developer/jya) has quit (Quit: jya)

[16:47:26]

jya (jya!~jyavenard@mythtv/developer/jya) has joined #mythtv

[16:49:02]

stuartm:

jya: we don't read any settings with -p, since at that point we have no access to the database, we do read in database connection details from config.xml but nothing else

[16:49:19]

jya:

stuartm: i was suspecting so...

[16:49:49]

jya:

i’m just so worried with patches supplied these days, that always turn out to break some obscure feature elsewhere

[16:49:57]

stuartm:

I've been meaning to save language/country to config.xml too so we don't need to keep prompting with -p

[16:50:26]

jya:

and what about the prompt backend stuff?

[16:50:26]

stuartm:

jya: yeah, especially a patch like that one which touches something so fundamental as we cache settings

[16:50:54]

stuartm:

jya: we only prompt for the backend when we're unable to access the database to read the backend IP and port

[16:51:18]

jya:

not if you start mythfrontend with some option, can’t remember which one

[16:51:20]

jya:

-n ?

[16:51:25]

stuartm:

if we can't connect to the database the info in config.xml is likely wrong

[16:51:32]

stuartm:

jya: that's -p

[16:51:39]

jya:

ah ok

[16:51:40]

jya:

my bad

[16:52:45]

stuartm:

-p will first prompt for the language so the following backend connection dialogs will be shown correctly translated

[16:54:32]

jya:

our access to the database, is in any case, far from optimised

[16:54:46]

Merlin83b (Merlin83b!~Daniel@office.34sp.com) has quit (Quit: Leaving)

[16:55:53]

stuartm:

yup

[16:56:10]

jya:

ok.. so it is our demuxer that incorrectly returns the number of channels as 0 for mp3 and ac3

[16:57:39]

jya:

using ffmpeg mpegts native demuxer prevents all crashes and invalid stream detection

[16:59:41]

jya:

stuartm: can you set extra arguments mythconverg is going to be called with ?

[17:10:45]

jya:

ffmpeg mpegts demuxer handle that video much more nicely. no decoding error artifacts etc

[17:16:00]

xris (xris!~xris@xris.forevermore.net) has quit (Changing host)

[17:16:00]

xris (xris!~xris@mythtv/developer/xris) has joined #mythtv

[17:18:17]

clever (clever!~clever@nwcsnbsc00w-142163138075.dhcp-dynamic.FibreOp.nb.bellaliant.net) has quit (Ping timeout: 252 seconds)

[17:30:47]

clever (clever!~clever@nwcsnbsc00w-142163138075.dhcp-dynamic.FibreOp.nb.bellaliant.net) has joined #mythtv

[17:32:44]

SteveGoodey (SteveGoodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has joined #mythtv

[18:05:56]

natanojl_ (natanojl_!~jonatan@mythtv/developer/natanojl) has joined #mythtv

[18:20:54]

jya (jya!~jyavenard@mythtv/developer/jya) has quit (Quit: jya)

[18:25:09]

jya (jya!~jyavenard@mythtv/developer/jya) has joined #mythtv

[18:38:07]

poolejj (poolejj!~poolejj@adsl-074-167-126-214.sip.mco.bellsouth.net) has joined #mythtv

[18:44:58]

jya:

paul-h: I see that in 0.27, there is the possibility to play media from a storage group.

[18:45:07]

jya:

so what exactly are the new changes in 0.28?

[18:46:11]

jya:

or it wasn’t used yet?

[18:46:57]

poolejj (poolejj!~poolejj@adsl-074-167-126-214.sip.mco.bellsouth.net) has quit (Quit: Leaving)

[18:50:16]

SteveGoodey (SteveGoodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has quit (Ping timeout: 260 seconds)

[18:50:16]

Steve-Goodey (Steve-Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has joined #mythtv

[19:00:55]

paul-h:

jya: you could play tracks from a storage group but all the other stuff like scanning for files and coverart was missing

[19:01:16]

jya:

how would you have added a file ?

