IRC Logs

*******************************************************************
[03:04] benow ([email protected]) joined #gstreamer.
[03:08] <taaz> walters: i need some make fu.  i have multiple rules of the for dest: src otherstuff <some sed shell stuff>.  how can i use a common rule for different src/dest pairs?
[03:08] <taaz> s/for/form/
[03:08] <ds> this esds stuff is tricky
[03:09] <walters> taaz: can you give an example?
[03:10] <taaz> debian/$(gst_prefix)-tools.manpages: debian/gstreamer-tools.manpages debian/rules
[03:10] <taaz> the some shell stuff to implement it that is common for all pairs of files
[03:11] <taaz> ie, i'm generating a bunch of debian/ files that need text replacement for various vars
[03:11] Action: BBB gets really interesting sound using gsmdec now
[03:11] <taaz> and for no apparent reason i'm using make rules rather than just generating them all the time ;)
[03:12] <walters> probably something like debian/$(gst_prefix)-%.manpages: debian/gstreamer-%.manpages:
[03:12] <taaz> well, it's more like .manpages, .links, .install, etc etc
[03:13] <walters> that's harder
[03:13] <taaz> and -tools -docs -blah -blah
[03:13] <taaz> yeah, which is why i asked ;)
[03:15] <walters> if you want to get your hands dirty you could use $(eval ...) :)
[03:17] <taaz> or i could punt the problem and depend on pre-build to do the work and use a for loop
[03:18] <walters> yeah
[03:18] <walters> unfortunately make is a pretty sucky language.
[03:18] <taaz> yeah.  good for some things, quite difficult for others
[03:20] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[03:22] <BBB> does anyone know gsm here?
[03:32] <taaz> i do not.
[03:35] <BBB> hm, it's causing issues for me ;)
[03:37] <BBB> ds && ds-work: wanna help me more?
[03:37] <ds> I don't know anything about gsm
[03:38] mxpxpod ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[03:39] <BBB> ds: does quicktime have something like block_align or so?
[03:41] <ds> not that I know og
[03:41] <ds> of
[03:43] <ds> but then, I'm running into a lot of stuff just now that is undocumented in qt
[03:46] harshy ([email protected]) left irc: Read error: 60 (Operation timed out)
[03:52] harshy ([email protected]) joined #gstreamer.
[03:53] jcsston ([email protected]) joined #gstreamer.
[03:54] Action: Company believes ffmpeg typefinding is borked
[03:54] mxpxpod ([email protected]) joined #gstreamer.
[03:55] <BBB> there's so much borked
[03:58] <jcsston> can gstreamer play directly from an audio cd?
[03:58] <Company> yes
[03:58] <Company> the plugin is called cdparanoia i believe
[03:59] Rotty ([email protected]) left irc: Remote closed the connection
[03:59] <jcsston> ah, ok
[04:00] synap ([email protected]) joined #gstreamer.
[04:01] <BBB> hey dennis
[04:01] <synap> ha hallo!
[04:01] <synap> ik denk ik join eens =]
[04:01] <Portable> Does someone know why gstreamer might be printing this out?  Is this supposed to be a bad thing?  GStreamer-WARNING **: pipeline: state change in callback 4 8
[04:02] <BBB> synap: goed plan!
[04:02] <BBB> Portable: own app? or a known app?
[04:02] <Portable> My own
[04:02] <BBB> well, yes, it's bad because otherwise it wouldn't print this out
[04:02] <BBB> you're using callbacks to do stuff to pipelines
[04:02] <BBB> it's better to register an idle handler using g_idle_add() and do the actual work in there
[04:03] <BBB> that also prevents thread issues
[04:03] <Portable> Hmm
[04:05] <Portable> Basically I have a callback setup on the eos signal from the file source, and it calls a couple functions to iterate through my playlist and start playing the next track.
[04:08] <Portable> What does the g_idle_add do, set a callback for when the thread is not playing?
[04:09] <BBB> glib idle loop handler
[04:14] <BBB> ds: read http://developer.apple.com/documentation/QuickTime/QTFF/QTFFChap3/chapter_4_section_38.html and seek for 'GSM' in the document
[04:14] <BBB> ds: might that help in fixing this?
[04:15] <BBB> [spam]
[04:15] <BBB> Samples per compression block
[04:15] <BBB>     A 16-bit unsigned integer specifying the uncompressed samples per compression block. A value of 0 is equivalent to a value of 1. 
[04:16] <BBB> If the bytes per compression block and/or the samples per compression block is greater than 1, than this ratio is used to translate a sample number into an actual byte offset.
[04:16] <BBB> This ratio mode is typically used for compressed audio tracks. Note that for QuickTime sound tracks, the bytes per compression block also factors in the number of sound channels in that stream, so a QuickTime stereo sound streams BPCB would be twice that of a mono stream of the same sound format.
[04:16] <BBB> (CB = NS * BPCB / SPCB)
[04:17] <BBB> where CB = compressed bytes, NS = number of samples, BPCB = bytes per compression block, and SPCB = samples per compression block.
[04:17] <BBB> An example:
[04:17] <BBB> A GSM compression block is typically 160 samples packed into 33 bytes.
[04:17] <BBB> So, BPCB = 33 and SPCB = 160.
[04:17] <BBB> The hint sample requests 33 bytes of data starting at the 161st media sample. Assume that the first QuickTime chunk contains at least 320 samples. So after determining that this data will come from chunk 1, and where chunk 1 starts, you must use this ratio to adjust the offset into the file where the requested samples will be found:
[04:17] <BBB> [/spam]
[04:24] spider-man2010 ([email protected]) left irc: Read error: 110 (Connection timed out)
[04:26] MuMart ([email protected]) left irc: "Leaving"
[04:27] spider-man2010 ([email protected]) joined #gstreamer.
[04:37] <taaz> walters: oh arch expert, i did ^c in the middle of a commit (typing in the wrong dir...) when it asked for my gpg pw, now a commit fails and says this:
[04:37] <taaz> arch_commit: unable to acquire revision lock (could not rename file.)
[04:37] <taaz>     tree: /home/dlehn/projects/debian/pkgs/gstreamer0.8/gstreamer0.8-0.8.1
[04:37] <taaz>     revision: [email protected]/gstreamer--debian--0.8--patch-7
[04:37] <walters> taaz: tla lock-revision -b [email protected]/gstreamer--debian--0.8--patch-7
[04:38] <walters> taaz: (i've been meaning for a while now to make it so lock-revision -b with no args uses the latest version)
[04:39] Action: taaz bows to the arch master
[04:39] <taaz> that fixed it, thanks ;)
[04:40] <taaz> the error message is not as informative as it could be
[04:40] <walters> heh well it does say to use lock-revision -b in the error message you know :)
[04:41] Misirlou ([email protected]) got netsplit.
[04:41] <taaz> eh?
[04:42] Misirlou ([email protected]) returned to #gstreamer.
