IRC Logs

*******************************************************************
[03:01] thomasvs ([email protected]) joined #gstreamer.
[03:02] rootcode ([email protected]) left irc: "Leaving"
[03:15] sub_pop ([email protected]) joined #gstreamer.
[03:33] chh ([email protected]) joined #gstreamer.
[04:42] aldug ([email protected]) joined #gstreamer.
[04:43] ChriHJW_log ([email protected]) left irc: Read error: 110 (Connection timed out)
[04:51] harshy ([email protected]) left irc: "I quit for now"
[04:51] harshy ([email protected]) joined #gstreamer.
[04:53] chh ([email protected]) left irc: Read error: 60 (Operation timed out)
[05:34] jcsston ([email protected]) left irc: "VDubMOD watchout TCME is coming!"
[05:34] jcsston ([email protected]) joined #gstreamer.
[05:42] harshy ([email protected]) left irc: "I quit for now"
[05:42] harshy ([email protected]) joined #gstreamer.
[06:02] NFusi0n ([email protected]) joined #gstreamer.
[06:28] ChrisHJW_log ([email protected]) joined #gstreamer.
[06:46] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[06:46] jcsston` ([email protected]) joined #gstreamer.
[06:51] mathrick ([email protected]) joined #gstreamer.
[06:54] <mathrick> yo
[07:03] steveb_ ([email protected]) joined #gstreamer.
[07:05] thaytoo ([email protected]) left irc: "hometime"
[07:21] steveb__ ([email protected]) left irc: Read error: 110 (Connection timed out)
[07:25] bluejay ([email protected]) left irc: "Leaving"
[07:29] harshy ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[07:30] Nick change: mathrick -> mathrick|Uni
[07:39] harshy ([email protected]) joined #gstreamer.
[07:41] walters ([email protected]) joined #gstreamer.
[08:07] thaytan ([email protected]) left irc: Read error: 113 (No route to host)
[08:28] NFusi0n ([email protected]) left irc: Read error: 110 (Connection timed out)
[08:36] jcsston` ([email protected]) left #gstreamer ("VDubMOD watchout TCME is coming!").
[08:47] thaytan ([email protected]) joined #gstreamer.
[09:00] Company ([email protected]) joined #gstreamer.
[09:09] <Company> note to self: xine buffer passing sucks
[09:13] kmaraas ([email protected]) left irc: "Leaving"
[09:41] wheels ([email protected]) left irc: "using sirc version 2.211+KSIRC/1.3.9"
[09:44] Company ([email protected]) left irc: Remote closed the connection
[10:00] KoRnouille ([email protected]) joined #gstreamer.
[10:03] thomasvs ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:03] <KoRnouille> hi all
[10:06] markey ([email protected]) joined #gstreamer.
[10:10] markey ([email protected]) left irc: Client Quit
[10:17] sxpert_work ([email protected]) joined #gstreamer.
[10:17] bilboed_ ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:30] jdahlin ([email protected]) joined #gstreamer.
[10:32] dolphy ([email protected]) joined #gstreamer.
[10:33] <dolphy> morning
[10:34] <KoRnouille> hello
[10:34] <jdahlin> lo
[10:35] <KoRnouille> im trying to get that video wall project to run here...
[10:35] <KoRnouille> but Im not a programmer
[10:35] <KoRnouille> and I don't really know how to compile the live_client.c and live_server.c files
[10:36] <KoRnouille> do I need gnome for it to run ?
[10:36] <KoRnouille> #include <gnome.h>
[10:36] <KoRnouille> I guess I do, huh ?
[10:36] <KoRnouille> gcc live_client.c -o live_client
[10:36] <KoRnouille> is that the correct syntax ?
[10:41] <KoRnouille> I get errors like :
[10:41] <KoRnouille> live_client.c:8:21: gst/gst.h: No such file or directory
[10:41] <KoRnouille> live_client.c:9:19: gnome.h: No such file or directory
[10:41] <KoRnouille> and I am compiling these files under gstreamer-0.7.4/
[10:41] <KoRnouille> so it should find gst/gst.h
[10:42] <KoRnouille> the gnome.h, I dont have it
[10:44] <thaytan> I haven't built the project
[10:44] <thaytan> but you'll need the dev packages for gnome and gstreamer
[10:45] <thaytan> and (if there is no makefile) something like gcc -o live_client `pkgconfig --cflags gstreamer-0.6 gnome-2.0` live_client.c `pkgconfig --libs gstreamer-0.6 gnome-2.0`
[10:46] <KoRnouille> i compiled gstreamer
[10:46] <KoRnouille> didn't install from rpms
[10:46] <thaytan> which version did you install?
