IRC Logs

*******************************************************************
[03:00] <walters> hmmm
[03:01] Action: walters wonders if he found the bug in audioconvert
[03:01] sub_pop ([email protected]) joined #gstreamer.
[03:19] steveb_ ([email protected]) joined #gstreamer.
[03:21] Zeenix ([email protected]) joined #gstreamer.
[03:29] <LeRoutier> good night ppl
[03:29] LeRoutier ([email protected]) left irc: "Leaving"
[03:37] steve_b ([email protected]) left irc: Read error: 110 (Connection timed out)
[03:37] spyder482 ([email protected]) left irc: Read error: 110 (Connection timed out)
[04:05] spyder482 ([email protected]) joined #gstreamer.
[04:07] harshy ([email protected]) joined #gstreamer.
[04:16] joeforker (dholth@3ffe:bc0:8000:0:0:0:0:262b) joined #gstreamer.
[04:16] <joeforker> should I expect cvs gst-player to work well?
[04:16] <joeforker> with cvs everything-else-gstreamer?
[04:22] Zeenix ([email protected]) left irc: "The whole problem with the world is that fools and fanatics are always so certain of themselves, but wiser people so full of 
[04:25] <walters> does this patch look sane?
[04:25] <walters> --- gstaudioconvert.c.~1.36.~	2004-03-06 15:48:24.000000000 -0500
[04:25] <walters> +++ gstaudioconvert.c	2004-03-09 22:24:58.456843232 -0500
[04:25] <walters> @@ -484,7 +484,7 @@
[04:25] <walters>          if (this->sinkcaps.sign) {
[04:25] <walters>            CONVERT_TO (cur, src, gint8, this->sinkcaps.sign, this->sinkcaps.endianness, GINT8_IDENTITY, GINT8_IDENTITY);
[04:25] <walters>          } else {
[04:25] <walters> -          CONVERT_TO (cur, src, guint16, this->sinkcaps.sign, this->sinkcaps.endianness, GUINT8_IDENTITY, GUINT8_IDENTITY);
[04:25] <walters> +          CONVERT_TO (cur, src, guint8, this->sinkcaps.sign, this->sinkcaps.endianness, GUINT8_IDENTITY, GUINT8_IDENTITY);
[04:25] <walters>          }
[04:25] <walters>          break;
[04:25] <walters>        case 16:
[04:29] wheels ([email protected]) joined #gstreamer.
[04:31] harshy ([email protected]) left irc: Read error: 113 (No route to host)
[04:40] sub_pop ([email protected]) left irc: "Client exiting"
[04:46] BBB ([email protected]) joined #gstreamer.
[04:48] <BBB> anyone here?
[04:49] <thaytoo> yo
[04:49] <joeforker> hi
[04:50] <BBB> thaytoo: can you confirm that seeking works at all for you in totem?
[04:50] <BBB> with any format
[04:50] <BBB> I don't care which
[04:52] <thaytoo> unfortunately, not from work I can't
[04:53] Action: BBB is gonna send rant emails if the code stays this way
[04:59] <thaytoo> which code?
[05:00] <BBB> osssink
[05:03] <thaytoo> ahr
[05:04] jcsston ([email protected]) joined #gstreamer.
[05:10] steveb__ ([email protected]) joined #gstreamer.
[05:15] walters ([email protected]) left irc: Client Quit
[05:16] <sri_> yes, I suppose that if you're using gint16, you should havd GUINT16_IDENTITY
[05:28] steveb_ ([email protected]) left irc: Read error: 110 (Connection timed out)
[05:32] <joeforker> still no peeps from gstreamer...
[05:33] <BBB> ?
[05:34] <joeforker> I am having a frustrating time getting my gst-all cvs compile to make any sound.
[05:34] <BBB> ?
[05:34] <BBB> as in?
[05:34] Action: BBB is just compiling HEAD right now
[05:35] <joeforker> well I run an example like player and it just freezes, and I've just gotten mad to compile so I can maybe get the mp3 'hello world' examples to work.
[05:36] <BBB> it freezes?
[05:36] <BBB> player in gst-plugins/examples/gstplay/?
[05:37] <joeforker> will try that
[05:37] <joeforker> no the gst-plugins player just waits doing nothing
[05:38] <BBB> how do you start it?
[05:38] <joeforker> ./player ~/Movies/Railroada.ai
[05:38] <joeforker> .avi
[05:38] <BBB> hmk, that should work
[05:39] <BBB> does nothing?
[05:39] <BBB> any messages on the stdout?
[05:39] <joeforker> no
[05:40] <BBB> er...
[05:40] <BBB> CPU usage? 0%? 100%?
[05:41] <joeforker> 0
[05:41] <joeforker> should I start over, check out gst-all, and 'make' and then try running them from their non-installed positions?
[05:42] <BBB> no...
[05:42] <BBB> 0% is bad
[05:42] <BBB> run with --gst-debug-level=5... where does it hang?
[05:44] <joeforker> DEBUG  spideridentity(12226) gstspideridentity.c(540):gst_spider_identity_handle_src_event: spider_identity src_event
[05:44] <joeforker> INFO         GST_PADS(12226) gstpad.c(3618):gst_pad_event_default_dispatch:<src_0:src> Sending event 0x8077690 to all internally linked pads
[05:44] <joeforker> LOG        GST_BUFFER(12226) gstdata.c(236):gst_data_unref: 0x8077690 1->0
[05:44] <joeforker> INFO        GST_EVENT(12226) gstevent.c(91):_gst_event_free: freeing event 0x8077690
[05:44] <joeforker> LOG   GST_REFCOUNTING(12226) gstobject.c(240):gst_object_unref:<src_0:src> unref 2->1
[05:44] <joeforker> now I'm checking to see a completely empty registry.xml
[05:47] spyder482 ([email protected]) left irc: "Leaving"
[06:25] mathrick|sleep ([email protected]) left irc: Read error: 60 (Operation timed out)
[06:32] sub_pop ([email protected]) joined #gstreamer.
[06:32] benow ([email protected]) joined #gstreamer.
[06:34] <benow> heyas... thomasvz, you around?
[06:34] <BBB> no
[06:34] <BBB> he's asleep
[06:35] <benow> ahh, hmmm perhaps someone else can help me... I'm _sooo_ close to getting the java bindings going.
[06:35] <BBB> heh, not me
[06:35] Action: BBB knows little about java bindings
[06:35] <BBB> sorry
[06:35] <benow> hmm, well, I'll spew, if only to get it in the logs
[06:36] <BBB> sure :)
[06:37] <benow> it's an environment problem, but I'll explain.  When calling gst_pipeline_new via jni I get:
[06:37] <benow> ** (process:15418): CRITICAL **: how to remove plugins?
[06:37] <benow>  (process:15418): GStreamer-CRITICAL **: file gstscheduler.c: line 910 (gst_scheduler_factory_create): assertion `factory->type != 0' failed
[06:37] <benow>  GStreamer-ERROR **: Critical error: could not get scheduler "opt"
[06:37] <benow> Are you sure you have a registry ?
[06:37] <benow> Run gst-register as root if you haven't done so yet.
[06:38] <benow> but when using the _same_ code, called from a c app, it works fine.
[06:39] <benow> i've seen this problem in two places now (the gst-java bindings, and calling a c lib that calls the gst code).  Basically, there's something missing in the JNI runtime environment that is causing the plugins reg to not be found... 
[06:39] <BBB> you need to have a registry
[06:39] <BBB> are you sure you have a registry? see warning ;)
[06:39] <BBB> and if you do... well... then I'm not totally sure
[06:40] <benow> yup, it's there... all installed.  when running as C (ie not wrapped) it works fine.  gst-launch also works well
[06:40] <BBB> does java find it?
[06:40] <BBB> (check the debug logs when running the java bindings)
[06:40] <benow> it finds the libs, there does not seem to be a linkage problem.
[06:41] <BBB> do you have another scheduler?
[06:41] <BBB> try --gst-scheduler=basicgthreads
[06:41] <BBB> or -s
[06:41] <BBB> (basicgthread)
[06:42] <benow> debug enabled with '--gst-debug-level=5' in gst_init args?
[06:43] <benow> well, I have a feeling all is right with the plugins, as gst-launch works fine.  it's the environment which is messed up due to the wrapped environment ... are any environment params being used for execution?
[06:45] <BBB> I really don't know
[06:45] <BBB> and yes to your previous question
[06:45] Action: BBB wants to go to bed
[06:46] <BBB> sorry
[06:46] <BBB> I need to go
[06:46] <BBB> this is a bad excuse for being too late at the university every day
[06:46] <BBB> ;)
[06:46] BBB ([email protected]) left irc: "Client exiting"
[06:56] <benow> hmmm, dies immediately with --gst-scheduler=basicgthreads, and no logging when passing --gst-debug-level=5
[06:58] mathrick ([email protected]) joined #gstreamer.
[07:01] <sri_> good evening mathrick
[07:01] <sri_> you talkative bastard ;)
[07:12] <benow> hmm, well, if anyone has ideas on the above, I can be contacted at: [email protected]
[07:12] benow ([email protected]) left irc: "goodnight"
[07:32] jcsston` ([email protected]) joined #gstreamer.
[07:33] thaytoo ([email protected]) left irc: "leaving"
[07:50] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[08:07] Nick change: jcsston` -> jcsston
[08:18] harshy ([email protected]) joined #gstreamer.
[08:23] escii ([email protected]) left irc: "My damn controlling terminal disappeared!"
[08:27] Company ([email protected]) joined #gstreamer.
[08:29] <Company> good morning angels
[08:33] <mathrick> mornin
[08:34] Action: mathrick wonders why loosing connection caused his nick to be reset to "mathrick" instead of "mathrick|sleep"
[08:36] <Company> ask your irc client
[08:36] <Company> probably because mathrick|sleep was already taken when you reconnected
[08:37] <mathrick> Company: nah, seems it took rather long for it reconnect, so it should've been already timed out by then. Maybe it's indeed Xchat's fault
[08:46] joeforker (dholth@3ffe:bc0:8000:0:0:0:0:262b) left irc: "Client exiting"
[08:46] harshy ([email protected]) left irc: "I quit for now"
[08:50] harshy ([email protected]) joined #gstreamer.
[09:03] <Company> abiword is 25 MB?
[09:04] <Company> that's bigger than mozilla
[09:19] markey ([email protected]) joined #gstreamer.