[04:43] <taaz> tools like this need an ARCH_HELP_MODE=newbie option to tell people like me what to do for an error like that.  i don't know what revision locks are, didn't know lock-revision existed, and wouldn't have known what to type if i did. ;)
[04:43] <taaz> btw, have you built new rb debs yet? ;)
[04:43] <walters> i thought the error message mentioned lock-revision
[04:44] <taaz> or is the gnome team doing that these days?
[04:44] <taaz> i pasted the error message i got
[04:44] <walters> gnome team maintains it
[04:44] <Company> walters is a guru who knows how to aenable newbie mode
[04:44] Action: Company yestereday googled what revision-locks are because he had the same issue
[04:47] <walters> btw, you guys may be interested in this: http://web.verbum.org/files/gnome-gpg-0.1.tar.bz2
[04:47] <walters> (not announced yet since the archive it's in isn't yet public)
[04:48] <walters> screenshot: http://web.verbum.org/files/gnome-gpg.png
[04:51] <BBB> walters: what's that?
[04:51] Action: BBB tries to fix bugs and notices how much info he misses
[04:51] <walters> BBB: a little wrapper for gpg that uses gnome-keyring
[04:53] <taaz> i do so much of my work over text terminals that gnome apps are not too useful ;)
[04:54] Action: BBB goes home
[04:54] <BBB> bbl :)
[04:55] <BBB> home?
[04:55] <BBB> hm
[04:55] <BBB> to_bed?
[04:55] <BBB> :)
[04:55] <BBB> later!
[04:55] BBB ([email protected]) left irc: "Client exiting"
[05:00] <taaz> ok, lintian clean core and plugins debs done
[05:05] <taaz> so what does libgst do that takes 1.1M?
[05:05] <taaz> that's with --enable-debug/DEBUG but still seems like a lot of code
[05:09] <Company> --disable-gst-debug shoulg save some code
[05:10] <Company> dunno what else causes that size - and i don't care either
[05:32] lilo ([email protected]) left irc: " "
[05:32] lilo ([email protected]) joined #gstreamer.
[05:43] sub_pop ([email protected]) left irc: "Client exiting"
[06:23] Portable ([email protected]) left irc: "Leaving"
[06:31] steve_b ([email protected]) joined #gstreamer.
[06:36] walters ([email protected]) left irc: "out"
[06:42] steve_b_ ([email protected]) joined #gstreamer.
[06:42] desrt ([email protected]) joined #gstreamer.
[06:51] Company ([email protected]) left irc: Read error: 110 (Connection timed out)
[06:58] <desrt> i'm trying to debug a gstreamer bug.... i need to create a .wav file...
[06:58] <desrt> so i tried following the capabilities example in the gst-launch manpage.... basically, i'm doing:
[06:58] <desrt> filesrc ! spider ! audio/x-wav ! filesink
[06:59] <desrt> and expecting that to decode the input and give me a wav.... unfortunately it gives me a 0 byte file with no error message
[07:01] jcsston` ([email protected]) joined #gstreamer.
[07:01] steve_b ([email protected]) left irc: Read error: 110 (Connection timed out)
[07:02] <desrt> the pipeline being constructed appears to contain oggdemux but no vorbisdecode or wavenc (for an ogg input)
[07:03] <desrt> (according to gst-inspect the only plugin i have that has a src with an audio/x-wav capability is wavenc)
[07:19] sub_pop ([email protected]) joined #gstreamer.
[07:24] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[07:39] Nick change: jcsston` -> jcsston
[07:45] ChrisHJW ([email protected]) joined #gstreamer.
[08:17] markey ([email protected]) joined #gstreamer.
[08:22] ChrisHJW ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[08:33] jcsston ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[08:35] tvon ([email protected]) joined #gstreamer.
[08:37] spider-man2010 ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[08:38] <tvon> Any clues what an 'erroneous pipleline' error might be from" I'm running some self-rolled RPM's here that worked on my last system, but after a new install it craps out
[08:38] <tvon> gst-launch-ext foo.mp3 spits:
[08:39] <tvon> WARNING: erroneous pipeline: no element "mad"
[08:39] <tvon> among other things...static  comes out of the speakers
[08:39] <tvon> perhaps I goofed in compiling?
[08:39] <tvon> gst-launch-0.8 filesrc location="/mnt/multimedia/Music/Dead Milkmen, The/Beelzebubba/01 Brat In The Frat.mp3" ! mad ! osssink
[08:39] <tvon> tis the gst-launch command that got used
[08:39] <tvon> oggs play fine
[08:53] <_246tNt_pc_> Are you sure the mad element has been compiled ?
[08:53] <_246tNt_pc_> /usr/lib/gstreamer-0.8/libgstmad.so ?
[08:54] <_246tNt_pc_> gst-inspect mad ?
[08:55] <tvon> ah
[08:55] <tvon> _246tNt_pc_: there you have it, not ther
[08:56] Action: tvon goes to rebuild
[08:57] <_246tNt_pc_> be sure to had libmad installed so it's detected and gstmad will be built.
[08:58] <tvon> yeah, I didnt have libmad-devel when I built it before
[08:58] Action: tvon adds it to BuildRequires
[09:19] Shoragan ([email protected]) joined #gstreamer.
[09:24] kmaraas ([email protected]) left irc: "Leaving"
[10:02] tvon ([email protected]) left #gstreamer.
[10:11] steve_b ([email protected]) joined #gstreamer.
[10:15] foser ([email protected]) joined #gstreamer.
[10:20] ChriHJW_log ([email protected]) joined #gstreamer.
[10:31] Rotty ([email protected]) joined #gstreamer.
[10:31] dolphy ([email protected]) joined #gstreamer.
[10:34] <dolphy> morning
[10:34] <steve_b> hi
[10:34] ChrisHJW_log ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:45] hallibaby ([email protected]) joined #gstreamer.
[10:46] hallibaby ([email protected]) left #gstreamer ("Client exiting").
[10:48] jdahlin ([email protected]) joined #gstreamer.
[10:53] Nick change: markey -> markey|afk
[11:03] thomasvs ([email protected]) left irc: Client Quit
[11:06] steve_b ([email protected]) left irc: Remote closed the connection
[11:12] thomasvs ([email protected]) joined #gstreamer.
[11:13] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) joined #gstreamer.
[11:20] <jdahlin> Ronald> Give me a few days to work on it... ;).
[11:20] <jdahlin> 4 hours later
[11:20] <jdahlin> Ronald> Fixed in CVS.
[11:22] <thomasvs> attaboy
[11:33] <thomasvs> sheesh, reading backlog of gnome-hackers now, I'm not sure BBB and Company "get it" wrt. gnome integration
[11:33] jacob_ ([email protected]) joined #gstreamer.
[11:33] <jacob_> Would it be possible to compile and install a mp3-plugin locally on my account ? (running FC2T2, there of the missing mp3 support)
[11:35] <thomasvs> jacob_: it would, but it's probably easier if you try to install the FC1 rpm for it, or try to rebuild that
[11:36] <jacob_> I don't wanna install 3hd party rpms yet, but I'll go for compiling/installing it locally if thats possible.