[10:46] <KoRnouille> so if i compiled it, does it have the -devel ?
[10:46] <KoRnouille> gstreamer-0.7.4/
[10:47] <thaytan> cool
[10:47] <KoRnouille> true.
[10:47] <thaytan> not sure if the videowall will work - I don't know that anyone has tried it on a 0.7 release
[10:48] <KoRnouille> aaah, really ? big changes ?
[10:49] <thaytan> lots of changes, yes. it will probably still compile, but I don't know if it will work
[10:49] <thaytan> it's worth getting it to though - it's a cool poster project
[10:50] <KoRnouille> Requirements : Atleast two computers having Red Hat Linux( >8.0) and Gstreamer( >0.5.0) installed
[10:51] <KoRnouille> yeah, I'm trying to build a video wall, and be able to play multiple streamer displayed on different TVs
[10:51] <KoRnouille> I just got my self 4 PCI cards
[10:51] <KoRnouille> xinerama works fine
[10:51] <KoRnouille> it's nice
[10:51] <KoRnouille> but now, I need to get that gstreamer to run
[10:52] <KoRnouille> and I'm not very good at it
[10:52] <thaytan> I'm planning to try it myself, for computerbank
[10:52] <thaytan> (local computer charity)
[10:52] <thaytan> KoRnouille: where did you install gstreamer?
[10:53] <KoRnouille> /usr/local/src/gstreamer/gstreamer-0.7.4
[10:53] <KoRnouille> aah
[10:53] <KoRnouille> non
[10:53] <KoRnouille> where
[10:53] <KoRnouille> well
[10:53] <KoRnouille> make ; make install
[10:54] <KoRnouille> /usr/local/bin/gst-launch
[10:54] <KoRnouille> /usr/local/include/gstreamer-0.7
[10:55] <thaytan> also installed gst-plugins?
[10:55] teemu ([email protected]) joined #gstreamer.
[10:56] <KoRnouille> /usr/local/lib/gstreamer-0.7
[10:56] <KoRnouille> /usr/local/share/doc/gstreamer-0.7.4
[10:56] <KoRnouille> btw : man gstreamer
[10:56] <KoRnouille> No manual entry for gstreamer
[10:56] <KoRnouille> yeah, I installed the plugs
[10:56] ChrisHJW_log ([email protected]) left irc: 
[10:57] <thaytan> KoRnouille: you may also need to export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig 
[10:57] <thaytan> I'm not sure where /usr/local is searched by default
[10:57] <KoRnouille> I did that
[10:57] <KoRnouille> well only at  the install
[10:57] <KoRnouille> will I need that everytime I wanna run gstreamer ,
[10:57] <KoRnouille> ?
[10:58] <thaytan> no, just to comile
[10:58] <thaytan> compile 
[10:58] <thaytan> for example, to run the compile line I gave above
[10:58] <thaytan> but substitute gstreamer-0.7 in place of gstreamer-0.6
[11:01] KoRnouille ([email protected]) got netsplit.
[11:02] KoRnouille ([email protected]) returned to #gstreamer.
[11:02] Rotty ([email protected]) joined #gstreamer.
[11:02] <KoRnouille> woah !
[11:02] <KoRnouille> cool
[11:02] <thaytan> wassup?
[11:05] sublett ([email protected]) joined #gstreamer.
[11:06] <KoRnouille> so
[11:06] <KoRnouille> why wouldn't it compile ?
[11:07] <KoRnouille> live_server only has this include : #include <gst/gst.h>
[11:07] <KoRnouille> gcc live_server.c -o live_server
[11:07] <KoRnouille> live_server.c:8:21: gst/gst.h: No such file or directory
[11:07] <thaytan> it needs the pkgconfig bits to get the right include flags
[11:07] <KoRnouille> that export stuff ?
[11:07] <thaytan> no, earlier
[11:08] <thaytan> gcc -o live_client `pkgconfig --cflags gstreamer-0.7 gnome-2.0` live_client.c `pkgconfig --libs gstreamer-0.7 gnome-2.0`
[11:08] <thaytan> something like that (I'm guessing)
[11:13] <KoRnouille> what is pkgconfig used for ?