[09:42] Company ([email protected]) left irc: Remote closed the connection
[09:44] harshy ([email protected]) left irc: Remote closed the connection
[10:03] sub_pop ([email protected]) left irc: "Client exiting"
[10:06] <mathrick> make[6]: Entering directory `/home/mathrick/Dev/gstreamer/gst-ffmpeg/gst-libs/ext/ffmpeg/ffmpeg/libavcodec'
[10:06] <mathrick> gcc -Wall  -g -O3 -DHAVE_AV_CONFIG_H -I.. -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_GNU_SOURCE -c -o i386/fdct_mmx.o i386/fdct_mmx.c
[10:06] <mathrick> i386/fdct_mmx.c:10:20: common.h: No such file or directory
[10:06] <mathrick> i386/fdct_mmx.c:35: error: parse error before "fdct_tg_all_16"
[10:06] <mathrick> ...etc.
[10:06] <mathrick> how do I get common.h?
[10:20] harshy ([email protected]) joined #gstreamer.
[10:20] Mithrandir ([email protected]) joined #gstreamer.
[10:22] <Mithrandir> is gstreamer ready for use in production?
[10:38] <wheels> Mithrandir: Yes, it's been running some of the world's most mission critical business applications since 1806.
[10:38] markey ([email protected]) left irc: "bbl"
[10:38] <Mithrandir> wheels: bah. :P
[10:39] <Mithrandir> wheels: it was actually a serious question -- I might be wanting to use it in a project which will (hopefully) have a lifetime of 5-10 years, with no possibility of running upgrades.
[10:39] <wheels> Mithrandir: Well, I don't doubt that it was a serious question, but it wasn't a very focused one.  Some things are quite mature, some aren't.
[10:40] <Mithrandir> what parts should I stay away from, then?
[10:41] <wheels> This works better the other way -- what are you interested in using?  :-)
[10:43] <Mithrandir> heh. :)
[10:43] <Mithrandir> the product I'm working on is basically an electronic, animated painting.
[10:44] <Mithrandir> so, I'll have some source elements for producing the image, probably some filters to change it around a bit, then an video output.
[10:45] sublett ([email protected]) joined #gstreamer.
[10:46] <wheels> Mithrandir: The main thing to check then will be the specific filter and codec plugins that you're interested in.  The core is reasonably (Ok, maybe not 0.7.x at the moment) stable so a lot of that will be up to the plugins.
[10:46] <Mithrandir> ok, thanks. :)
[10:47] thomasvz ([email protected]) left irc: Client Quit
[10:49] <Mithrandir> gstreamer looks like a fairly nice arch.
[10:53] Shoragan ([email protected]) joined #gstreamer.
[11:03] thaytan ([email protected]) joined #gstreamer.
[11:07] <thaytan> yoyo
[11:13] <mathrick> yo thaytan
[11:16] alley_cat ([email protected]) left irc: Read error: 60 (Operation timed out)
[11:22] markey ([email protected]) joined #gstreamer.
[11:24] pb_ ([email protected]) joined #gstreamer.
[11:26] alley_cat ([email protected]) joined #gstreamer.
[11:34] wheels ([email protected]) left irc: "using sirc version 2.211+KSIRC/1.3.9"
[11:40] jcsston ([email protected]) left irc: Read error: 60 (Operation timed out)
[11:42] robsta ([email protected]) joined #gstreamer.
[11:42] <robsta> hello
[11:43] <robsta> i'm having problems building HEAD on linux/ppc
[11:43] <robsta> using jhbuild
[11:47] <thaytan> what's up?
[11:47] <sxpert_work> morning
[11:47] <thaytan> yo, sxpert
[11:47] <thaytan> sxpert_work: I didn't find the leaks you're seeing
[11:48] <thaytan> but I'm also seeing the sound-server growing on each song change
[11:48] <thaytan> by varying amounts
[11:49] <sxpert_work> thaytan: ok, so, it's not just me ;D
[11:50] <thaytan> I was trying to track them, but valgrind keeps throwing a bus-error on exit before it reports :-/
[11:50] <sxpert_work> thaytan: the fact that it's growing means that some buffers get allocated and never freed, but still pointed to somewhere
[11:51] <sxpert_work> imho, it should 'settle' on memory usage at some point...
[11:51] <thaytan> I agree, I just couldn't find them
[11:51] <sxpert_work> hmmm
[11:52] <sxpert_work> thaytan: ah, you're getting a bus error too ? (how interesting)
[11:54] <robsta> thaytan: gst-plugins bail on multiple symbols in libgsttuner
[11:54] Rotty ([email protected]) joined #gstreamer.
[11:55] smoke ([email protected]) joined #gstreamer.
[11:55] <thaytan> robsta: which symbols?
[11:56] <robsta>  multiple definition of `gst_tuner_marshal_VOID__OBJECT_INT'
[11:56] <robsta> but there are more above
[11:57] <robsta> of type gst_mixer_marshal_X and gst_tuner_marshal_Y
[12:01] <thaytan> hrmn, those are built files
[12:01] <thaytan> made by glib-genmarshal
[12:01] <thaytan> which version of glib do you have?
[12:01] <thaytan> oh, you said - jhbuild
[12:01] <robsta> maybe 2-3 days old
[12:02] <thaytan> does it say which files the  multiple symbols are from?
[12:03] <thaytan> which .o files, I mean
[12:04] <mathrick> make[6]: Entering directory `/home/mathrick/Dev/gstreamer/gst-ffmpeg/gst-libs/ext/ffmpeg/ffmpeg/libavcodec'
[12:04] <mathrick> gcc -Wall  -g -O3 -DHAVE_AV_CONFIG_H -I.. -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_GNU_SOURCE -c -o i386/fdct_mmx.o i386/fdct_mmx.c
[12:04] <mathrick> i386/fdct_mmx.c:10:20: common.h: No such file or directory
[12:04] <mathrick> i386/fdct_mmx.c:35: error: parse error before "fdct_tg_all_16"
[12:04] <mathrick> i386/fdct_mmx.c:35: warning: type defaults to `int' in declaration of `fdct_tg_all_16'
[12:04] <mathrick> ...
[12:04] <mathrick> ghrrr
[12:04] <robsta> thatyan: libgsttuner_la-tunermarshal.o prev def in tunermarshal.c
[12:05] <robsta> i could upload some bigger log somewhere
[12:06] <thaytan> sounds like it is including tunermarshal.o twice, or somesuch??
[12:06] iain ([email protected]) joined #gstreamer.
[12:07] <thaytan> perhaps it's automake or libtool's fault
[12:08] <sxpert_work> thaytan: I know jhbuild fails for me because of libtool fscking up
[12:09] Action: iain really needs to get mono working
[12:10] <thaytan> hi iain
[12:11] markey ([email protected]) left irc: "bbl"
[12:11] Nick change: robsta -> robsta-lunch
[12:11] <mathrick> lo iain
[12:14] <iain> hello
[12:49] thomasvs ([email protected]) joined #gstreamer.
[12:49] jdahlin ([email protected]) joined #gstreamer.
[12:58] Nick change: robsta-lunch -> robsta
[13:00] <Mithrandir> has the bindings page moved to somewhere else?
[13:03] <jdahlin> hello
[13:03] <jdahlin> who wants to review my opt changes? :)
[13:05] <alley_cat> i can't exactly review them as i don't understand schedulers at all, but i could test them, i seem to get hit by all oddities in opt, so may be a good subject :)
[13:06] <jdahlin> I'm only fixed the case where you got a warning
[13:06] <jdahlin> -      g_warning ("opt: unlink elements in same group: implement me");
[13:07] <alley_cat> hmm, didn't get that one recently
[13:07] <thaytan> jdahlin: did you actually implement it? :)
[13:08] <jdahlin> thaytan: yes
[13:08] <thaytan> great! :)
[13:12] <jdahlin> http://async.com.br/~jdahlin/misc/gst-opt-unlink.patch.txt
[13:14] <jdahlin> alley_cat: would be nice if you can test it with some weird pipelines
[13:16] <robsta> thaytan: sorry for the delay, put log to http://stereolyzer.net/gst-plugins.txt
[13:16] <iain> jdahlin: ohhh, I'll try that with marlin
[13:17] <jdahlin> iain: thanks!
[13:22] <thaytan> robsta: oh, weird!
[13:23] <thaytan> glib-genmarshal --version ?
[13:23] <robsta> glib-genmarshal version 2.3.5
[13:24] <thaytan> ds changed stuff in there the other day
[13:24] <thaytan> make sure you're up to date in that dir?
[13:24] <thaytan> s/make/made/
[13:25] <iain> jdahlin: everything seems a lot faster now :)
[13:25] markey ([email protected]) joined #gstreamer.
[13:25] Nick change: markey -> markey|afk
[13:26] <jdahlin> iain: really?
[13:26] <thomasvs> iain: that's just your imagination :)
[13:26] Action: jdahlin jumps up and down
[13:27] <robsta> thaytan: cvs u only affected po stuff
[13:28] <thaytan> robsta: that's strange, because there isn't meant to BE a mixermarshal.c
[13:28] <jdahlin> iain: what kind of pipelines are you using
[13:28] <thaytan> it should be mixer-marshal.c and ditto for the others
[13:28] <iain> thomasvs: it probably is my imagination
[13:29] <thaytan> iain: oh, I snuck in and upgraded your cpu
[13:29] <iain> jdahlin: filesrc ! spider ! audioconvert ! deinterleave ! my-sink
[13:29] <iain> thaytan: sweet, thats why my feet are getting hotter
[13:29] <jdahlin> iain: oggs?
[13:29] <iain> jdahlin: mp3s
[13:29] <iain> trying with ogg now
[13:30] <iain> but its not working
[13:30] <jdahlin> hmm, I got a segfault
[13:30] <iain> I think I might have as well...
[13:30] <iain> trying to cancel the operation hangs, which generally means the gstreamer thread has crashed
[13:31] <jdahlin> okay, I caused it
[13:31] <jdahlin> thanks, I'll look into why
[13:32] <jdahlin> I need a small testcase, hmm
[13:32] <jdahlin> err, small media file
[13:33] <thomasvs> robsta: what's the problem ?
[13:33] <iain> jdahlin: a small ogg do?
[13:33] <robsta> thomasvs: build failure on linux/ppc; see http://stereolyzer.net/gst-plugins.txt
[13:33] <robsta> (jhbuild)
[13:33] <jdahlin> iain: I can only reproduce crash with a large mp3
[13:34] Action: jdahlin runs with logging
[13:34] <iain> jdahlin: hmm, both my small and my normal sized mp3s work
[13:34] <iain> jdahlin: what size is "a large mp3"?