[11:37] <jacob_> Just wanted to be sure it was possible before I dove into it 8)
[11:38] <thomasvs> http://gstreamer.freedesktop.org/pkg/fedora/1/i386/RPMS.gst/gstreamer-plugins-extra-audio-0.8.1-0.lvn.1.1.i386.rpm
[11:39] <thomasvs> this is our rpm containing that, and probably works well with fc2t2
[11:39] <thomasvs> jacob_: you don't want to install rpms, but want to install from source ? why ?
[11:40] <jacob_> thomasvs, thanks, ill look into that one. I mean, I don't wanna install rpms, but I don't mind cluttering my $HOME file structure. Thats what I'm hoping is possible.
[11:40] Rotty ([email protected]) left irc: Read error: 110 (Connection timed out)
[11:42] steve_b__ ([email protected]) joined #gstreamer.
[11:44] <thomasvs> jacob_: it sort of should be ... but it will require some extra things to be done
[11:44] <jacob_> thomasvs, root access in some way ?
[11:45] <thomasvs> jacob_: no, setting env vars for extra dirs and so on
[11:45] <thomasvs> dolphy: can you restart your buildbot slave ?
[11:47] jacob ([email protected]) joined #gstreamer.
[11:47] <dolphy> thomasvs: yup
[11:47] <jacob> thomasvs, ah ok, that shouldn't be any problems i hope.
[11:47] <dolphy> thomasvs: i did major modifications yesterday to the machine :)
[11:47] <thomasvs> dolphy: like ?
[11:48] <thomasvs> someone *please* fix MPEG for real
[11:48] <dolphy> thomasvs: switching my router to pppoe
[11:48] <thomasvs> it sort of works for me to the effect that it works too well to actually debug it, since I can't technically define what the problem is
[11:49] <dolphy> thomasvs: so that i can try to get gnomemeeting to work
[11:49] <thomasvs> hehe :)
[11:49] <jdahlin> thomasvs: something wrong with mpeg, apart from scheduler issues?
[11:49] <thomasvs> you don't need gnomemeeting, you can write it yourself, no ?
[11:49] <thomasvs> jdahlin: "apart from" ? :)
[11:49] <thomasvs> jdahlin: it just needs to play and play reliably
[11:55] <thomasvs> it's incredibly silly to have company and bbb arguing for totem-gst in gnome 2.8 *when we cannot even play mpeg well*
[12:00] Rotty ([email protected]) joined #gstreamer.
[12:00] steve_b_ ([email protected]) left irc: Read error: 110 (Connection timed out)
[12:03] jacob_ ([email protected]) left irc: Client Quit
[12:05] <jdahlin> this is just plain ugly:
[12:05] <jdahlin> gst_debug_log_valist (GstDebugCategory * category, GstDebugLevel level,
[12:05] <jdahlin>     const gchar * file, const gchar * function, gint line,
[12:05] <jdahlin>     GObject * object, const gchar * format, va_list args)
[12:05] bilboed_school ([email protected]) joined #gstreamer.
[12:06] <thomasvs> jdahlin: it doesn't matter since it doesn't get called except in one place :)
[12:06] <jdahlin> thomasvs: it does, since I found a bug in the code and have to read it
[12:07] <jdahlin> the function prototypes used to look good
[12:11] <thomasvs> ah, ugly in that way you mean
[12:11] <thomasvs> that's a matter of perspective and what you're used to :)
[12:12] <jdahlin> gtk+ like indentation?
[12:12] <thomasvs> see, that's what I mean :) you're used to one style.
[12:12] <thomasvs> personally I don't care, I care about consistency.
[12:13] <jdahlin> I care a lot more about readability than consistency
[12:14] <thomasvs> it's not unreadable, you're just not used to it.  entirely different problem.
[12:14] <thomasvs> I got used to it fairly quickly when going through it
[12:14] <jdahlin> we can run it through some loop-unroller scripts if you like, we'll still have consistency
[12:16] <jdahlin> I was used to the old style, which is commonly used in the gnome projects etc
[12:16] <jacob> Small Q: In the configure script, is there a way to disable building all plugins except for one ?
[12:19] <thomasvs> jacob: --with-plugins=mad (which will still build a bunch of others, but ignore those)
[12:19] <thomasvs> jacob: then, cd ext/mad
[12:19] <thomasvs> make
[12:20] <jacob> ok, thanks
[12:24] Nick change: markey|afk -> markey
[12:29] bilboed_school ([email protected]) left irc: "Leaving"
[12:35] Company ([email protected]) joined #gstreamer.
[12:41] wheels ([email protected]) left irc: "work"
[12:45] jacob_ ([email protected]) joined #gstreamer.
[12:47] ploum ([email protected]) joined #gstreamer.
[12:47] <ploum> hello
[12:48] <ploum> ds, I've reported the bug.  I've recompiled my linux kernel. Now, ALSA works for audio application  but for gstreamer I have the following error : 
[12:48] <ploum> gst-launch-0.8: pcm_plug.c:882: snd_pcm_plug_hw_params:  l'assertion « err >= 0 » a échoué.
[12:49] <ploum> don't know what it means !
[12:49] <thomasvs> ploum: set your alsasink to "alsasink device=hw:0" and try again
[12:51] <ploum> thomasvs, thx
[12:51] <Company> or better: help me debug that issue
[12:51] <ploum> it gives me :
[12:51] <ploum> ERROR: from element /pipeline0/spider0/audioconvert0: Internal GStreamer error: negotiation problem.  File a bug.
[12:51] <ploum> Additional debug info:
[12:51] <ploum> gstaudioconvert.c(246): gst_audio_convert_chain: /pipeline0/spider0/audioconvert 0:
[12:51] <ploum> Sink pad (connected to vorbisdec0:src) not negotiated before chain function
[12:51] <ploum>  
[12:51] <ploum> (process:11944): GStreamer-WARNING **: push on peer of pad vorbisdec0:src but pe er is not active
[12:51] <ploum> the last line is reproduced several times
[12:52] <Company> ploum: what alsalib version do you use?
[12:52] <thomasvs> Company: are you actively discussing this with the alsa guys ? Some people have been telling me there might be better workarounds than using hw:0
[12:53] <thomasvs> Company: e.g., dmix for example
[12:53] <thomasvs> Company: but I have no clue about alsa
[12:53] jacob ([email protected]) left irc: Client Quit
[12:53] <Company> thomasvs: the alsa guys closed my bug because i couldn't help them debug that issue
[12:53] <ploum> Company, the kernel 2.6.5 one
[12:54] <Company> thomasvs: people that had that issue weren't interested in helping to fix it
[12:54] <Company> ploum: alsalib, not the drivers
[12:55] <ploum> Company : libasound2:  1.0.4-1   (in Debian, alsalib package is deprecated)
[12:56] iain ([email protected]) joined #gstreamer.
[12:56] <ploum> is that what you want ?
[12:56] <iain> maybe
[12:56] <iain> ?