[11:13] <KoRnouille> i don't understand really
[11:13] <thaytan> it supplies compile flags for various gnome packages
[11:14] <thaytan> it's used throughout the gnome stack
[11:14] <thaytan> brb
[11:14] thaytan ([email protected]) left irc: "See y'all"
[11:23] thaytan ([email protected]) joined #gstreamer.
[11:24] pb_ ([email protected]) joined #gstreamer.
[11:28] Shoragan ([email protected]) joined #gstreamer.
[11:31] ensonic ([email protected]) joined #gstreamer.
[11:31] <ensonic> hi ho
[11:32] <thaytan> hi ho :)
[11:33] markey ([email protected]) joined #gstreamer.
[11:52] thomasvs ([email protected]) joined #gstreamer.
[12:03] markey ([email protected]) left irc: "bbl"
[12:13] <ensonic> ds: can you please have a look at bug_id+133528 again
[12:16] <ensonic> anyone knows how to get the buffer_size and the sampling_rate used by an audio_sink (e.g. from esdsink) to compute the resolution of synchronous dparam changes
[12:18] <ensonic> 2nd question, why have audio+pads not just a depth (bit resolution), but a width as well (and what does the width refers to)
[12:26] <thomasvs> ensonic: it's possible to have audio with width 16 and depth 12
[12:26] <thomasvs> which means, 12 bit audio transported in 16 bits
[12:28] <sxpert_work> 12 bit audio would be from 4 channel DV recording
[12:28] <sxpert_work> (for instance)
[12:49] <ensonic> thomasvs: & sxpert_work: ahh understoof
[12:49] <ensonic> ...ood
[12:49] pb_ ([email protected]) left irc: Remote closed the connection
[12:50] pb_ ([email protected]) joined #gstreamer.
[12:54] <thomasvs> ensonic: sounded like someone punched you in the stomach while typing
[13:00] <KoRnouille> hello thomasvs
[13:00] <KoRnouille> :)
[13:02] <thomasvs> hi
[13:26] <ensonic> ds: are you awake?
[13:47] smoke ([email protected]) joined #gstreamer.
[13:47] sublett ([email protected]) left irc: Read error: 113 (No route to host)
[13:58] Rotty ([email protected]) left irc: Remote closed the connection
[14:19] thaytan ([email protected]) left irc: Read error: 60 (Operation timed out)
[14:26] thaytan ([email protected]) joined #gstreamer.
[14:35] Rotty ([email protected]) joined #gstreamer.
[14:47] Shoragan ([email protected]) left irc: Read error: 110 (Connection timed out)
[14:57] Shoragan ([email protected]) joined #gstreamer.
[15:09] foser ([email protected]) joined #gstreamer.
[15:21] <KoRnouille> that's the command for the man pages of gstreamer ? :/
[15:24] ibsteveog ([email protected]) joined #gstreamer.
[15:27] Nick change: mathrick|Uni -> mathrick
[15:27] <mathrick> re
[15:29] sublett ([email protected]) joined #gstreamer.
[15:34] <shawarma> re
[15:36] <KoRnouille> where are the man pages for gstreamer please ?
[15:37] <shawarma> mathrick: Yeah, I figured. :-(  The gnomevfs plugin system appears to be better documented.. That's why I was thinking about going that way.
[15:37] <shawarma> Whoops.
[15:44] iain_ ([email protected]) joined #gstreamer.
[15:47] <thomasvs> KoRnouille: if you installed gst, just do man gst-launch-0.7 for example
[15:56] markey ([email protected]) joined #gstreamer.
[16:02] trow ([email protected]) joined #gstreamer.