[19:01:23]

jya:

from the SG that is

[19:04:07]

paul-h:

IIRC you could fiddle it by scanning a directory as normal and then creating a Music storage group to point to that directory and it would use that – it was never publicised so no one should have been using it in 0.27

[19:05:44]

jya:

i see

[19:06:02]

jya:

i’ve backported my recent changes to avfd there anyway

[19:06:08]

jya:

having a bigger buffer can only help

[19:06:26]

jya:

especially with #11995

[19:06:26]

** MythLogBot http://code.mythtv.org/trac/ticket/11995 **

[19:18:17]

paul-h:

jta: don't know if it my end or yours but in all your commit messages the apostrophes get changed to odd characters

[19:26:25]

jya:

am I jta ? :)

[19:26:32]

jya:

like which commit ?

[19:26:54]

jya:

is that just in the commit log or in the file itself?

[19:31:56]

paul-h:

All commits with a ' character – they look OK in cgit so it's probably thunderbird can't handle whatever encoding the commit messages are sent in

[19:32:31]

Steve-Goodey (Steve-Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has quit (Ping timeout: 244 seconds)

[19:32:46]

Steve-Goodey (Steve-Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has joined #mythtv

[19:33:33]

paul-h:

[19:35:12]

paul-h:

I't no big deal just curious if it's just me or others ses it as well

[19:35:42]

jya:

maybe using utf-8

[19:36:00]

jya:

I’ve seen xcode converting — into a — character

[19:42:04]

sanmarcos (sanmarcos!~sanmarcos@unaffiliated/sanmarcos) has left #mythtv ()

[19:42:14]

Steve_Goodey (Steve_Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has joined #mythtv

[19:42:25]

Steve-Goodey (Steve-Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has quit (Ping timeout: 272 seconds)

[19:48:03]

sphery:

xris: did you realize you sent that message to -dev list, and not developers

[19:49:11]

xris:

ah dammit.. trying to do too many things at once.

[19:49:21]

sphery:

hehe

[19:49:23]

xris:

I even double checked but in my brain -dev was the correct one

[19:49:37]

xris:

oh well.. users can get some free ones.

[19:49:54]

xris:

that's twice now that I've done that.

[19:50:04]

sphery:

yeah, too bad though we can't make sure they go to users who have helped with the project somehow

[19:52:04]

sphery:

anyway, not a big deal for me--I have been paying for mine each year specifically so I can say, "It's the best $25 I spend each year" (especially when people complain that "information wants to be free" and the like)

[19:59:35]

xris:

I can always generate more if mythtv wants to specifically give some away

[20:12:52]

knightr:

xris, that's so funny, I was wondering if, for once, I would use one (I never did, I always pay for my subscription), ROTFL... ;-)

[20:13:19]

xris:

I haven't paid yet. gave myself an expiration of 2023 or something crazy like that.

[20:14:01]

xris (xris!~xris@mythtv/developer/xris) has left #mythtv ("Goodbye")

[20:14:10]

xris (xris!~xris@mythtv/developer/xris) has joined #mythtv

[20:14:12]

xris:

oops.

[20:14:19]

knightr:

:-)

[20:14:53]

xris:

my computer should know that I want to close the window I'm *looking* at, not the one that actually has focus.

[20:30:19]

stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv

[20:50:23]

cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has quit (Quit: Konversation terminated!)