[13:35] <thaytan> gah! damnit, who rolled the CVS revision up to 0.8? :)
[13:35] <thaytan> <- just noticed
[13:35] <thomasvs> thaytan: who do you think ? :)
[13:36] <jdahlin> iain: 4.5M
[13:36] <iain> jdahlin: oh, thats a normal sized one for me
[13:37] <jdahlin> iain: it's a bit too large to be able to use gdb
[13:38] <thaytan> thomasvs: annoyed because I _just_ finished building nautilus-media and friends, and they're against the now obsolete 0.7 install :)
[13:38] <thaytan> well, not annoyed as such, but you know 
[13:38] Action: jdahlin starts gdb
[13:39] sxpert_work ([email protected]) left irc: "Leaving"
[13:40] <thaytan> hey ya, fortunately this laptop is fast :)
[13:40] <mathrick> thomasvs: make[6]: Entering directory `/home/mathrick/Dev/gstreamer/gst-ffmpeg/gst-libs/ext/ffmpeg/ffmpeg/libavcodec'
[13:40] <mathrick> gcc -Wall  -g -O3 -DHAVE_AV_CONFIG_H -I.. -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_GNU_SOURCE -c -o i386/fdct_mmx.o i386/fdct_mmx.c
[13:40] <mathrick> i386/fdct_mmx.c:10:20: common.h: No such file or directory
[13:40] <mathrick> i386/fdct_mmx.c:35: error: parse error before "fdct_tg_all_16"
[13:41] <mathrick> i386/fdct_mmx.c:35: warning: type defaults to `int' in declaration of `fdct_tg_all_16'
[13:41] Action: mathrick hopes someone will finally get interested in his build issues :)
[13:43] <thomasvs> thaytan: heh, sorry :)
[13:43] <thomasvs> mathrick: I'm going import ffmpeg soon
[13:43] <mathrick> thomasvs: hmm, so gst-ffmpeg is unbuildable atm?
[13:44] <iain> jdahlin: well, marlin has just loaded my 70meg mp3, so I guess mp3s work here
[13:44] <thaytan> mathrick: try editing it to ../common.h
[13:44] <thaytan> (just guessing)
[13:45] <mathrick> thaytan: nah, I'm lazy, will just wait for thomasvs :)
[13:45] <jdahlin> iain: I'm trying to track it down
[13:45] <jdahlin> iain: I think it's a bug in deinterleave
[13:46] <jdahlin> yes, it's the same with other schedulers
[13:47] <jdahlin> (process:32261): GStreamer-CRITICAL **: file gstpad.c: line 1393 (gst_pad_try_set_caps): assertion `GST_IS_REAL_PAD (pad)' failed
[13:47] <jdahlin> It has a GParamSpec, when it thinks it's a GstRealPad
[13:48] <iain> strange
[13:48] <iain> I think I have a bug in deinterleave too, but it doesn't crash, it just corrupts one channel
[13:49] <jdahlin> I can give you an mp3 if you like
[13:51] <jdahlin> so, that's not my fault
[13:57] wingo ([email protected]) joined #gstreamer.
[13:59] robsta ([email protected]) left irc: "Client exiting"
[14:03] <Mithrandir> what is the status of the C++ bindings?  according to viewcvs, they haven't been updated in something like 11 months
[14:04] <mathrick> Mithrandir: are you looking at sourceforge repository?
[14:04] <Mithrandir> no, looked fd
[14:04] <Mithrandir> http://freedesktop.org/cgi-bin/viewcvs.cgi/gstreamer/gst-bind/c%2B%2B/examples/, for instance.
[14:04] <mathrick> Mithrandir: mxpxpod has updated them very recently, not sure if it's commited yet
[14:04] <mathrick> mxpxpod: ping?
[14:06] <wingo> hey folks
[14:06] <thaytan> hi wingo!
[14:07] <Mithrandir> mathrick: I am looking in the right place, am I not?
[14:07] <mathrick> yo wingo
[14:07] <mathrick> Mithrandir: I guess so, but I'm not the one who played with them :) mxpxpod is best one to ask
[14:07] pb_ ([email protected]) left irc: "Client exiting"
[14:07] <Mithrandir> idle 12 hours..
[14:08] <mathrick> Mithrandir: I guess that means he'll be awake soon :)
[14:08] <Mithrandir> hopefully, yes.
[14:14] Action: Mithrandir looks around for a release of gst-bind
[14:17] <alley_cat> cool, avidemux can handle avis with multiple audio streams :)
[14:17] <alley_cat> jdahlin: no problems with your opt changes so far
[14:19] <wingo> jdahlin: are you into the scheduler now? /me happy there's someone else :)
[14:19] <jdahlin> alley_cat: cool
[14:19] <jdahlin> wingo: yeah, couldn't resist :)
[14:20] LeRoutier ([email protected]) joined #gstreamer.
[14:20] <LeRoutier> hello
[14:20] <jdahlin> wingo, http://async.com.br/~jdahlin/misc/gst-opt-unlink.diff
[14:20] <jdahlin> err
[14:20] <jdahlin> wingo, http://async.com.br/~jdahlin/misc/gst-opt-unlink.diff.txt
[14:21] <alley_cat> now if someone could fix bug 136686 i'll be very happy ;)
[14:21] Action: wingo checks
[14:21] <alley_cat> hmm, i really need to add clocking to nassink, it's quite useless without
[14:22] <jdahlin> Would be nicer with a better testcase
[14:22] <alley_cat> i can't reproduce it with gst-launch, only gst-player and totem do it
[14:23] <alley_cat> is there a way to use libgstplay from gst-launch or some test program?
[14:23] <wingo> alley_cat: check gstaudioclock in gst-libs/gst/audio
[14:23] <wingo> it's pretty easy actually
[14:23] <wingo> jdahlin: dude i had a patch earlier that i lost thatdid the same thing,
[14:23] <jdahlin> wingo: really?
[14:24] <jdahlin> wingo: I'm not sure if my method to measure the "linkcount" is correct.
[14:24] <wingo> well, sorta -- nowthat i look more it was still for the pad-unlink-with-same-sched case, 
[14:24] <jdahlin> I sent a mail to wim, let us hope he can take some time to review it
[14:25] <wingo> maybe. but he's no necessary.
[14:25] <wingo> the key is, if you've scheduled a link between different groups, you've increased the link count.
[14:25] <wingo> if an unlink causes an element to be removed,
[14:25] <wingo> you need to dec the link count *from that element* manually.
[14:25] <wingo> *not neccessary
[14:26] <wingo> i mean, the better thing is for us to become the experts :)
[14:26] <wingo> although that's a ways away.
[14:26] <wingo> dude i can't believe that. our patches were almost exactly the same.
[14:26] <jdahlin> It took a couple of hours of thinking to get it right
[14:27] <wingo> yeah. not so much code, but losts of thinking.
[14:27] <jdahlin> but it's good sign if they're the same
[14:27] <wingo> *lots
[14:27] <wingo> we're only seeing these problems because
[14:27] sxpert_work ([email protected]) joined #gstreamer.
[14:27] <wingo> before the 1.61 version, groups were being over-reffed.
[14:27] <wingo> they'dnever be destroyed.
[14:28] <wingo> seems my space bar doesn't want to work :/
[14:28] Action: jdahlin checks out 1.61
[14:29] <jdahlin> http://freedesktop.org/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/gst/schedulers/gstoptimalscheduler.c.diff?r1=1.60&r2=1.61
[14:29] <wingo> the 1.61 patch fixed a lot of problems, but potentially brought old ones to the surface.
[14:30] <jdahlin> we need a *good* testsuite
[14:31] Nick change: markey|afk -> markey
[14:31] <wingo> yes, i agree
[14:31] <wingo> it's difficult to make one, though.
[14:32] <wingo> i think the set of different topologies is infinite.
[14:32] <jdahlin> somewhat true
[14:32] <jdahlin> but we can test the general ones, used by common applications
[14:33] <jdahlin> and some corner cases
[14:34] <wingo> true. how would you test success? empty bufpens? a handoff signal?
[14:34] <wingo> i want to start profiling soon.
[14:34] <sxpert_work> wingo: hi there
[14:34] <wingo> howdy sxpert
[14:35] <wingo> and sxpert_work
[14:35] <jdahlin> You can try to send in some data where you know what the output data is supposed to be
[14:35] <jdahlin> similar to what I've been doing in gst-media-test
[14:35] <sxpert_work> wingo: heh
[14:36] <jdahlin> wingo, http://build.fluendo.com:8080/FC1-media/events/9/log
[14:37] Action: wingo wonders if he fucked things up somehow
[14:37] <wingo> ah, ok ;)
[14:37] <jdahlin> hmm, a little bit broken (as usual)
[14:37] <wingo> jdahlin: are you working for fluendo?
[14:37] <jdahlin> wingo: yes
[14:39] <wingo> cool.
[14:41] <jdahlin> it's nice to be able to spend work time on fixing gstreamer
[14:42] <LeRoutier> bbl
[14:42] LeRoutier ([email protected]) left irc: "Leaving"
[14:43] kmaraas ([email protected]) joined #gstreamer.
[14:45] <wingo> jdahlin: i bet.
[14:46] <jdahlin> wingo: especially on infrastrucutre
[14:46] <jdahlin> that's really needed
[14:48] <sxpert_work> wingo: I have some worries about memory usage... I have a program (available at http://sxpert.esitcom.org/sound-server ) that plays your entire set of ogg files randomly, and have noticed that the process grows from about 4M to 30 in 24 hours.. and gets killed by the OOM killer after a few days...
[14:55] <jdahlin> wingo: do you think it would be safe to include my patch before 0.8?
[14:55] Action: jdahlin hides from thomasvs
[14:56] <wingo> sxpert: i have similar problems for dynamic synthesizer patches. this stuff will get ironed out over the next few weeks, i think.