[12:56] <ploum> thomasvs, I try to ask on #alsa, but nobody answer
[12:57] <Company> ploum: yeah, that's what i meant
[12:57] <ploum> iain, I was replying to Company ;)
[12:57] <iain> ploum: I knew it wasn't me :)
[12:57] steve_b__ ([email protected]) left irc: Remote closed the connection
[12:57] <Company> ploum: there's a bug report about this: http://bugzilla.gnome.org/show_bug.cgi?id=134007
[12:58] <thomasvs> Company: I saw a FC2test user add additional info to your bug report because walters asked to
[12:58] <thomasvs> Company: so maybe it can be resolved with the help of that user ?
[12:59] <Company> thomasvs: yeah, i'm gonna get back to him, too
[12:59] <ploum> I've recompiled my kernel this morning and now alsa is working perfectly for totem and mplayer but not for BMP and alsaplayer (!).
[13:00] <ploum> Company, it's not the same bug because it doesn't work for me with hw:0
[13:00] <ploum> But now, gstreamer has a lot of debugging error (with my old kernel, I had only a "tube failed" )
[13:01] <Company> ploum: in that case you probably should contact the alsa guys directly (or their bugzilla)
[13:01] Action: jacob_ got the mad decoder plugin working, but only if libid3tag & libmad were dynamically linked to the libgstmad.so plugin, not statically
[13:02] <thomasvs> Company,ploum: with hw=0 he's getting a different, gstreamer, error, which is something else to be debugged, no ?
[13:02] <Company> yeah, i've had my fair share of problems with building plugins from static libs, too
[13:03] <Company> thomasvs: most likely he needs audioconvert
[13:03] <jacob_> I should perhaps file this in bugzilla...
[13:03] bilboed_school ([email protected]) joined #gstreamer.
[13:03] <Company> s/audioconvert/audioscale/
[13:05] <Company> spider doesn't plug audioscale...
[13:06] Nick change: bilboed_school -> bilboed
[13:06] <ploum> thomasvs, you are right !  With hw=0, this is a gstreamer error.  Other error is an alsa error
[13:06] <Company> what error do you get?
[13:08] <Company> ploum: anyway, https://bugtrack.alsa-project.org/alsa-bug/bug_view_page.php?bug_id=0000116 is the bug we opened for that issue (click on guest on login page)
[13:08] <ploum> with a MP3 file and mad plugin :
[13:08] <ploum> ERROR: from element /pipeline0/mad0: Internal GStreamer error: pad problem.  File a bug.
[13:08] <ploum> Additional debug info:
[13:08] <ploum> gstpad.c(2403): gst_pad_set_explicit_caps: /pipeline0/mad0:
[13:08] <ploum> failed to negotiate (try_set_caps with "audio/x-raw-int, endianness=(int)1234, signed=(boolean)true, width=(int)16, depth=(int)16, rate=(int)44100, channels=(int)2" returned REFUSED)
[13:08] <Company> try adding audioconvert between mad and osssink
[13:10] <jdahlin> Is it feasible to extend/modify spider to create threads with queues?
[13:10] <ploum> Company : same ! (but this is with alsasink.  With osssink, I have no error but no sound and a 100% CPU !)
[13:10] <ploum> now with an OGG file and spider
[13:10] <ploum> ERROR: from element /pipeline0/spider0/audioconvert0: Internal GStreamer error: negotiation problem.  File a bug.
[13:10] <ploum> Additional debug info:
[13:10] <ploum> gstaudioconvert.c(246): gst_audio_convert_chain: /pipeline0/spider0/audioconvert0:
[13:10] <ploum> Sink pad (connected to vorbisdec0:src) not negotiated before chain function
[13:10] <ploum>  
[13:11] <ploum> (process:12829): GStreamer-WARNING **: push on peer of pad vorbisdec0:src but peer is not active
[13:11] <ploum> Last line is reported several time..
[13:11] <ploum> Very strange
[13:13] <Company> could you mail me the output of this command:
[13:14] <Company> ploum: try adding audioscale after audioconvert
[13:16] <ploum> Company !!! with autoscale in my pipeline and alsa hw0, all work perfectly ! both mp3 and ogg !
[13:16] <ploum> but why ?  is my audio driver ?
[13:17] <Company> ploum: forget the mailing me
[13:17] <Company> ploum: most likely your audio driver or soundcard does not support 44.1kHz audio
[13:18] <Company> ploum: which unfortunately is the default audio format used in ogg, mp3 and cds
[13:18] <Company> ploum: so you need to resample by using audioscale (most likelyt it resamples to 48kHz)
[13:19] <iain> hmmm
[13:19] <ploum> oh.. Thx
[13:19] <Company> setting the gconf audiosink to audioscale ! alsasink device=hw:0 should do the trick for now i guess, but i would still greatly appreciate someone fixing this stupid alsalib bug :)
[13:19] <ploum> But it's very strange to not support 44.1 !!
[13:19] <iain> my midiverb ii is broken :/
[13:20] <ploum> Company, alsa people don't seem to be very reactives ::(
[13:20] <Company> ploum: i agree that it's very strange - that's why we get so little bug reports about it ;)
[13:21] <Company> ploum: and no, the alsa guys aren't very reactive - especially on IRC or the mailing lists
[13:21] <Company> ploum: their responsiveness on their bug tracker is ok though
[13:22] <thomasvs> Company: so wouldn't setting it to dmix instead of hw cause alsa to resample instead ?
[13:22] <Company> thomasvs: leaving it at just alsasink should cause alsa to resample instead
[13:23] <Company> thomasvs: setting it to dmix opens a whole other can of worms
[13:23] <iain> thomasvs: dmix doesn't do the resampling, its the plug thing that does
[13:23] <iain> dmix just uses plug
[13:24] <iain> and if you have your asoundrc file set up right, you don't need to explicitly use dmix
[13:24] <thomasvs> ok, so all bugs that currently are solved by using hw:0 can be solved with a correct .asoundrc ?
[13:25] <iain> possibly
[13:25] <Company> i wouldn't call working around bugs "correct"
[13:25] <iain> we should probably defintely be using plughw:0 I think
[13:25] <Company> but can be solved by working around the issue with a different .asoundrc
[13:25] <iain> or just default
[13:25] <Company> we use default
[13:26] <iain> ok
[13:26] <jdahlin> what about defaulting to plughw:0?
[13:26] <Company> jdahlin: plughw:0 is default
[13:26] <Company> (unless you mess with .asoundrc)
[13:29] <jdahlin> ok
[13:37] <jdahlin> "Personally, I think the stability issues are partly related to the use of C."
[13:37] <Company> what are you reading now?
[13:39] <jdahlin> markey's weblog
[13:39] <jdahlin> http://www.kdedevelopers.org/node/view/426
[13:41] <iain> Without a working KIO plugin, GStreamer will only be able to play local files
[13:41] <iain> that should read "Without a working KIO plugin, GStreamer will only be able to play remote files, but won't use KIO"
[13:41] <jdahlin> well, from their point of view, gnomevfs is not an option
[13:42] <jdahlin> making a good kio plugin is definitely important for kde adoption
[13:42] <iain> they can still use it, and not be aware of it
[13:42] <iain> I'm not denying that, but I think he painted it a bit too drastically
[13:42] <jdahlin> Agreed
[13:43] <iain> whats mlt though?