[20:57:08]

dekarl1 is now known as dekarl

[21:02:16]

jya (jya!~jyavenard@mythtv/developer/jya) has quit (Quit: jya)

[21:11:25]

dekarl:

stuarta, what about a try scheduler for the buildbot? we can then kick off builds for commits that are not even pushed yet (and later even hook it into merge requests)

[21:15:14]

Chutt (Chutt!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has quit (Read error: Connection reset by peer)

[21:15:41]

Chutt (Chutt!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has joined #mythtv

[21:19:44]

robink (robink!~quassel@unaffilated/robink) has quit (Remote host closed the connection)

[21:19:53]

robink (robink!~quassel@unaffilated/robink) has joined #mythtv

[21:22:06]

jafa2 (jafa2!~jafa@50.242.92.249) has quit (Read error: Connection reset by peer)

[21:26:01]

jpabq:

stuartm: In MythRenderOpenGL2::DrawRoundRectPriv() Mark did shrunk the QRect r area by half of the line width. That does not do the right thing. If you have a buttonlist with a line width of 15, then each button ends up having a gap between them. Getting rid of the +/- halfline when defining that area makes it draw correctly.

[21:27:05]

jpabq:

With it the way it is, is is impossible for a theme to have two button areas butt right up next to each other (at least in a buttonlist).

[21:27:44]

jpabq:

"Fixing" it though, could cause a theme that has tried to compensate for that behavior, to suddenly look wrong, though.

[21:27:53]

jpabq:

Thoughts?

[21:28:22]

Steve_Goodey (Steve_Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has quit (Quit: Konversation terminated!)

[21:33:01]

jpabq:

The similar problem exists with the OpenGL1 renderer.

[21:59:16]

Tobbe5178 (Tobbe5178!~asdf@h104n2-sv-a13.ias.bredband.telia.com) has quit (Read error: Connection reset by peer)

[22:04:15]

stuartm:

jpabq: it's late here so I'll have to take a look at it tomorrow, I'm too tired :)

[22:04:53]

jpabq:

stuartm: NP. I completely understand. Good night.

[22:08:48]

stuartm:

fwiw, in principle I've no problem with breaking themes to fix layout/positioning bugs, at the end of the day it's more frustrating to themers, new and old, that they have to implement tricks/hacks for stuff to work correctly and I'm sure they won't really mind

[22:09:16]

stuartm:

but I really should take a look at the code to understand exactly what the issue is

[22:12:53]

MythBuild_:

build #831 of master-win8-msvc-2010–32bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . t/builds/831

[22:18:02]

jpabq:

stuartm: all three renders have the problem. The thicker the line, the larger the gap between the buttons.

[22:21:03]

paul-h:

Something must have changed because I know the shapes lined up exactly as they should at least in the qt and opengl painters don't know about the opengl2 one

[22:23:45]

wagnerrp:

jya: i think we might call it to see if the database is alive

[22:28:34]

jya (jya!~jyavenard@mythtv/developer/jya) has joined #mythtv

[22:32:16]

paul-h (paul-h!~Paul@94.6.239.28) has quit (Quit: Konversation terminated!)

[22:35:03]

jya:

paul-h: your addition of --checkrecordings is awesome… however, the call to mythcommflag should include the current application path, otherwise it won’t work if the PATH isn’t set or you have two different mythtv version installed

[22:46:12]

natanojl_ (natanojl_!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 255 seconds)

[22:59:01]

dreamcat4 (dreamcat4!~dreamcat4@62.49.10.153) has joined #mythtv

[23:05:33]

moparisthebest (moparisthebest!~quassel@cpe-74-140-228-202.swo.res.rr.com) has quit (Ping timeout: 240 seconds)

[23:06:12]

jya (jya!~jyavenard@mythtv/developer/jya) has quit (Quit: jya)

[23:09:50]

jya (jya!~jyavenard@mythtv/developer/jya) has joined #mythtv

[23:40:30]

jya (jya!~jyavenard@mythtv/developer/jya) has quit (Quit: jya)

[23:47:08]

Chutt (Chutt!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has quit (Ping timeout: 240 seconds)

[23:47:31]

Chutt (Chutt!~ijr@cpe-76-190-199-73.neo.res.rr.com) has joined #mythtv

[23:57:58]

moparisthebest (moparisthebest!~quassel@cpe-74-140-228-202.swo.res.rr.com) has joined #mythtv

MythLogBot@irc.freenode.net :: #mythtv

Current users (68):