[14:57] Action: thomasvs pulls out hair and plans a 0.6.7 release this week
[14:57] <alley_cat> jdahlin: with your opt patch: ** ERROR **: internal error: cannot schedule get to loop in multi-loop based group
[14:57] <alley_cat> without the patch gst-player just segfaults without outputting that error :P
[14:57] <jdahlin> oh :)
[14:57] <alley_cat> only sometimes though, most problems i have appear and disappear randomly
[14:58] <sxpert_work> wingo: ah, ok, so it's not "just me" :D
[14:58] <wingo> jdahlin: i would say yes -- i mean, it basically only affects that section that had a g_warning
[14:58] <jdahlin> wingo: yes, and that was pretty broken before anyway
[14:58] <wingo> indeed
[14:59] Action: sxpert_work drops coins in 'put back hair on thomasvs' head once 0.8 is out' cash box
[14:59] Action: thomasvs should collect a list of the instances of when "it's just a small part of the code" broke something very important in the final release
[14:59] <sxpert_work> thomasvs: lol
[14:59] Action: thomasvs remembers doing a release that had aasink as the default video sink
[14:59] Action: thomasvs remembers a plugin that fucked up typefinding so the player could not play anything
[15:00] Action: thomasvs stops remembering and finds something to eat
[15:00] <jdahlin> thomasvs: we have lots of paper bags here in the officfe
[15:00] <wingo> that aasink release was funny :)
[15:00] <thaytan> thomasvs: "In the first case, it is an official release of GStreamer." ... "if n == 0"
[15:00] <thaytan> missed the last bit
[15:00] <thomasvs> jdahlin: you think those will cover the wounds I'm about to inflict on you ?
[15:00] <jdahlin> thomasvs: no, I have insurance
[15:00] <thomasvs> thaytan: you *read* that shit ? cool !
[15:00] <thomasvs> you're going to bandage your wounds using your insurance papers ?
[15:00] <thomasvs> good luck :)
[15:01] <thomasvs> I'd get another insurance against self-inflicted stupidity :)
[15:01] <jdahlin> hihi
[15:01] Action: thomasvs actually heard that giggle
[15:01] <alley_cat> #0  0xffffe410 in ?? () <= i really love backtraces like this one
[15:02] <thomasvs> thaytan: what did I miss ? I did not want to say "n == 0" because there are no source tarballs out there with 0.x.y.0
[15:02] <thaytan> thomasvs: it doesn't actually say what the 'first case' is ...
[15:03] <thaytan> perhaps it's the first case 'if n doesn't exist'?
[15:03] <sxpert_work> ===9392=== collecting stats on lost blocks
[15:03] <sxpert_work> Bus error
[15:03] <sxpert_work> ^ I love *those* valgrind messages
[15:03] <thaytan> thomasvs: oh, or I could read it better
[15:03] <thaytan> I missed it in the earlier paragraph :)
[15:04] <thaytan> it's good :)
[15:05] khalek ([email protected]) joined #gstreamer.
[15:07] <wingo> hm
[15:07] <wingo> i gotta head out
[15:08] <thaytan> cya wingo
[15:08] <thaytan> I'm trying to cut gst-editor 0.7.0 before I sleep
[15:08] <wingo> good luck ;)
[15:10] Action: sxpert_work passes the knife to thaytan
[15:11] <iain> sxpert_work: Bus error is more like "Segmentation fault" in that its not a Valgrind error surely?
[15:11] <thomasvs> thaytan: btw, was user_data on mpeg2dec something you used for dvd ?
[15:11] sublett ([email protected]) left irc: Read error: 113 (No route to host)
[15:11] <thaytan> thomasvs: not yet, no
[15:12] <thaytan> DVD subtitles come from the system stream, not the user_data in the video stream
[15:12] <thaytan> although we'll need it back later when we go to do closed captioning
[15:12] <sxpert_work> iain: that's the last message I get from running valgrind.. next is a command promp
[15:12] <sxpert_work> t
[15:12] <iain> yeah
[15:12] <iain> it means balgrind crashed
[15:12] <iain> valgrind
[15:12] <thaytan> badgrind
[15:12] <thomasvs> as opposed to
[15:12] <jdahlin> ballgrind
[15:12] <thomasvs> balgrind, the sex toy ?
[15:13] <sxpert_work> iain: yeah, what I thought, very useful though :D
[15:13] <thomasvs> I guess it shows what environment the two of us here work in :)
[15:13] <sxpert_work> thomasvs: lol
[15:13] <iain> thomasvs: you're working on a media streaming framework...
[15:14] <iain> thomasvs: you know fine well the only real application for that is streaming porn
[15:14] <sxpert_work> iain: heh
[15:14] <sxpert_work> iain: well, it's a car stereo in my case :D
[15:17] <mathrick> sxpert_work: porn i car stereo? you're sick
[15:17] <mathrick> s/i/in/
[15:17] <sxpert_work> mathrick: gah...
[15:18] <mxpxpod> Mithrandir: around?
[15:18] <sxpert_work> mathrick: "lady on the passenger seat giving head to driver" is a classic :D
[15:18] <Mithrandir> mxpxpod: yes
[15:19] <mxpxpod> Mithrandir: sorry, I just woke up
[15:19] <Mithrandir> mxpxpod: np. :)
[15:19] <Mithrandir> mxpxpod: I'm looking at the C++ bindings, it seems like viewcvs on fd.o is somewhat fucked wrt dates. :)
[15:20] <Mithrandir> mxpxpod: but the bindings don't compile.
[15:20] <mxpxpod> Mithrandir: there's an open bug for gstmm with a patch to update it for 0.7.4
[15:20] <Mithrandir> bugzilla?
[15:20] <mxpxpod> yeah
[15:21] Action: Mithrandir goes to find the bu
[15:21] <Mithrandir> +g
[15:23] <Mithrandir> mxpxpod: bugzilla is dog slow, you don't happen to have the # handy?
[15:23] <mxpxpod> lemme get it
[15:25] Action: Mithrandir found it
[15:26] <mxpxpod> oh, ok
[15:26] <mxpxpod> good
[15:26] <Mithrandir> 135308, is it.
[15:31] wingo ([email protected]) left irc: Read error: 110 (Connection timed out)
[15:34] Action: Mithrandir compiles
[15:34] <mxpxpod> Mithrandir: that should compile... if not, email me
[15:34] <mxpxpod> I gotta go
[15:34] <Mithrandir> structure.cc: In member function `bool Gst::Structure::get(const 
[15:34] <Mithrandir>    Glib::ustring&, bool&) const':
[15:34] <Mithrandir> structure.cc:161: error: non-lvalue in unary `&'
[15:34] <mxpxpod> dang
[15:35] <mxpxpod> ok, I'll have to take a look at it today
[15:35] <Mithrandir> I'll give it a shot
[15:35] <mxpxpod> alright
[15:35] <mxpxpod> talk to ya later
[15:35] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[15:37] Nick change: khalek -> kh_zZz
[15:46] KoRnouille ([email protected]) left irc: Read error: 110 (Connection timed out)
[15:53] alley_cat ([email protected]) left irc: "May the Source be with you!"
[15:53] smoke ([email protected]) got netsplit.
[15:55] LeRoutier ([email protected]) joined #gstreamer.
[15:56] <LeRoutier> re
[15:56] alley_cat ([email protected]) joined #gstreamer.
[15:59] smoke ([email protected]) got lost in the net-split.
[16:06] <LeRoutier> thomasvs, are you around ?
[16:06] <thomasvs> yeah
[16:07] <LeRoutier> did you have time to review my updated patch (with fallbacks) ?
[16:07] <thomasvs> no, bug link ?
[16:08] <LeRoutier> http://bugzilla.gnome.org/show_bug.cgi?id=131270
[16:09] <LeRoutier> perhaps some fallback aren't really needed for packages which provide a .pc file for a long time but in doubt, i did one for every check i changed in the past
[16:15] <alley_cat> gst-launch-0.7: error while loading shared libraries: libgstreamer-0.7.so.6: cannot enable executable stack as shared object requires: Invalid argument
[16:15] <alley_cat> anyone ever got such a message from valgrind?
[16:15] <LeRoutier> thomasvs: and this file seems unused : gst-plugins/m4/libdv.m4  (as libdv detection use exclusively pkg-config)
[16:15] psyduck ([email protected]) joined #gstreamer.
[16:16] <thaytan> thomasvs: do you manually make the .bz2 archive from the .gz, or is there a magical way?
[16:17] <thomasvs> thaytan: I have "release:" added as a target, just copy/paste that stuff
[16:17] <thomasvs> you're doing a real release ?
[16:17] <thomasvs> rock :)
[16:18] <thomasvs> thaytan: is it working with 0.8 as well for the pkgconfig check ?
[16:18] <thaytan> yep
[16:18] <thaytan> my first tarball fails on configure though
[16:18] <thaytan> ./config.status: line 1233: ./intltool-extract.in: No such file or directory
[16:18] <thomasvs> you are running make distcheck, right ?
[16:18] <thaytan> mv: cannot stat `intltool-extract.out': No such file or directory
[16:18] <thomasvs> you should add it to EXTRA_DIST iirc
[16:18] <thomasvs> the intltool stuff might be oldish
[16:18] <thaytan> any examples?
[16:19] <thomasvs> any gnome tarball :)
[16:20] <thaytan> ta. totem.
[16:23] psyduck ([email protected]) left irc: Remote closed the connection
[16:26] <thaytan> ok, 2 more questions.
[16:26] <thaytan> where do I put the tarballs in the www checkout tree I have
[16:26] psyduck ([email protected]) joined #gstreamer.
[16:26] <thaytan> and how do I sync them to the real website?
[16:29] Action: bitshifter finds that in a   filesrc ! typefind ! spider ! fakesink  pipeline the MAD element doesn't seem to be finalized/disposed of when the pipeline is unrefed
[16:30] <thomasvs> thaytan: simple
[16:31] <thomasvs> thaytan: data/src contains tarballs
[16:31] <thomasvs> put it in the right location
[16:31] <thomasvs> then, bin/data-put
[16:31] <thaytan> I don't need a full checkout of the data first?
[16:31] mxpxpod ([email protected]) joined #gstreamer.
[16:34] <mxpxpod> Mithrandir: any luck?
[16:34] <thomasvs> thaytan: ah, hm
[16:34] <thomasvs> good question
[16:34] <Mithrandir> seems like g++ doesn't like the static_casts; I'm going through and fixing it now.
[16:34] <thomasvs> I thought you had already
[16:34] markey ([email protected]) left irc: "bbl"
[16:34] <Mithrandir> not sure if I do this in the right way, though.. I don't really know C++ :)
[16:34] <thaytan> no, I stopped when you said it was massive 
[16:34] <mxpxpod> Mithrandir: which version of g++?