[13:43] <jdahlin> dunno
[13:43] <Company> mlt is a framework that thinks about using GStreamer in the future
[13:44] <jdahlin> Company: have you ever worked on the autopluggers/spider?
[13:44] <Company> mlt.sf.net
[13:44] <thomasvs> Company: no, it's thinking about a way to bridge the two, slightly different
[13:44] <Company> jdahlin: i'm currently working on the new one
[13:44] <jdahlin> Company: oh, nice
[13:44] <jdahlin> Company: I'm currently wondering if it's possible to change the current one to use threads and queues
[13:45] <Company> jdahlin: ugh
[13:45] <Company> jdahlin: i've kept that in mind for the new one, but even in that case it's most likely not an easy task
[13:46] jacob__ ([email protected]) joined #gstreamer.
[13:46] <jdahlin> Company: I haven't worked enough on to find any things that'll stop it from working, but why not just create a thread for each new element, add a queue to the thread and then the element?
[13:46] <lilo_booter> jdahlin: that comment about stability being related to use of c is kinda funny :-) - haven't read the article to see the context...
[13:47] <jdahlin> lilo_booter: it's not totally of context, but I don't think C++ (which I guess markey suggest that people should use instead) is the solution :)
[13:47] <markey> I'm not saying C is the root of all evil :)
[13:47] <Company> jdahlin: feel free to try it
[13:48] <lilo_booter> markey: heh - i should hope not :-)
[13:49] <Company> markey: the current segfaults with KIO are all nicely drowned in KDE/Qt code, so I can't really do much there
[13:49] <iain> markey: I guess trying to get KIO split so that the problems that company came across could be fixed would be out of the question?
[13:49] <jdahlin> Company: I might just do.
[13:50] <jdahlin> Company: Do you know where I can look where it "adds" upp element that are not demuxers (with multiple pads) ?
[13:50] <Company> jdahlin: grep the code for gst_bin_add
[13:50] <markey> iain: I'm not in a position to split KIO, but of course one can discuss ways to improve on stuff
[13:51] <iain> markey: but you would be in a position to suggest it?
[13:51] <iain> and to be able to argue for it if you thought it might help?
[13:52] <jdahlin> Company, currently reading http://article.gmane.org/gmane.comp.video.gstreamer.devel/10217
[13:52] <jdahlin> Company: I'd just like to sort out the difference between a "mpegdemux" like object and the others
[13:53] <Company> jdahlin: spider doesn't care
[13:53] Action: Company grabs lunch now
[13:53] <jdahlin> Company: okay
[13:53] Company ([email protected]) left irc: Remote closed the connection
[13:53] <jdahlin> Maybe I can check if it's parent is called spider_indent :-)
[13:53] <jdahlin> ident
[13:54] jacob_ ([email protected]) left irc: Client Quit
[14:00] ploum ([email protected]) left irc: "Leaving"
[14:00] markey ([email protected]) left irc: "bbl"
[14:05] Action: iain wonders how he loses his driving license so much
[14:06] <jdahlin> huh?
[14:06] tlg ([email protected]) joined #gstreamer.
[14:13] BBB ([email protected]) joined #gstreamer.
[14:13] <BBB> thomasvs: awake?
[14:15] <jdahlin> BBB, he just left for lunch
[14:16] <jdahlin> he'll be back in an hour and a half or so (maximum)
[14:17] <jdahlin> BBB, I have a couple of question for you, if you're not too busy working
[14:17] <BBB> I'm having breakfast ;)
[14:17] <BBB> tell thomasvs to go through the gnome-media patchlist
[14:17] <BBB> :p
[14:17] <BBB> there's some that's really easy that need applying, some even before we release 2.6.1
[14:18] <jdahlin> Will do
[14:18] <jdahlin> he'll probably read up on his backlog when he comes back
[14:18] <jdahlin> BBB, so. I'm pondering modifying spider to create threads with queues, so we can have decent mpeg playback [with opt]
[14:19] <BBB> oh my god, you're gonna touch spider?
[14:19] <BBB> it was nice having known you
[14:19] Action: BBB RIPs jdahlin
[14:19] <BBB> :p
[14:19] <jdahlin> :D
[14:20] <jdahlin> I figured I've already hacked opt so I'm looking for new "adventures"
[14:20] <BBB> http://bugzilla.gnome.org/buglist.cgi?short_desc_type=allwordssubstr&short_desc=&product=gnome-media&long_desc_type=allwordssubstr&long_desc=&status_whiteboard_type=allwordssubstr&status_whiteboard=&keywords_type=anywords&keywords=PATCH&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&emailtype1=substring&email1=&emailtype2=substring&email2=&bugidtype=include&bug_id=&changedin=&chfieldfrom=&chfieldto=Now&chfieldvalue
[14:20] <BBB> =&cmdtype=doit&newqueryname=&order=Reuse+same+sort+as+last+time&field0-0-0=noop&type0-0-0=noop&value0-0-0= <- for thomasvs
[14:20] <BBB> it sounds cool... if you know how spider does it, please go for it
[14:20] <BBB> I guess that for each newly created sometimes pad, you could do this
[14:20] <jdahlin> For all elements which parent isn't the original pipeline (sink_ident)
[14:21] <jdahlin> that'd solve the mpeg issue, but might break other things, so maybe I'll have to implement it as a property of some kind
[14:21] <BBB> what would it break?
[14:21] <BBB> I don't think it'd break anything
[14:23] <jdahlin> It changes the semantics slightly
[14:23] <jdahlin> might break everything :)
[14:23] <BBB> yes, but I think we all agree that this is the best workaround for now
[14:23] <jdahlin> why haven't anyone tried to fix it before then?
[14:23] <BBB> I don't understand spider a slight bit
[14:23] <BBB> I only know how typefinding works in spider
[14:24] <BBB> the plugging bit is complete magic for me
[14:24] <jdahlin> so? as do I
[14:24] <BBB> so I didn't want to touch it ;)
[14:24] <BBB> given that ds is working on a new one anyway
[14:24] <BBB> (don't forget that my time is limited, if if takes me 40 hours to learn spider, that's 4-5 weeks!)
[14:25] <jdahlin> hmm, right
[14:25] <jdahlin> I think we all agree that it's better if you spend your time hacking on ffmpeg and plugins in general
[14:25] <jdahlin> :-)
[14:26] <desrt> anyone see my issue last night?
[14:26] <desrt> i'm basically doing filesrc ! spider ! audio/x-wav ! filesink and expecting to get a wav file output...
[14:26] <desrt> and instead i'm getting a 0 byte file and no error messages
[14:27] <BBB> jdahlin: it takes less time
[14:27] <BBB> desrt: spider only works for decoding
[14:27] <BBB> for encoding, anything could happen
[14:28] <BBB> especially given that wavenc has no rank, so it'll never select a wav encoder
[14:28] <desrt> BBB; won't it just try to find an element that has output caps of audio/x-wav?
[14:28] <desrt> i see.