[16:02] sublett ([email protected]) left irc: Read error: 113 (No route to host)
[16:03] markey ([email protected]) left irc: "leaving"
[16:16] <KoRnouille> ok thomasvs. I did try "man gst-launch" only
[16:20] <KoRnouille> I have an error
[16:20] <KoRnouille> gst seems to crash
[16:20] <thomasvs> ok, spill
[16:20] <KoRnouille> gst-launch filesrc location=/root/info.mpg ! mpegdemux ! mpeg2dec ! xvideosink
[16:20] <KoRnouille> Caught SIGSEGV accessing address (nil)
[16:20] <KoRnouille> #0  0xffffe002 in ?? ()
[16:20] <KoRnouille> #1  0x402fd759 in ?? ()
[16:20] <KoRnouille> #2  0x40290342 in ?? ()
[16:20] <KoRnouille> #3  0x08049ac3 in ?? ()
[16:20] <KoRnouille> #4  <signal handler called>
[16:20] <KoRnouille> #5  0x4006866b in ?? ()
[16:20] <KoRnouille> #6  0x40068c54 in ?? ()
[16:20] <KoRnouille> #7  0x400634ce in ?? ()
[16:20] <KoRnouille> #8  0x4006341b in ?? ()
[16:20] <KoRnouille> #9  0x0804a50c in ?? ()
[16:20] <KoRnouille> #10 0x42015574 in ?? ()
[16:20] <KoRnouille> Caught interrupt.
[16:20] <KoRnouille> Spinning.  Please run 'gdb gst-launch 4323' to continue debugging, Ctrl-C to quit, or Ctrl-\ to dump core.
[16:20] <KoRnouille> Caught interrupt.
[16:21] <KoRnouille> is ti bad doc ?
[16:21] <KoRnouille> is it bad doc ?
[16:25] <KoRnouille> thomasvs ?
[16:26] <KoRnouille> gst-launch fakesrc ! fakesink
[16:26] <KoRnouille> I get the same kind of error
[16:26] <KoRnouille> (not exactly the same though)
[16:27] Sphair ([email protected]) joined #gstreamer.
[16:27] <thomasvs> ok, start with the simple one
[16:27] <thomasvs> what do you get on fakesrc ! fakesink ?
[16:28] <KoRnouille> gst-launch fakesrc ! fakesink
[16:28] <KoRnouille> Caught SIGSEGV accessing address (nil)
[16:28] <KoRnouille> #0  0xffffe002 in ?? ()
[16:28] <KoRnouille> #1  0x402fd759 in ?? ()
[16:28] <KoRnouille> #2  0x40290342 in ?? ()
[16:28] <KoRnouille> #3  0x08049ac3 in ?? ()
[16:28] <KoRnouille> #4  <signal handler called>
[16:28] <KoRnouille> #5  0x4006866b in ?? ()
[16:28] <KoRnouille> #6  0x40068c54 in ?? ()
[16:28] <KoRnouille> #7  0x400634ce in ?? ()
[16:28] <KoRnouille> #8  0x4006341b in ?? ()
[16:28] <KoRnouille> #9  0x0804a50c in ?? ()
[16:28] <KoRnouille> #10 0x42015574 in ?? ()
[16:28] <KoRnouille> Spinning.  Please run 'gdb gst-launch 4345' to continue debugging, Ctrl-C to quit, or Ctrl-\ to dump core.
[16:28] <KoRnouille> there, I press CTRL + C
[16:28] <KoRnouille> Caught interrupt.
[16:28] <Sphair> hermes is dead, we need new maintainers, anyone?
[16:30] <KoRnouille> if you need me for apt-get under RH 9, let me know... i'm still here... I will be much easier for me to compile and use gstreamer
[16:31] Marsupilami23 ([email protected]) joined #gstreamer.
[16:34] Action: thomasvs gets really confused at the way gstreamer handles colorspaces
[16:36] Shoragan ([email protected]) left irc: "Leaving"
[16:36] Marsupilami23 ([email protected]) left irc: "I don't wanna grow up, I'm a Toys R Us Kid. There's a million toys at Toys R Us that I can play with!"
[16:38] <KoRnouille> maybe I should try withg that GStreamer 0.6.5 ?
[16:38] sxpert_work ([email protected]) got netsplit.
[16:38] Sphair ([email protected]) got netsplit.
[16:38] foser ([email protected]) got netsplit.
[16:38] Jara[zZ] ([email protected]) got netsplit.
[16:38] ds-work ([email protected]) got netsplit.
[16:38] herzi ([email protected]) got netsplit.
[16:38] dilinger ([email protected]) got netsplit.
[16:38] foser ([email protected]) returned to #gstreamer.
[16:39] Jara[zZ] ([email protected]) returned to #gstreamer.
[16:39] ds-work ([email protected]) returned to #gstreamer.
[16:39] herzi ([email protected]) returned to #gstreamer.
[16:39] dilinger ([email protected]) returned to #gstreamer.
[16:39] sxpert_work ([email protected]) returned to #gstreamer.