[16:34] <thomasvs> thaytan: ok, don't do data-put :)
[16:35] Action: bitshifter hands a bag of amphetamines to bugzilla
[16:35] <thaytan> I'll just rsync that dir across
[16:35] <Mithrandir> mxpxpod: g++ (GCC) 3.3.3 (Debian 20040306)
[16:35] <thomasvs> thaytan: let me fix some stuff
[16:35] <mxpxpod> Mithrandir: ah, stupid 3.3
[16:35] <mxpxpod> :)
[16:35] <thaytan> thomasvs: ok
[16:35] <Mithrandir> mxpxpod: oh?  fixed in 3.4?
[16:35] <mxpxpod> Mithrandir: no, I used 3.2
[16:35] <thaytan> is it meant to have updated the site when I did cvs commit, btw?
[16:35] <mxpxpod> Mithrandir: which cast is it choking at?
[16:35] <thomasvs> thaytan: depends on what you commited
[16:36] <thomasvs> thaytan: not all the make rules are correct yet
[16:36] <Mithrandir>   return gst_tag_list_get_boolean_index(const_cast<GstTagList*>(gobj()), tag.c_str(), index, &(static_cast<int>(*value)));
[16:36] <Mithrandir> the static_casts
[16:36] <Mithrandir> ag.cc:154: error: non-lvalue in unary `&'
[16:36] <Mithrandir> typically
[16:36] <thaytan> thomasvs: it seemed to try and do a cvs up then complain about a missing .cvspass file
[16:36] <mxpxpod> whoa, that static cast is wrong
[16:36] <thaytan> which seems like an update attempt gone awry
[16:36] <Mithrandir> I might have fucked around with it
[16:37] <mxpxpod> Mithrandir: change the static_cast<int> to (int*)
[16:37] <Mithrandir> willdo
[16:37] <mxpxpod> just on that one
[16:38] smoke ([email protected]) joined #gstreamer.
[16:44] <mathrick> RB fails to seek for me, checkout from today's morning
[16:44] <mathrick> thomasvs: btw, is gst-ffmpeg buildable yet?
[16:45] <thaytan> mathrick: it was never broken for me...
[16:45] <thaytan> which automake are you using?
[16:45] <mathrick> thaytan: that missing common.h thing it was
[16:45] <mathrick> thaytan: gimme a sec
[16:46] <thaytan> mathrick: mine has -I$srcdir/libaccodec or something on the compile line, so finds common.h happily
[16:46] <mathrick> 1.7.9
[16:47] <LeRoutier> mathrick, the problem is different
[16:47] <mathrick> LeRoutier: oh?
[16:47] <LeRoutier> mathrick, it still uses the old configure system instead of ronald's new one
[16:48] <LeRoutier> because autogen fails i think (look for autogen complaining on non defined AC_CHECK_IMLIB, or AM_PATH_SDL)
[16:48] <mathrick> LeRoutier: checkout to fresh dir then?
[16:48] <thomasvs> thaytan: did you add gst-editor to the www module ?
[16:48] <thomasvs> because it's not here
[16:48] <LeRoutier> mathrick: won't change anything, there are missing m4 macros
[16:49] <LeRoutier> mathrick: http://bugzilla.gnome.org/show_bug.cgi?id=136366
[16:49] <thaytan> thomasvs: yes, but it seems to fail on the commit hook, I think it's trying to update the live site?
[16:49] <mathrick> well, I don't see any macros complaints
[16:49] <thomasvs> thaytan: yeah, it is
[16:49] <thaytan> the commit succeeds, but the update doesn't
[16:49] <thomasvs> lemme check
[16:49] <thaytan> AFAICS I've put everything in as should be
[16:50] <thomasvs> as for the tarball ... if you updated cvs, feel free to run data-put now
[16:50] <thomasvs> forced a regen of the site, that's ok now
[16:51] <thomasvs> thaytan: 0.7.0 is meant to compile against which gst core ?
[16:51] <thaytan> 0.7 or 0.8
[16:56] <thaytan> thomasvs: the release notes 0.7.0.html file didn't make it to the website
[16:57] <thaytan> did I miss something?
[16:57] <thomasvs> thaytan: did you add the .xml file ?
[16:57] <thomasvs> me /checks locally
[16:57] <thaytan> yep
[16:58] <thomasvs> looks to me like you didn't add 0.7.0.xml
[16:58] <thaytan> woops!
[16:58] <thaytan> I added 0.7.1.xml when I got confused earlier :-/
[17:00] <thaytan> ok, committed
[17:00] jcsston ([email protected]) joined #gstreamer.
[17:00] <thaytan> still a perms problem on the commit hook
[17:03] <thaytan> cvs [update aborted]: cannot make directory src/copy/data/src: Permission denied
[17:03] <thomasvs> can you paste it to me privately ?
[17:03] Action: thomasvs checks
[17:04] mxpxpod ([email protected]) left irc: "Leaving"
[17:04] <thomasvs> hm, should set the umask on update
[17:06] <thaytan> woohoo! I've done my first release! :)
[17:09] <thomasvs> thaytan: now make the release notes say 0.7.0 :)
[17:09] <thaytan> doh
[17:09] <thomasvs> did you like the setup for writing release notes ?
[17:09] <thaytan> totally, it is awesome, dude!
[17:10] <thaytan> you've made the whole thing nicely streamlined
[17:10] <thomasvs> good, nice to see it pay off
[17:11] <thaytan> Another beer to buy you in 5 weeks
[17:11] <thaytan> can I trigger a rebuild of the website, or do I need you to?
[17:12] <thomasvs> you can
[17:12] <thomasvs> bin/www-update
[17:12] <thomasvs> that's for forcing it when the makefile rules were borked
[17:12] <thomasvs> thaytan: don't forget to add an item to news, at the bottom
[17:12] <thaytan> done already
[17:12] <thomasvs> thaytan: if you figure out an easy way to reverse order of items in newsl.xml, that'd be nice too :)
[17:13] <thomasvs> thaytan: but if stuff doesn't autoupdate after a commit, let me know what failed in a mail so I can colelct and look at it later
[17:13] <thaytan> only that perms problem. do you want a mail for that?
[17:14] <thomasvs> nope, should be fixed now
[17:14] <thomasvs> just need to chmod a little now
[17:14] <thomasvs> thaytan: can you run this too:
[17:14] <thomasvs>  chmod g+w -R *
[17:14] <thomasvs> on fdo, on our website ?
[17:14] Action: thomasvs thinks about a better fix
[17:15] <thaytan> stickyness on the dirs
[17:15] <thomasvs> well, they were sticky already
[17:15] <thomasvs> hm, groupsticky
[17:15] <thaytan> yeah
[17:15] <thomasvs> well, they are
[17:15] Action: thomasvs is confused
[17:15] <thomasvs> yet your tarballs are owned by you
[17:16] <thaytan> they should be owned by me
[17:16] <thomasvs> well, that's fine
[17:16] <thomasvs> but not groupwritable
[17:16] <thomasvs> so, what's wrong then ?
[17:16] <thomasvs> I guess you have to make them groupwritable ?
[17:16] <thaytan> yep
[17:16] <thaytan> the umask should help
[17:16] <thaytan> oh, no
[17:17] <thaytan> it won't because rsync is in archive mode
[17:17] <thomasvs> I'll add chmod -R g+w in data-put
[17:17] <thaytan> yeah, I think that's the best
[17:17] <thaytan> might want to redirect that to dev/null too
[17:18] <thomasvs> ok, commited, try running it
[17:21] <thaytan> looks like it worked great 
[17:23] <thaytan> ok, 3:20am. 4 hours to sleep. G'night!
[17:24] <mathrick> nite thaytan
[17:25] <LeRoutier> good night thaytan
[17:25] markey ([email protected]) joined #gstreamer.
[17:30] <thomasvs> night jan, thanks for the release
[17:36] mathrick ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[17:38] teuf ([email protected]) joined #gstreamer.
[17:38] mathrick ([email protected]) joined #gstreamer.
[17:39] mxpxpod ([email protected]) joined #gstreamer.
[17:41] kwm ([email protected]) joined #gstreamer.
[17:49] <LeRoutier> lut teuf. hi mxpxpod, kwm
[17:49] <mxpxpod> ?
[17:50] <mxpxpod> hi
[17:52] <kwm> hi
[17:59] walters ([email protected]) joined #gstreamer.
[17:59] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[18:03] <jdahlin> http://build.fluendo.com:8080/commits/info
[18:04] <thomasvs> jdahlin: now make the files clickable :)
[18:05] <jdahlin> and reverse
[18:08] <LeRoutier> jdahlin, visually, i prefer this one http://cia.navi.cx/stats/project/GStreamer/ (but i suppose your one is more functionnal)
[18:09] <jdahlin> LeRoutier: I prefer bonsai, but it's not installed at fd.o
[18:15] trow ([email protected]) joined #gstreamer.
[18:16] psyduck ([email protected]) left irc: "Leaving"
[18:22] <thomasvs> LeRoutier: applied and closed, thanks
[18:30] teuf ([email protected]) left irc: "Client exiting"
[18:30] sub_pop ([email protected]) joined #gstreamer.
[18:30] psyduck ([email protected]) joined #gstreamer.
[18:32] smoke ([email protected]) got netsplit.
[18:32] smoke ([email protected]) returned to #gstreamer.
[18:35] <LeRoutier> thomasvs, ok, thx
[18:37] <sxpert_work> jdahlin: Bonzaï (records) ?
[18:38] <jdahlin> sxpert_work: ?
[18:40] <sxpert_work> jdahlin: http://www.bonzai-records.com/
[18:41] <jdahlin> what does that have to do with anything?
[18:43] <sxpert_work> jdahlin: well, you mentionned bonsai :D
[18:43] <jdahlin> sxpert_work, http://bonsai.mozilla.org/cvsqueryform.cgi
[18:45] apoc ([email protected]) joined #gstreamer.
[18:45] <sxpert_work> jdahlin: ah heh :D
[18:45] <apoc> hi
[18:46] <LeRoutier> lut apoc
[18:47] walters ([email protected]) left irc: Client Quit
[18:48] <mathrick> hi apoc
[18:55] kmaraas ([email protected]) left irc: Client Quit
[18:55] <apoc> can someone try to play a mod file ?
[18:55] <apoc> with modplug
[18:57] <thomasvs> apoc: it seems to give core rrors
[18:58] <apoc> thomasvs: about pad negotiated ?
[19:00] <thomasvs> apoc: yeah
[19:01] <apoc> thomasvs: you propably don't have last version of modplug 
[19:01] <apoc> thomasvs: Company fixed this yesterday
[19:02] BBB ([email protected]) joined #gstreamer.