[14:28] <BBB> sorry ;)
[14:28] <desrt> so i want spider ! wavenc ! filesink?
[14:28] <desrt> oo.  that rhymed.
[14:28] <BBB> encoders in general have no rank
[14:28] <BBB> so you really want to not use spider
[14:28] <desrt> excellent.  this one is working
[14:29] <BBB> wavenc might actually work
[14:29] <desrt> it did =)
[14:29] <BBB> the result is still unclear though
[14:29] <BBB> it might decode the file to audio/x-raw-int
[14:29] <BBB> but it also might transcode mp3 files to mp3-in-wav
[14:29] <BBB> both are valid
[14:29] <desrt> foo.wav: RIFF (little-endian) data, WAVE audio, Microsoft PCM, 16 bit, stereo 44100 Hz
[14:30] <desrt> looks like it did ok.
[14:30] <BBB> :)
[14:30] <desrt> in gstreamer i trust :)
[14:31] <desrt> gstreamer has a PPC problem where the vorbis plugin does odd things
[14:31] <desrt> so i'm trying to make sure it is in fact the vorbis plugin by outputting to a wav and playing on my PC
[14:31] <desrt> (i get hissy output)
[14:31] <BBB> ah :)
[14:32] <desrt> thanks for the help :)
[14:32] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) left irc: "bbl"
[14:32] <BBB> n
[14:32] <BBB> np
[14:32] Action: BBB gone again
[14:32] <BBB> jdahlin: tell thomasvs I'll be online somewhat today
[14:34] <desrt> hmm.  36 seconds to decode a 5 minute song.
[14:34] <jdahlin> BBB, okay
[14:34] <jdahlin> spider is ugly FYI
[14:35] <desrt> i like spider.  it's nice to me.
[14:35] <desrt> and afaik there exists nothing better
[14:35] jacob__ ([email protected]) left irc: Client Quit
[14:35] <BBB> jdahlin: gah ;)
[14:36] <BBB> desrt: true
[14:36] Company ([email protected]) joined #gstreamer.
[14:36] <BBB> desrt: that's long... I ususally transcode video faster
[14:36] <desrt> hmm
[14:36] <BBB> ohwell
[14:36] <jdahlin> well, spider behaves quite nice, but that doesn't mean the code is nice :-)
[14:36] Action: BBB gone now
[14:36] <BBB> later
[14:36] <desrt> ok.  i think my wav output broke :)
[14:36] <Company> half past 8 and BBB is up?
[14:36] <BBB> yes
[14:36] <BBB> odd uh? :)
[14:37] <Company> yeah
[14:37] <BBB> that's why I'm going back to take a cold shower and go to work after that
[14:37] <BBB> bye ;)
[14:37] BBB ([email protected]) left irc: "Client exiting"
[14:37] <Company> as long ds doesn't show up now
[14:38] <desrt> hm
[14:38] <desrt> wavenc also looks to be broken on ppc
[14:40] <Company> wavenc takes LE only
[14:40] <Company> so you need audioconvert
[14:41] Action: desrt scratches head
[14:41] <Company> but i guess BBB told you that already
[14:41] Action: desrt scrolls
[14:41] <iain> ohh marlin should then add audioconvert before wavenc then...
[14:41] <desrt> no.  don't think he did
[14:41] md` ([email protected]) left irc: Read error: 60 (Operation timed out)
[14:42] <iain> Company: did you see the crash with entry/marlin/wav?
[14:42] md` ([email protected]) joined #gstreamer.
[14:43] <Company> iain: no, what crash?
[14:44] <iain> with entrygthread loading a wav does nothing
[14:44] <iain> then when I stop the pipeline I get
[14:44] <iain> ** ERROR **: file entryscheduler.c: line 890 (gst_entry_scheduler_yield): assertion failed: (ELEMENT_PRIVATE (element))
[14:44] <iain> aborting...
[14:45] <Company> i'll check that when i'm back
[14:45] <Company> i wonder who'd call _yield though in that pipeline
[14:47] <desrt> alas... no help from audioconvert
[14:47] Action: desrt tries something easier
[14:48] <iain> Company: for what its worth, totem/entrygthread crashes with similar error when I play an  asf file
[14:52] <Company> iain: hm, will checkwhen back
[14:54] <iain> ok
[14:58] md` ([email protected]) left irc: Read error: 60 (Operation timed out)
[14:59] ploum ([email protected]) joined #gstreamer.
[15:02] _246tNt_ ([email protected]) joined #gstreamer.
[15:03] <ploum> hello _246tNt_
[15:19] Company ([email protected]) left irc: Remote closed the connection
[15:42] mxpxpod ([email protected]) left irc: "Ex-Chat"
[15:55] smoke ([email protected]) joined #gstreamer.
[16:17] markey ([email protected]) joined #gstreamer.
[16:28] jpetersen ([email protected]) joined #gstreamer.
[16:35] <thomasvs> dolphy: urk / can you connect your home machine again ?
[16:49] md` ([email protected]) joined #gstreamer.
[16:52] spider-man ([email protected]) joined #gstreamer.
[17:02] kwm ([email protected]) joined #gstreamer.
[17:06] markey ([email protected]) left irc: "leaving"
[17:06] teuf ([email protected]) joined #gstreamer.
[17:07] bitshifter ([email protected]) joined #gstreamer.
[17:07] Company ([email protected]) joined #gstreamer.
[17:07] Company ([email protected]) left irc: Remote closed the connection
[17:09] Company ([email protected]) joined #gstreamer.
[17:09] Action: benow yawns
[17:09] trow ([email protected]) joined #gstreamer.
[17:10] jimmy_dean ([email protected]) left irc: Read error: 110 (Connection timed out)
[17:12] alley_cat ([email protected]) left irc: Read error: 110 (Connection timed out)
[17:14] alley_cat ([email protected]) joined #gstreamer.
[17:14] <Misirlou> Company: Are you a debugging master?
[17:18] <jpetersen> is there a specific reason why gstalsamixer uses a [store volume; set volume to 0; set mute flag] algortihm for mute?
[17:19] <Company> Misirlou: i hope so
[17:20] <Misirlou> Company: In GDB, is "backtrace full" more useful than "backtrace", and if so, is that the most verbose that it gets?
[17:20] <Company> Misirlou: i'm not a gdb master :)
[17:20] <Misirlou> You suck :P
[17:21] <teuf> thread apply all backtrace will give you more info in threaded programs
[17:21] <Company> Misirlou: i can _do_ debugging, i don't _know_ it
[17:21] <Company> Misirlou: i thought that makes me a master :(
[17:22] <Misirlou> :(
[17:25] <Company> sink vs src is so confusing
[17:26] <Company> you link a src to a sink
[17:26] <Company> but in elements you transfer data from sinks to srcs
[17:26] <Company> that so confuses me everytime
[17:27] <teuf> that confused me at first, but is actually more or less logical
[17:27] <Company> yes
[17:28] <Company> it gets more confusing if you write autopluggers
[17:29] <Company> there you keep track of paths and whatnot that have a src that's a sinkpad and a sink that's a srcpad
[17:30] <jdahlin> Is there a difference between pad.link(other_pad) and pad.link_filtered(other_pad, NULL) ?