[16:39] <mathrick> KoRnouille: porting it to 0.7 would be _much_ better
[16:39] <KoRnouille> ok, but I get that error
[16:39] <mathrick> KoRnouille: even tho 0.6 is certainly easier to get going
[16:41] Marsupilami23 ([email protected]) joined #gstreamer.
[16:41] <KoRnouille> hummm
[16:42] <mathrick> KoRnouille: for starters, replace xvideosink with xvimagesink and try again
[16:42] <KoRnouille> what ever I do, I get that error stuff
[16:43] <mathrick> ah, didn't notice that
[16:43] Action: mathrick needs to rebuild recent CVS, my current one is shamefully out of date
[16:44] Sphair ([email protected]) got lost in the net-split.
[16:48] <KoRnouille> i got the 0.7.4 in tar.gz
[16:49] Marsupilami23 ([email protected]) left irc: "I don't wanna grow up, I'm a Toys R Us Kid. There's a million toys at Toys R Us that I can play with!"
[16:55] <thomasvs> KoRnouille: did you compile gst yourself, to get those backtraces ?
[16:58] trow ([email protected]) left irc: "Leaving"
[16:58] kwm ([email protected]) joined #gstreamer.
[17:02] markey ([email protected]) joined #gstreamer.
[17:28] markey ([email protected]) left irc: "leaving"
[17:31] Marsupilami23 ([email protected]) joined #gstreamer.
[17:33] Uraeus ([email protected]) joined #gstreamer.
[17:33] <mathrick> lo Uraeus
[17:33] <Uraeus> ello
[17:34] <jdahlin> hello norweigian boy
[17:35] walters ([email protected]) left irc: Client Quit
[17:35] <Uraeus> norwegian == tall, strong and handsome
[17:37] yippi ([email protected]) joined #gstreamer.
[17:37] kmaraas ([email protected]) joined #gstreamer.
[17:49] foser ([email protected]) left irc: "[ I want to believe ]"
[17:52] <thomasvs> Uraeus: is Dom on irc ?
[17:52] <thomasvs> Uraeus: I have an endless loop in librsvg
[17:53] <thomasvs> Uraeus: and I would like to discuss it with him
[18:06] jcsston ([email protected]) joined #gstreamer.
[18:08] Nick change: harshy -> harshyWork
[18:13] <Uraeus> thomasvs: yes, in #librsvg on irc.gnome.org
[18:14] vegan_linuxguy ([email protected]) joined #gstreamer.
[18:15] <vegan_linuxguy> Is gstreamer capable of doing dvd to divx transcoding?  Would it be possible/smart to make an app doing this?  Or should I make a frontend to mencoder or transcode?
[18:19] <Uraeus> vegan_linuxguy: gstreamer should be be capable of doing that
[18:20] <Uraeus> vegan_linuxguy: making an application for it would also be a good idea as using gst-launch pipelines is not that userfriendly :)
[18:22] <vegan_linuxguy> Oh, of course.  I'm planning on making an application that is really simple to do this.  I'd like to generate thumbnails of all the titles on the DVD, etc, if at all possible.  I'm modeling it loosely off of soundjuicer.
[18:23] <Uraeus> vegan_linuxguy: well both our DVD playback and our divc encoder is something which has seen a lot of love lately so I think they should be up to this task
[18:23] ChrisHJW ([email protected]) joined #gstreamer.
[18:25] <vegan_linuxguy> Wonderful.  I just thought it seemed like a relatively easy hole to fill in the GNOME desktop, so I thought I'd give it a shot.  Thanks for the info.
[18:25] <Uraeus> np
[18:26] foser ([email protected]) joined #gstreamer.
[18:31] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[18:32] <KoRnouille> <thomasvs> KoRnouille: did you compile gst yourself, to get those backtraces ?
[18:32] <KoRnouille> yes :)
[18:34] Company ([email protected]) joined #gstreamer.
[18:34] <Uraeus> hi foser and Company
[18:35] <Company> ho
[18:35] dolphy ([email protected]) joined #gstreamer.
[18:35] thaytan ([email protected]) left irc: Ping timeout: 14400 seconds
[18:36] <mathrick> yo Company, foser
[18:36] vegan_linuxguy ([email protected]) left #gstreamer.