[19:02] <apoc> yo BBB
[19:03] <thomasvs> BBB: ah, just wanted to discuss something with you
[19:04] <thomasvs> about gst-ffmpeg
[19:04] <thomasvs> so, here's the thing (wrt your mail about it last time)
[19:04] <thomasvs> if our CVS contains ffmpeg + patches applied, which then gets checked out as an additional module by other people put into gst-ffmpeg
[19:04] <thomasvs> then that works fine for users
[19:04] <thomasvs> but
[19:04] <thomasvs> that means, that for you ...
[19:04] <thomasvs> ... you have to
[19:04] <thomasvs> a) hack on ffmpeg in the mirror/ module
[19:04] <thomasvs> b) commit it
[19:04] <thomasvs> c) update gst-ffmpeg to get the fixes
[19:04] <thomasvs> d) test
[19:05] <thomasvs> are you ok with that ?
[19:05] <LeRoutier> hello ronald
[19:08] <mxpxpod> is there a rhythmbox that works with gstreamer-0.7.5?
[19:08] <psyduck> hello everybody
[19:09] <psyduck> when i use the "adder" element with gstreamer 0.6.4, the first channel stops after 15 seconds
[19:09] <psyduck> and when i use it with gstreamer 0.7.6
[19:10] <psyduck> it doesn't output any sounds
[19:10] <psyduck> I also tried examples/mixer/mixer.c
[19:10] <psyduck> from gstreamer 0.7.6 and doesn't work neither
[19:10] <psyduck> am I missing something?
[19:11] <thomasvs> the example could be outdated
[19:11] <thomasvs> I hope to check it again next week
[19:12] <thomasvs> as for "channel stops after 15 seconds"; how are you using it with 0.6.4 ?
[19:12] <psyduck> i have filesrc1 -> decoder1 -> adder
[19:12] <psyduck> filesrc2 -> decoder2 -> adder
[19:12] <psyduck> and adder -> audiosink
[19:13] <psyduck> i also tried putting the decoding in threads
[19:13] <thomasvs> psyduck: yeah, but I mean - is this an app you wrote yourself, or one of ours ?
[19:13] <psyduck> wrote myself
[19:13] <psyduck> but with the mixer example, it did the same
[19:13] <psyduck> the mixer example from 0.7.5 with the libs from 0.6.4 to be precise :)
[19:15] <apoc> thomasvs: i'm trying to update gst-plugins and I get this error 
[19:15] <apoc> thomasvs: cvs [update aborted]: connect to mplayerhq.hu(192.190.173.45):2401 failed: Connection timed out
[19:15] <LeRoutier> apoc: gst-plugins or gst-ffmpeg ?
[19:16] <BBB> thomasvs: I don't like it particularly, it's too complicated... I know you proposed it before, but there must be an easier way...
[19:16] <apoc> LeRoutier: gst-plugins .. that's the problem ;)
[19:16] <BBB> thomasvs: it'd be cool if you could set it up in such a way that I can put a symlink to my "freedesktop ffmpeg module" in gst-libs/ext/ffmpeg/ffmpeg, so that I don't actually have to checkout twice
[19:17] <BBB> thomasvs: then it becomes a whole lot easier for me
[19:17] <BBB> I use the quilt version of the hack module
[19:17] <BBB> you guys use the production module
[19:17] <BBB> which is the hack module with all patches applied
[19:17] <BBB> so it'd be cool if it was all combined
[19:17] <thomasvs> BBB: heh :) the problem I just described comes from YOUR suggestion :)
[19:18] <thomasvs> BBB: which is why I'm verifying if it's really what you want
[19:18] <apoc> hm ok I found the problem ...
[19:18] <BBB> yup, so it's not
[19:18] <BBB> I was just thinking by then
[19:18] <BBB> and it's not a good solutio, I fear
[19:18] <BBB> thomasvs: don't worry about it now
[19:18] <BBB> the current way is good enough for 0.8.0
[19:18] <BBB> please fix osssink
[19:19] <BBB> I *want* osssink/ossmixer and alsasink/alsamixer to work in 0.8.0
[19:19] <BBB> with gnome-volume-control and totem/RB
[19:19] <BBB> without that, there's no point in releasing a stable version
[19:19] <BBB> I mean, 0.6.x had a working osssink!
[19:20] <thomasvs> BBB: what's not working ?
[19:21] <BBB> after a seek, it often fails
[19:21] <BBB> wait
[19:21] <BBB> didn't you read my email?
[19:21] Action: BBB slaps thomas
[19:21] <BBB> I have some ideas for tonight
[19:21] <BBB> point is that it's failing badly
[19:22] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[19:23] <thomasvs> yeah, I read it
[19:23] <thomasvs> quickly
[19:23] <thomasvs> was hoping to get back to it
[19:23] <thomasvs> basically I see the same lag
[19:23] <BBB> I probaly know how to fix it
[19:23] <BBB> partly
[19:24] <BBB> but I still need documentation
[19:24] <BBB> basically, the "drop sound" part in osssink needs to die
[19:24] <BBB> that's what's killing it
[19:24] <BBB> it should just be removed completely
[19:35] teuf ([email protected]) joined #gstreamer.
[19:35] bilboed ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[19:36] <BBB> if I'm correct, osssink will play movies again and not output "noise"
[19:38] pb_ (~pb@2002:5160:45ef:0:240:95ff:fe30:aa84) joined #gstreamer.
[19:38] Action: BBB lunch
[19:38] <BBB> bbl
[19:38] BBB ([email protected]) left #gstreamer.
[19:41] <alley_cat> gst-ffmpeg still looks for gstreamer-0.7 packages
[19:43] <thomasvs> alley_cat: yeah, needs updating
[19:45] bilboed ([email protected]) joined #gstreamer.
[19:46] <LeRoutier> alley_cat, same for gst-player
[19:51] <ds-work> do we care if it doesn't work with 0.7 anymore
[19:56] <thomasvs> ds-work: do we care if what doesn't what ?
[19:58] <LeRoutier> bbl
[19:58] LeRoutier ([email protected]) left irc: "Leaving"
[20:00] <ds-work> thomasvs: if gst-ffmpeg looks for just gstreamer-0.8
[20:00] <ds-work> and not gstreamer-0.7
[20:00] psyduck ([email protected]) left irc: Remote closed the connection
[20:02] psyduck ([email protected]) joined #gstreamer.
[20:21] thomasvs ([email protected]) left irc: Read error: 113 (No route to host)
[20:23] bitshifter ([email protected]) left irc: "bye"
[20:23] mxpxpod ([email protected]) joined #gstreamer.
[20:24] thies ([email protected]) joined #gstreamer.
[20:29] steveb_ ([email protected]) joined #gstreamer.
[20:31] BBB ([email protected]) joined #gstreamer.
[20:35] kmaraas ([email protected]) joined #gstreamer.
[20:36] bitshifter ([email protected]) joined #gstreamer.
[20:39] steveb__ ([email protected]) left irc: Read error: 60 (Operation timed out)
[20:40] <BBB> where is dolphy?
[20:40] <Mithrandir> mxpxpod: around?
[20:41] <mxpxpod> Mithrandir: yeah
[20:41] <Mithrandir> /usr/lib/gtkmm-2.0/proc/gtkmmproc -I ../../tools/m4 --defs . propsentry . ./../gstmm
[20:41] <Mithrandir> No conversion from GstPropsType to PropsType defined (line: , parameter name: gst_props_entry_get_props_type`'(ifelse(`1',1,const_cast<__CNAME__*>(gobj()),gobj())`'ifelse(`',,,`, ')))
[20:41] <Mithrandir> m4 failed with exit code 1.  Aborting...
[20:41] <jdahlin> c++ is just so beautiful
[20:41] <jdahlin> NOT
[20:41] <Mithrandir> after a cvs up
[20:41] <mxpxpod> Mithrandir: aw suck
[20:41] <mxpxpod> Mithrandir: what version of gstreamer are you trying to compile against?
[20:41] <ds-work> dolphy has been missing for a while
[20:42] <Mithrandir> mxpxpod: CVS.
[20:42] <Mithrandir> (cvs as of a few hours ago)
[20:42] <mxpxpod> Mithrandir: ah, ok... I'll have to update the defs and such (I think)... I'll take a look at it tonite
[20:42] <Mithrandir> ok, thx.
[20:42] <Mithrandir> :)
[20:42] <Mithrandir> is there a stable tarball I can download somewhere?
[20:43] <mxpxpod> of gstmm?
[20:43] <Mithrandir> yup
[20:43] <mxpxpod> no
[20:43] <mxpxpod> I'm not even the maintainer... I just submitted a patch
[20:43] <Mithrandir> oh, ok.
[20:44] <Mithrandir> do you know if there's a stable version planned for sometime?
[20:44] <mxpxpod> nope
[20:44] <mxpxpod> you'd have to ask on the mailing list
[20:44] <Mithrandir> ok
[20:44] <Mithrandir> thx anyhow. :)
[20:53] jdahlin ([email protected]) left irc: "Leaving"
[20:56] Shoragan ([email protected]) left irc: "Leaving"
[20:57] lupusBE ([email protected]) joined #gstreamer.
[21:00] pb_ (~pb@2002:5160:45ef:0:240:95ff:fe30:aa84) left irc: "Client exiting"
[21:12] <BBB> ds-work: but where *is* he? ;)
[21:13] <ds-work> BBB: ask thomasvs or jdahlin :)
[21:14] <ds-work> ugh.  /me is impatient for 0.8.0, so I can close all the 0.6 bugs :)
[21:15] <BBB> like 0.8.0 is bugfree
[21:15] <BBB> ds-work: could you help me with osssink?
[21:15] <BBB> you have totem, right?
[21:15] <BBB> can you test something for me?
[21:16] <ds-work> no totem here
[21:16] <ds-work> I can compile it
[21:16] <BBB> please do :)
[21:16] <ds-work> as long as it doesn't have a million deps
[21:16] thomasvs ([email protected]) joined #gstreamer.
[21:16] sublett ([email protected]) joined #gstreamer.
[21:16] <BBB> play an AVI, heat up your CPU while playing an AVI (so you've got 100% CPU)... then, wait and see
[21:16] <BBB> it depends on gnome-2.5, basically
[21:17] <BBB> the test is: remove the throw_out code in osssink and see if it still does the same bad thing that it did before (like not playing any audio at all anymore)
[21:17] <BBB> once AVI plays perfectly fine, I'll start fixing MPEG
[21:18] <BBB> I hope you'll implement seeking qtdemux while we're at it
[21:18] <BBB> then I'll fix matroskademux too... it segfaults on seeking
[21:18] <BBB> just before 0.8.0 :)
[21:18] <ds-work> I added some code
[21:18] <ds-work> it doesn't seem as pressing as some other bugs
[21:18] <BBB> placeholders? :)
[21:18] <BBB> or actual seeking code?