[17:30] <Company> no
[17:31] <Company> that goes for all link functions
[17:31] <jdahlin> ok
[17:31] <jdahlin> (process:29919): GStreamer-WARNING **: internal error: push on pad queue1:sink but it has no chainhandler
[17:31] <jdahlin> I get a couple of these, but I can't really figure out why.
[17:31] <Company> because the sink has no chainhandler most likely ;)
[17:31] <jdahlin> True, but why?
[17:32] <Company> scheduler
[17:32] <jdahlin> how are they usually attached?
[17:32] <Company> opt and basic mess with them pretty much
[17:33] spider-man ([email protected]) left irc: Read error: 110 (Connection timed out)
[17:33] <jdahlin> I bet that's how they're designed
[17:43] <Company> gngngngngngngn
[17:44] <Company> -me puts export GST_SCHEDULER=entrygthread into his .bashrc
[17:48] <jpetersen> http://jpetersen.org/patches/gstreamer-alsamixer-1.patch implements "volume-changed" and "mute-toggled" signal support in gstalsamixer
[17:51] sub_pop ([email protected]) left irc: "Client exiting"
[18:00] walters ([email protected]) joined #gstreamer.
[18:07] sub_pop ([email protected]) joined #gstreamer.
[18:23] markey ([email protected]) joined #gstreamer.
[18:27] teuf ([email protected]) left irc: "Client exiting"
[18:38] Company ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[18:39] Company ([email protected]) joined #gstreamer.
[18:41] jimmy_dean ([email protected]) joined #gstreamer.
[18:42] Pamels ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[18:47] mathrick ([email protected]) joined #gstreamer.
[18:49] <mathrick> ho
[18:53] alley_cat ([email protected]) left irc: Read error: 60 (Operation timed out)
[18:56] bilboed ([email protected]) left irc: Remote closed the connection
[19:04] Uraeus ([email protected]) joined #gstreamer.
[19:04] <Uraeus> hello
[19:07] <mathrick> yo Uraeus
[19:07] <mathrick> how's move going?
[19:07] <Uraeus> fine, although I took a break today (partly due to lack of car :)
[19:08] <mathrick> Uraeus: did you decide where to settle after trip yet?
[19:14] <Uraeus> mathrick: it is not really up to me to make the final decision :) 
[19:16] markey ([email protected]) left irc: Read error: 60 (Operation timed out)
[19:19] <mathrick> Uraeus: so who's in charge? immigration office or something?
[19:19] alley_cat ([email protected]) joined #gstreamer.
[19:22] <Uraeus> mathrick: no, its up julien :)
[19:23] <mathrick> Uraeus: hmm? are you being hired by fluendo?
[19:23] <Uraeus> mathrick: thats the plan :)
[19:23] <mathrick> Uraeus: w00t :)
[19:24] <mathrick> Uraeus: as a hacker ot press person? ;)
[19:24] <mathrick> s/ot/or/
[19:24] <Uraeus> mathrick: as juliens pool cleaner ;)
[19:24] <mathrick> hehe :)
[19:28] harshy ([email protected]) left irc: "I quit for now"
[19:29] harshy ([email protected]) joined #gstreamer.
[19:31] pb_ (~pb@2002:5160:4550:1:240:95ff:fe30:aa84) joined #gstreamer.
[19:31] mxpxpod ([email protected]) joined #gstreamer.
[19:38] Nick change: apoc_ -> apoc
[19:38] <apoc> hi
[19:39] <mathrick> hi apoc
[19:41] <Uraeus> hello apoc
[19:56] Action: Uraeus spent 220 Euros on vacinations today :)
[19:56] <jdahlin> hoho
[19:57] <Uraeus> then again I am now safe for many of the major diseases known to mankind
[19:57] <mathrick> Uraeus: you're never safe, d00d ;P
[19:58] <Uraeus> I can even practice unsafe sex, yet still be sure not to get Hepatites B
[19:58] <mathrick> pfff
[20:05] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[20:09] Uraeus ([email protected]) left irc: "Client exiting"
[20:10] markey ([email protected]) joined #gstreamer.
[20:11] dilinger ([email protected]) left irc: "Terminated with extreme prejudice - dircproxy 1.0.5"
[20:11] dilinger ([email protected]) joined #gstreamer.
[20:15] Shoragan ([email protected]) got netsplit.
[20:16] jdahlin ([email protected]) left irc: "Leaving"
[20:16] Nick change: markey -> markey|afk
[20:20] teuf ([email protected]) joined #gstreamer.
[20:20] Shoragan ([email protected]) returned to #gstreamer.
[20:23] wheels ([email protected]) joined #gstreamer.
[20:26] markey|afk ([email protected]) left irc: Read error: 60 (Operation timed out)
[20:26] ChrisHJW ([email protected]) joined #gstreamer.
[20:45] thomasvs ([email protected]) left irc: Read error: 60 (Operation timed out)
[20:48] kwm ([email protected]) left irc: "brb"
[20:48] Uraeus ([email protected]) joined #gstreamer.
[20:54] kwm ([email protected]) joined #gstreamer.
[21:13] markey ([email protected]) joined #gstreamer.
[21:16] jdahlin ([email protected]) joined #gstreamer.
[21:19] markey ([email protected]) left irc: "leaving"
[21:20] markey ([email protected]) joined #gstreamer.
[21:23] <ds-work> Uraeus: the one you want to worry about is HepC
[21:25] thomasvs ([email protected]) joined #gstreamer.
[21:28] <Uraeus> ds-work: didn't see any vacine for that one
[21:28] <Uraeus> hi thomasvs
[21:29] markey_ ([email protected]) joined #gstreamer.
[21:30] markey ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[21:30] Nick change: markey_ -> markey
[21:41] Nick change: mathrick -> mathrick|food
[21:41] <ds-work> Uraeus: there isn't a vaccine for HepC.  that's why you have to worry about it
[21:43] cschalle__ ([email protected]) joined #gstreamer.
[21:44] <cschalle__> argh, I hate this laptop
[21:44] Uraeus ([email protected]) left irc: Nick collision from services.
[21:45] Nick change: cschalle__ -> Uraeus
[21:48] spyder482 ([email protected]) joined #gstreamer.
[21:48] markey_ ([email protected]) joined #gstreamer.
[21:48] markey ([email protected]) left irc: Connection reset by peer
[21:48] BBB ([email protected]) joined #gstreamer.
[21:50] sublett ([email protected]) joined #gstreamer.
[21:52] <BBB> thomasvs: I noticed the gnome-media work, ty for that!
[21:52] <BBB> I'll do some "relevant" patching work after 2.6.1
[21:52] <BBB> (there's loads of uncommitted patches in bugzilla, esp. for gnome-cd)
[21:54] <BBB> bbl... ;)
[21:54] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[21:54] Nick change: mathrick|food -> mathrick
[21:55] ernstp ([email protected]) joined #gstreamer.
[21:55] <ernstp> I just discovered the wierdest bug on my system.