[18:39] <ensonic> ds-work: (2nd try) can you please have a look at bug_id=133528 again
[18:40] <ensonic> Company: can you please reply to my mail on the ML about the GstClock implementations
[18:42] kmaraas ([email protected]) left irc: "Leaving"
[18:47] <Company> the only thing i know about clocking, clocks and everything related is that it's extremely brittle
[18:47] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[18:49] <Company> i haven't even looked at how/if that part works
[18:52] <Company> which is exactly why i haven't answered to the mail btw
[18:52] markey ([email protected]) joined #gstreamer.
[18:53] Nick change: Uraeus -> Ura_gym
[18:56] <thomasvs> http://thomas.apestaart.org/download/screenshots/gst-player-error-spider.png
[18:56] <thomasvs> Company: the funny thing is, after I click OK on that error, then press play again, it thinks it's theora :)
[18:56] <ensonic> Company: I had the feeling, that is the part you are doing/designing/working-at?
[18:57] <thomasvs> Company: actually, it does that again and again, no matter what the original type
[19:00] <ensonic> ok, cu
[19:00] ensonic ([email protected]) left #gstreamer.
[19:03] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[19:09] vegan_linuxguy ([email protected]) joined #gstreamer.
[19:09] <vegan_linuxguy> What kind of sink would a visualization like goom hook up to?  I can't get it to connect to any video sinks.
[19:11] <mathrick> vegan_linuxguy: you need ffcolorspace in between probably
[19:11] <vegan_linuxguy> Thanks.
[19:11] <mathrick> vegan_linuxguy: what pipeline do you use?
[19:12] <vegan_linuxguy> just osssrc ! goom ! xvideosink
[19:12] <vegan_linuxguy> I just started playing with gstreamer a few minutes ago.
[19:12] <mathrick> vegan_linuxguy: don't use xvideosink (if using 0.7)
[19:13] <mathrick> vegan_linuxguy: it's deprecated in favor of x[v]imagesink
[19:13] <vegan_linuxguy> OK, thanks again.
[19:13] <mathrick> thomasvs: Making all in po
[19:13] <mathrick> make[2]: Entering directory `/home/mathrick/Dev/gstreamer/gstreamer/po'
[19:13] <mathrick> rm -f nl.gmo && /usr/bin/msgfmt -c --statistics -o nl.gmo nl.po
[19:13] <mathrick> nl.po:555:2: parse error
[19:13] <mathrick> nl.po:555: keyword "nl" unknown
[19:13] <mathrick> /usr/bin/msgfmt: found 2 fatal errors
[19:13] <mathrick> thomasvs: checkout from 2 hours ago
[19:17] hallibaby ([email protected]) joined #gstreamer.
[19:17] hallibaby ([email protected]) left irc: Client Quit
[19:18] jimmy_dean ([email protected]) got netsplit.
[19:18] jimmy_dean ([email protected]) returned to #gstreamer.
[19:19] <mathrick> wb jimmy_dean :)
[19:20] <KoRnouille> gstreamer needs gnome to run ?
[19:20] <mathrick> KoRnouille: no
[19:20] <KoRnouille> it runs only under X ?
[19:20] <KoRnouille> ok
[19:20] <KoRnouille> no
[19:20] <mathrick> KoRnouille: no
[19:20] <KoRnouille> I know it doesn't
[19:20] <KoRnouille> i didn't mean that
[19:21] <KoRnouille> could someone compile the program live_client.c and live_server.c for me please ?
[19:21] <KoRnouille> I can't get it to compile
[19:22] <mathrick> jimmy_dean: btw, I've just started attending computer networks lecture this semester, looks like I'll later need some help with long line transmission, signal spectral analysis and such :)
[19:24] <thomasvs> mathrick: delete the po file you have a conflict
[19:24] <mathrick> thomasvs: k
[19:26] vegan_linuxguy ([email protected]) left #gstreamer.
[19:37] pb_ ([email protected]) left irc: "bbl"
[19:38] Action: KoRnouille is compiling gstreamer0.6.5
[19:40] <KoRnouille> Making install in docs
[19:40] <KoRnouille> mails
[19:40] <KoRnouille> fails
[19:40] <KoRnouille> :/
[19:40] <KoRnouille> mkdir -p /usr/local/share/doc/gstreamer-0.6.5/faq
[19:40] <KoRnouille> cp -pr  /usr/local/share/doc/gstreamer-0.6.5/faq
[19:40] <KoRnouille> cp: missing destination file
[19:40] <KoRnouille> Try `cp --help' for more information.