[21:19] <BBB> I know, but we need some basica video formats working
[21:19] <BBB> else, nobody *at all* will use totem-gst
[21:19] <BBB> why would they?
[21:19] <BBB> it sucks
[21:19] <ds-work> placeholders
[21:19] <BBB> it doesn't play back any video format right now
[21:19] <BBB> I'd like to make some basic ones work
[21:19] <ds-work> it's definitely on the list of things to do
[21:21] <ds-work> uhg, I'm tired of building gnome from CVS
[21:21] <mxpxpod> ds-work: I hear ya there
[21:22] <ds-work> I want snapshot .debs
[21:23] <harshy> dont we all
[21:23] <mxpxpod> hey harshy 
[21:23] <harshy> hey mxpxpod
[21:24] <mxpxpod> harshy: why don't you set up an apt repository with snapshot debs? ;)
[21:24] <mxpxpod> harshy: it's not like you do anything ever
[21:24] <harshy> haha
[21:24] <mxpxpod> harshy: have you checked out my page recently?
[21:24] <harshy> ya it grew
[21:25] <mxpxpod> :)
[21:25] <mxpxpod> all pages should be skinned now
[21:25] <harshy> nice
[21:26] <mxpxpod> harshy: and I'll have to get you a new copy of my gpg pub key with my new email address in it
[21:26] <harshy> mxpxpod: just add the new email to the current one
[21:26] <mxpxpod> I know, that's what I did
[21:26] <harshy> ok cool
[21:27] <mxpxpod> I'll email it to you tonite
[21:28] pb_ (~pb@2002:5160:45ef:0:240:95ff:fe30:aa84) joined #gstreamer.
[21:28] <harshy> cool
[21:30] psyduck ([email protected]) left irc: "bye everybody"
[21:34] <thomasvs> BBB: totem-gst plays mpeg and avi's fine here
[21:34] <thomasvs> well
[21:35] <thomasvs> sometimes sync wanders off :)
[21:37] mathrick ([email protected]) left irc: Read error: 60 (Operation timed out)
[21:39] <ds-work> BBB: how do I add a patch to gst-ffmpeg?
[21:40] mathrick ([email protected]) joined #gstreamer.
[21:41] robsta ([email protected]) joined #gstreamer.
[21:49] <BBB> ds-work: send it to me, or install quilt :)
[21:49] robsta ([email protected]) left irc: "Client exiting"
[21:49] <BBB> thomasvs: MPEGs? without dropping sound, losing sound, going to 1 FPS and more?
[21:49] <BBB> lucky bastard
[21:49] <BBB> anyone mind me re-patching the limits inside queue?
[21:50] <BBB> the current limits are insane
[21:50] <ds-work> go ahead
[21:50] Action: BBB adds to todo list for tonight
[21:50] <BBB> (todo list that says: fix totem)
[21:56] Company ([email protected]) joined #gstreamer.
[21:58] markey ([email protected]) left irc: "leaving"
[22:11] <ds-work> BBB: I have all but one chunk of asm fixed locally
[22:11] <ds-work> untested, of course :)
[22:11] <BBB> d'oh :D
[22:11] <BBB> test it ;)
[22:13] <BBB> to add a patch, simply add it to configure.ac (of gst-ffmpeg) in the patch list
[22:13] <BBB> I think it's slightly overkill to add it to the autotools patch
[22:13] Action: ds-work craves meat
[22:16] Action: BBB notes asm.diff in gst-ffmpeg CVS on gst-cvs
[22:17] <BBB> do I need to give you the line num in configure.ac?
[22:17] <BBB> :P
[22:17] <BBB> or will you find out yourself?
[22:17] <ds-work> I'll do it when it works :)
[22:17] <ds-work> it's already added locally
[22:18] <BBB> oh, and in case we want to send patches back: ffmpeg people don't like tabs, so please use spaces when possible
[22:18] <BBB> except in Makefile.am
[22:18] Action: BBB likes to send back patches
[22:18] <ds-work> ok
[22:18] <ds-work> I don't like spaces either
[22:19] Nick change: kh_zZz -> kh_out
[22:23] kmaraas ([email protected]) left irc: Read error: 110 (Connection timed out)
[22:33] LeRoutier ([email protected]) joined #gstreamer.
[22:33] <LeRoutier> re
[22:34] <LeRoutier> ds: are you around ?
[22:35] <ds-work> moo
[22:37] kmaraas ([email protected]) joined #gstreamer.
[22:40] <BBB> so, ds-work, will you test it? :)
[22:40] <BBB> or...
[22:41] <BBB> brb
[22:41] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[22:52] <Company> i didn't think esd was _that_ bad
[22:55] Nick change: LeRoutier -> LeRaway
[23:11] kwm ([email protected]) left irc: "Leaving"
[23:11] <grub_booter> esd bad - very bad - needs large buffers/introduces high latency and does an on the fly resampling that sets my teeth on edge (think it's the duplicate/drop sample approach, though not checked the code)
[23:12] smoke ([email protected]) left irc: Remote closed the connection
[23:12] <Company> and you can'T pause
[23:12] <Company> which is worse than all of that combined
[23:15] <mathrick> and you have absolutely _no_ control on if and when your sample will be played, which is the worst
[23:16] <grub_booter> nasty stuff
[23:18] <grub_booter> arts isn't much better from what i can tell - slightly smaller buffer, better resampling though
[23:18] <mathrick> and 40% CPU for free :P
[23:20] <grub_booter> still, there are worse things (take the avi file format for example... [being an aribtrary thing that i happen to be banging my head against at the moment])
[23:23] matt_101 ([email protected]) joined #gstreamer.
[23:25] lupusBE ([email protected]) left irc: "Bezig met verlaten"
[23:33] <sxpert> grub_booter: esd sucks (didn't you know that already ?)
[23:34] <mathrick> truth is, there isn't any sound server that doesn't suck badly, at least I don't know any
[23:35] <mxpxpod> which is why alsa has dmix now
[23:35] <grub_booter> heh - well, i haven't been giving it any support here, but I was pointing out why 'it sucks' :-)
[23:35] <sxpert> mxpxpod: what is dmix ?
[23:36] <mxpxpod> software mixing within alsa
[23:36] <grub_booter> cool
[23:36] <sxpert> mxpxpod: ah, I see
[23:36] thaytan ([email protected]) left irc: Read error: 60 (Operation timed out)
[23:36] <mxpxpod> yeah, it's pretty slick
[23:36] <grub_booter> does it have decent resampling to support that?
[23:36] <mxpxpod> no clue
[23:36] <sxpert> mxpxpod: ah yeah, I could launch multiple media apps this afternoon and you could hear mixed sounds...
[23:37] <mxpxpod> sxpert: ;)
[23:37] <sxpert> grub_booter: yeah, alsa has proper resampling (I use it on the carpc, the usb sound card knowing only 48000 Hz (why in hell did phillips settle on 44100 Hz is beyond me...
[23:39] <grub_booter> well, mixing is useful for those intermittent blippy things (e-mail clients that bleat are annoying, but people like it apparently...)
[23:40] <sxpert> grub_booter: lol
[23:40] <sxpert> grub_booter: it will be useful for my car gps app, so that the speech server that will do the "turn left" voice over can be separate from the music server
[23:41] <mxpxpod> sxpert: sweet!
[23:41] <mxpxpod> sxpert: when you get that all done, could you post that online somewhere?
[23:41] <grub_booter> well - you want the one that says 'don't go that way or you'll kill yourself' to be heard over everything
[23:41] <sxpert> mxpxpod: the sound server is available for testing at http://sxpert.esitcom.org/sound-server
[23:41] <mxpxpod> sxpert: no, I mean the whole car appliance thing
[23:41] <sxpert> grub_booter: yeah, volume hacking is also in order ;D
[23:42] <sxpert> mxpxpod: ah, that's already at http://sxpert.esitcom.org/projects/navsys (but is an OOOOOLD version )
[23:42] <grub_booter> mxpxpod, you could always just steal his car.,. 
[23:42] Action: sxpert kicks grub_booter
[23:42] <grub_booter> heh
[23:44] <grub_booter> i will be forced to get gps soon - my wife and i get lost every time we go somewhere new.. it's getting annoying :-/
[23:44] <sxpert> ha !
[23:44] BOFH ([email protected]) joined #gstreamer.
[23:44] <sxpert> then you'll be able to create free maps for belgium :D
[23:45] <grub_booter> heh
[23:46] smoke ([email protected]) joined #gstreamer.
[23:46] Company ([email protected]) left irc: Remote closed the connection
[23:47] <grub_booter> going by our recent experiences, that would involve lots of circles and weird directions guaranteed to confuse everyobne
[23:48] <grub_booter> we could provide a 'how to get from a to b via c to z' type of map if you think that'd help
[23:48] <sxpert> grub_booter: lol
[23:49] <sxpert> grub_booter: the trick is to have the actual shape of streets and roads, which is what is copyrighted. then we can generate street maps from that (and add street names at a later time...
[23:50] <iain> ds-work: ping?
[23:50] <ds-work> iain: yr ma
[23:51] <iain> ds-work: eff aff
[23:51] <iain> :)
[23:51] <BOFH> gstbasicscheduler.c: In function `gst_basic_scheduler_chainhandler_proxy':
[23:51] <BOFH> gstbasicscheduler.c:471: error: invalid lvalue in assignment
[23:51] <iain> ds-work: I'm trying to write a fixate function for audioconvert so that it won't always change the number of channels unless it has to
[23:51] <grub_booter> iain, you from scotland by any chance?
[23:52] <BOFH> that's the error I get building gstreamer
[23:52] <iain> ds-work: but I'm not sure a) how or b) if thats even what I'm meant to do
[23:52] <iain> grub_booter: Northern Ireland
[23:52] <grub_booter> iain, ah - k - wasn't so far off then
[23:53] <sxpert> lol
[23:53] <iain> grub_booter: no, nice and close
[23:53] <sxpert> iain: too close for comfort ?
[23:53] <grub_booter> heh
[23:54] <iain> nah, I don't mind scotland
[23:54] <iain> I lived in it for 3 years after all
[23:54] <grub_booter> yeah? which part?