[21:56] <ernstp> If I have libgstmad installed, xvimagesink doesn't work.
[21:56] <Uraeus> ernstp: that is weird
[21:56] <ernstp> Remove libgstmad, "gst-register", and the xvimagesink shows up in gstreamer-properties again!
[21:56] <ernstp> (then)
[21:57] <ernstp> It could be gcc related, so I'm compiling everything again, but I don't think so...
[21:57] <Uraeus> ernstp: get any error messages when running gst-register with the mad plugin installed
[21:58] markey_ ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[21:58] markey ([email protected]) joined #gstreamer.
[21:58] <ernstp> Uraeus: Nope, and the mad plugin works! Fine! Been playing mp3's in rhythmbox all day.
[21:59] <ernstp> Uraeus: Only disrupts the upper test in gstreamer-properties actually.
[21:59] <Uraeus> ernstp: if you try playing a mpeg with gst-launch-ext with mad plugin installed, what kinda message do you get?
[21:59] <Uraeus> hmm
[22:00] <ernstp> I don't know the whole pipeline. How do I tell which file to play from?
[22:00] <Uraeus> ernstp: just use gst-launch-ext filename and it should play it
[22:02] <ernstp> Uraeus: Sound, nothing else
[22:03] <ernstp> That was mpeg.
[22:03] <Uraeus> ernstp: try this pipeline 'gst-launch filesrc location=filename ! spider ! ffmpegcolorspace ! xvimagesink
[22:03] <ernstp> Tried an avi file, a balck window shoed up and quickly disapeared again.
[22:04] <Uraeus> hmm
[22:05] Zeenix ([email protected]) joined #gstreamer.
[22:05] <ernstp> I just recompiled libid3, libmad and gst-plugins-mad
[22:06] <ernstp> cool, it kinda works..
[22:06] <ernstp> ok, so it was totally gcc related and probably was libmad's fault
[22:07] <ernstp> but what dependecy does the imagesink Test in gstreamer-properties have on libmad?
[22:09] <Company> hrm
[22:09] <Company> GStObjects don't have a name by default
[22:09] <Uraeus> ernstp: maybe some kind of symbol conflict?
[22:09] <ds-work> Company: it's hard to create a GstObject without a name
[22:10] <Zeenix> hello
[22:10] <Company> ds-work: subclass GstElement in your private class and don't create it with gst_element_factory_make
[22:10] markey_ ([email protected]) joined #gstreamer.
[22:11] markey ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[22:15] ernstp ([email protected]) left irc: "Lämnar"
[22:15] tlg ([email protected]) left irc: "Client exiting"
[22:24] markey_ ([email protected]) left irc: Read error: 60 (Operation timed out)
[22:26] steve_b ([email protected]) joined #gstreamer.
[22:28] <Uraeus> ok, time to head home
[22:28] Uraeus ([email protected]) left irc: "Client exiting"
[22:35] Zeenix ([email protected]) left irc: Read error: 60 (Operation timed out)
[22:38] mxpxpod ([email protected]) left irc: "Ex-Chat"
[22:44] ploum ([email protected]) left irc: "Leaving"
[22:54] dolphy ([email protected]) joined #gstreamer.
[22:55] mxpxpod ([email protected]) joined #gstreamer.
[23:09] jcsston ([email protected]) joined #gstreamer.
[23:10] jcsston ([email protected]) left irc: Client Quit
[23:20] thomasvs ([email protected]) left irc: Read error: 60 (Operation timed out)
[23:21] thomasvs ([email protected]) joined #gstreamer.
[23:31] Zeenix ([email protected]) joined #gstreamer.
[23:34] <Zeenix> hi again
[23:41] <ds-work> thomasvs: is the 'sinclude $(srcdir)/../upload.mak' something I should understand?
[23:42] Company ([email protected]) got netsplit.
[23:42] jpetersen ([email protected]) got netsplit.
[23:42] wtay_ ([email protected]) got netsplit.
[23:42] danb ([email protected]) got netsplit.
[23:42] sri ([[email protected]) got netsplit.
[23:42] rener ([email protected]) got netsplit.
[23:42] desrt ([email protected]) got netsplit.
[23:42] sxpert_work ([email protected]) got netsplit.
[23:42] shawarma ([email protected]) got netsplit.
[23:42] <thomasvs> ds-work: the directive, or the other file ?
[23:42] <ds-work> sinclude
[23:43] <thomasvs> the directive just includes the snippet at the make level, but only if it exists
[23:43] <thomasvs> include would always include, and error if missing, and works at the automake level
[23:43] <thomasvs> (this is from memory, I may have forgotten something)
[23:43] <ds-work> in automake-1.8?
[23:43] <thomasvs> the snippet itself is just a bunch of simple upload rules for the docs
[23:43] <thomasvs> ds-work: no, since 1.4 I think
[23:43] Makki_FT ([email protected]) left irc: "Client Exiting"
[23:43] <thomasvs> why, are you getting an error for it ?
[23:44] <ds-work> yes
[23:44] Company ([email protected]) returned to #gstreamer.
[23:44] jpetersen ([email protected]) returned to #gstreamer.
[23:44] desrt ([email protected]) returned to #gstreamer.
[23:44] danb ([email protected]) returned to #gstreamer.
[23:44] rener ([email protected]) returned to #gstreamer.
[23:44] sxpert_work ([email protected]) returned to #gstreamer.
[23:44] wtay_ ([email protected]) returned to #gstreamer.
[23:44] sri ([[email protected]) returned to #gstreamer.
[23:44] shawarma ([email protected]) returned to #gstreamer.
[23:44] danb ([email protected]) left irc: Remote closed the connection
[23:44] danb ([email protected]) joined #gstreamer.
[23:45] <ds-work> any reason why upload.mak should only sometimes exist?
[23:48] <ds-work> I only see documentation for sinclude as being an m4 macro
[23:50] <ds-work> it's actually quite amazing the amount of crap you can put in Makefiles before GNU make will barf
[23:51] <ds-work> http://www.schleef.org/~ds/patch-solaris-gstreamer
[23:51] <ds-work> also someone take a look at pkgconfig/Makefile.am
[23:51] <ds-work> there's a GNU make extension in there
[23:53] <Company> i'm fine with requiring gnu make
[23:53] <Company> we already require gnu autoconf ;)
[23:53] <walters> er...
[23:53] <ds-work> not for compiling tarballs
[23:53] <walters> autoconf shouldn't be needed on end user machines
[23:53] <ds-work> we do require bison and flex, though
[23:54] <Company> i just would require it
[23:54] <Company> when i last tried on solaris, normal make wouldn't compile autotools output anyway
[23:55] <walters> perhaps because of the bug ds found?
[23:55] <Company> no, that was something else
[23:55] <ds-work> worked fine here after I fixed our bugs
[23:55] <Company> but it's been a long time ago
[23:56] <Company> automake 1.6 times or so
[23:56] <ds-work> automake output has always worked with non-GNU makes
[00:00] --- Tue Apr 20 2004


-------------------------------------------------------
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