[23:54] <iain> aberdeen
[23:54] <grub_booter> ah - sheep shagging territories
[23:54] <ds-work> iain: audioconvert should ideally not change the number of channels
[23:55] <ds-work> iain: but that's a pad_link() thing
[23:55] <sxpert> grub_booter: lol
[23:55] <sxpert> grub_booter: sounds like Jackass sort of thing
[23:55] <iain> ds-work: okay...
[23:55] <iain> ds-work: it is changing them and company said it was a fixate function problem
[23:55] <iain> grub_booter: indeed.
[23:56] <grub_booter> aberdeen's cool - my brother lives there
[23:56] <iain> grub_booter: my flatmate was very amused once when he was coming to aberdeen from edinburgh by train because he saw a farmer with a sheep over his shoulders
[23:56] <grub_booter> heh
[23:57] <grub_booter> the cliche goes something like 'aberdeen - where men are men... and the sheep are very afraid...'
[23:57] <iain> yup
[23:57] <iain> ah, the memories
[23:57] <grub_booter> lol
[23:58] <iain> what do you call a sheep tied to a lamppost in aberdeen?
[23:58] <iain> A leisure centre
[23:58] <grub_booter> heh
[23:58] <sxpert> iain: lol
[23:59] <grub_booter> what took you to aberdeen? uni?
[23:59] <iain> aye
[00:00] --- Thu Mar 11 2004
[00:00] <sxpert> iain: would that be a male or female sheep ?
[00:00] <grub_booter> sxpert: i don't think that matters 
[00:00] <grub_booter> ;-)
[00:00] <iain> any reason why gstreamer stuff is now being installed into gstreamer-0.8 directories
[00:00] <sxpert> grub_booter: heh
[00:00] <iain> but the version number is still 0.7.x?
[00:01] <grub_booter> anyway - i'm turning in.. nite
[00:01] <iain> night
[00:02] <sxpert> iain: someone forgot something ?
[00:04] spyder482 ([email protected]) joined #gstreamer.
[00:20] snikkt ([email protected]) joined #gstreamer.
[00:22] <thomasvs> iain: it's in the relnotes
[00:22] <thomasvs> iain: to shake out maj/min bugs
[00:28] jacob ([email protected]) joined #gstreamer.
[00:31] <iain> thomasvs: ok
[00:35] mxpxpod ([email protected]) left irc: "leaving"
[00:38] Shoragan ([email protected]) joined #gstreamer.
[00:38] <alley_cat> hmm, gst-ffmpeg is broken in a new way now
[00:38] Action: alley_cat wonder if there will ever be a projet using ffmpeg that isn't always borked in some way 
[00:39] <sxpert> alley_cat: well, ffmpeg itself is b0rken, ogg ac3 & mp3 don't work
[00:39] <alley_cat> ffmpeg is a pita when it comes to packaging, but it just is the best video decoder lib
[00:40] <sxpert> alley_cat: much progress is needed in the encoding dept though
[00:40] <alley_cat> although i never had problems with ffmepg inside gst-plugins, always have problems with gst-ffmpeg
[00:40] <alley_cat> hope that gets sorted out soon
[00:45] <alley_cat> well, ffmpeg has a focus on decoding
[00:45] <alley_cat> although ffmpegs mpeg4 encoder is quite good
[00:46] <alley_cat> the decoder totally blows away xvid
[00:46] <mathrick> g'night
[00:46] Nick change: mathrick -> mathrick|sleep
[00:46] <sxpert> alley_cat: heh
[00:47] <sxpert> alley_cat: I would like the encoder to work nicely too, because I do a hell of a lot of it...
[00:47] <alley_cat> every decoder sucks for xvid4 packed bitstream so far, but ffmpeg sucks the least ;)
[00:47] <sxpert> alley_cat: and being constrained to mpeg4 with mp2 sound blows...
[00:48] <alley_cat> sxpert: that's only ffmpeg itself, i never actually used that
[00:48] <alley_cat> i wouldn't use ffmpeg/libavcodec for anything audio related anyway, the seperate audio decoders are much better in most cases
[00:49] <sxpert> alley_cat: well, yeah, how would you encode a video that includes sound then ?
[00:50] <alley_cat> sxpert: encode video with ffmpeg, audio with seperate app, mux them together
[00:50] <sxpert> alley_cat: ah...
[00:50] <sxpert> uh..
[00:50] <alley_cat> got not much choice anyway right now, i usually create matroska files, so i have to use several tools
[00:50] <sxpert> alley_cat: ah...
[00:51] <alley_cat> oggenc for audio, currently mencoder for video and mkvmerge for muxing
[00:51] <alley_cat> one day i hope to do all that with gstreamer without any external tools :)
[00:52] <jacob> I have a problem with playing Ogg-files in Rhythmbox, using GStreamer 0.7.5. I think is is fixed in Gstreamer 0.7.6, am I right ?
[00:53] <jacob> (perhaps its GStreamer-plugins)
[00:54] Action: sxpert has wierd noisy sound on his motherboard integrated sound device (works fine with mplayer and xmms)
[00:54] <jacob> Well, my Ogg-files play fine when I use mplayer too.
[00:55] <sxpert> it's as though i'd be listening to am radio...
[00:55] <sxpert> badly tuned
[00:55] <alley_cat> using audioscale ! audiosink or audioscale ! audio/x-raw-int,rate=48000 ! audiosink fixed the noisy ogg output for me
[00:56] <alley_cat> depending on the audiosink
[00:57] <sxpert> alley_cat: ah, will try tomorrow
[00:57] Action: sxpert guesses it's more of the same "why did philips & sony choose 44100 Hz for CD..."
[01:02] teuf ([email protected]) left irc: "Client exiting"
[01:02] alley_cat ([email protected]) left irc: "May the Source be with you!"
[01:03] thaytoo ([email protected]) joined #gstreamer.
[01:07] <BOFH> ok...  I can;t gete gstreamer to build...  I wonder if anyone has built it w/ gcc 3.5 yet? (maybe that is my prob... I dunno)
[01:08] <ds-work> sxpert: http://www.isip.msstate.edu/data/mailing_lists/dsp/1999/archive/msg00047.html
[01:12] smoke ([email protected]) left irc: Remote closed the connection
[01:12] <BOFH> meh
[01:12] <BOFH> looks like gcc 3.5 is the bug in this case.
[01:13] <BOFH> Now, the question is...  The rest of gnome eup to this point has built error free...  Do we blame gcc or gstreamer? :)
[01:14] Shoragan ([email protected]) left irc: "Leaving"
[01:17] smoke ([email protected]) joined #gstreamer.
[01:18] <snikkt> why does this work: gst-launch osssrc ! audioscale ! audio/x-raw-int,rate=48000 ! volume volume=0.2 ! osssink
[01:18] <snikkt> and not this: gst-launch osssrc ! audioscale ! volume volume=0.2 ! osssink
[01:18] mxpxpod ([email protected]) joined #gstreamer.
[01:18] <ds-work> snikkt: because the latter is an unconstrained pipeline
[01:19] <snikkt> And it can't figure out how to connect it ?
[01:19] <BOFH> no comments / ideas?
[01:20] <ds-work> BOFH: gcc
[01:20] <ds-work> snikkt: it probably does connect it fine, it just doesn't do what you want
[01:21] <BOFH> hmm..   I figured it would be.
[01:21] <BOFH> wonder why gstream is the only one so far to puke...
[01:24] <ds-work> well, don't think too long about it, as gcc-3.5 isn't even remotely close to a stable compiler
[01:25] smoke ([email protected]) got netsplit.
[01:31] smoke ([email protected]) got lost in the net-split.
[01:32] sublett ([email protected]) left irc: "Nice of you to put me on the money"
[01:34] Rotty ([email protected]) left irc: Remote closed the connection
[01:41] <BOFH> now, after going back to gcc 3.3.2, why is it I am still getting the xoverlay X Includes error that was supposed to have been fixed?
[01:41] <BOFH> (in plugins)
[01:41] <BOFH> ../../../gst-libs/gst/xoverlay/xoverlay.h:27:22: X11/Xlib.h: No such file or directory
[01:44] pb_ (~pb@2002:5160:45ef:0:240:95ff:fe30:aa84) left irc: "Client exiting"
[01:48] arana ([email protected]) joined #gstreamer.
[01:52] spyder482 ([email protected]) left irc: Read error: 110 (Connection timed out)
[01:58] LeRaway ([email protected]) left irc: Read error: 110 (Connection timed out)
[01:58] LeRaway ([email protected]) joined #gstreamer.
[02:14] Nick change: LeRaway -> LeRoutier
[02:14] <LeRoutier> re
[02:15] <LeRoutier> BOFH: can't assign lvalue errors while compiling gstreamer is because gcc 3.5 is more strict than before and no more accept certain gcc-isms
[02:15] Nick change: kh_out -> khalek
[02:15] <LeRoutier> there has been a lot of patches lately for the kernel about this kind of error
[02:16] <BOFH> LeRoutier, ahh...   Very much appreciated - a REAL answer...
[02:16] <BOFH> (sorry)
[02:17] <BOFH> nice to know someone is paying attention, and it's nice to know that if nothing else, the error is expected, has a legitimate reason to it, and not the generic "don;t use CVS" response ;)
[02:17] harshy ([email protected]) left irc: Remote closed the connection
[02:18] <LeRoutier> well, this was working code before 3.5 but all gcc-ism are progressively no more supported to be more C99 compliant and more strict
[02:18] spyder__ ([email protected]) joined #gstreamer.
[02:18] <BOFH> indeed..  I changed my $CC spec to use the 3.3.2 ver, and it built fine...
[02:19] <BOFH> so it is definately a gcc 3.5 "bug" - but it's nice to hear an answer that is not "it's CVS and broken"
[02:19] <BOFH> Even if the answer means little or nothing to me...
[02:19] <BOFH> :)
[02:19] arana ([email protected]) left irc: Read error: 113 (No route to host)
[02:19] <LeRoutier> well no, it isn't a gcc bug
[02:20] <BOFH> thus the "quote" around bug..  
[02:21] <LeRoutier> actual code use a fuzzy written piece of C code which is no really readable. linus took a rather strict position on this against lvalue cast assignments. however, no one has to follow his opinion blindly
[02:23] thomasvs ([email protected]) left irc: Read error: 110 (Connection timed out)
[02:32] mathrick|sleep ([email protected]) left irc: Read error: 110 (Connection timed out)
[02:45] <LeRoutier> good night guys
[02:45] LeRoutier ([email protected]) left irc: "Leaving"


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click