[19:41] <KoRnouille> was this the last ting to be installed ?
[19:41] <thomasvs> yeah, pretty much
[19:41] <KoRnouille> so i don't care i guess
[19:43] <KoRnouille> all ask my questions here
[19:43] <KoRnouille> instead of RTFM
[19:43] <KoRnouille> :)
[19:43] <KoRnouille> so gstreamer-0.6.5.tar.gz is installed
[19:43] <KoRnouille> ./configure ; make ; make install
[19:43] <KoRnouille> seemed to work fine
[19:43] <KoRnouille> what version of gst-plugins should I use ,
[19:43] <KoRnouille> ?
[19:44] <mathrick> KoRnouille: 0.6.5 of course
[19:44] <thomasvs> 0.6.4, the latest 0.6.x one
[19:44] <thomasvs> mathrick: there is no such thing
[19:44] <taaz> neat.  build page full of red ;)
[19:44] <mathrick> thomasvs: ahh, ok :)
[19:44] <thomasvs> yeah
[19:44] <thomasvs> conflicts on po files
[19:45] <taaz> could probably use dependencies in there to avoid tests and install if compile fails
[19:45] <KoRnouille> I can't find the plugin
[19:45] <KoRnouille> where can I get gst-plugins-0.6.4.tar.gz ? it's not in the download directory
[19:46] <taaz> and success of core should trigger plugins/apps to build
[19:46] <thomasvs> taaz: yeah, but not that easy
[19:46] <KoRnouille> im sorry
[19:46] <thomasvs> taaz: we're slowly working on it
[19:46] <KoRnouille> ../
[19:46] <thomasvs> KoRnouille: you sure ?
[19:46] <KoRnouille> no
[19:46] <KoRnouille> it's okay
[19:46] <thomasvs> ok
[19:46] <KoRnouille> it's in ../
[19:46] Action: thomasvs goes home
[19:46] <taaz> thomasvs: yeah, i'm just thinking out loud, nevermind me ;)  it's pretty cool right now too ;)
[19:46] <KoRnouille> bye thomasvs
[19:46] <thomasvs> taaz: well, it's python, so you can join in on the hacking fun
[19:47] <taaz> i just might ;)
[19:47] <taaz> xp ideas like continuous testing make coding more fun
[19:56] <ds-work> moo
[19:59] jdahlin ([email protected]) left irc: "Klienten avslutas"
[20:00] pb_ (~pb@2002:5160:45ef:0:240:95ff:fe30:aa84) joined #gstreamer.
[20:05] thomasvs ([email protected]) left irc: Read error: 113 (No route to host)
[20:23] thomasvs ([email protected]) joined #gstreamer.
[20:33] doml ([email protected]) joined #gstreamer.
[20:34] <doml> thomasvs: there's not much you can do about already-installed copies of librsvg
[20:35] <thomasvs> doml: hm, I figured as much
[20:35] <thomasvs> doml: I was toying with setting up an alarm before calling the close
[20:35] <doml> thomasvs: my personal hope is that people upgrade
[20:35] <doml> though i find it odd - i've never encountered the bug that you're seeing
[20:36] <doml> and i've opened lots of svgz files
[20:36] <thomasvs> doml: yeah, it is odd
[20:36] <thomasvs> doml: have you opened a normal tarball ? :=)
[20:36] <thomasvs> doml: I don't doubt svgz files work fine
[20:36] <doml> heh
[20:36] <thomasvs> I'm not sure though if any random file, compressed with gzip, will work well though :)
[20:36] <thomasvs> but, I was trying to reproduce it with the same tarball and another application, but I didn't know what to try :)
[20:37] <thomasvs> I have no idea how pixbuf loaders, librsvg, or libgsf work :)
[20:37] <doml> just out of curiousity, why is librsvg getting passed a tarball?
[20:37] <thomasvs> doml: here's how it works
[20:37] <thomasvs> * player gets a file
[20:37] <thomasvs> * player tries to figure out the type
[20:37] <thomasvs> * gstreamer calls all of the typefind functions it knows about
[20:37] <thomasvs> * one of them is gdk_pixbuf
[20:37] <thomasvs> * gdk_pixbuf hands the tar.gz to all of its loaders
[20:38] <thomasvs> * one of them is the rsvgz one
[20:38] <thomasvs> * boom
[20:38] <thomasvs> I was working on making the player/gstreamer robust for files that aren't perfect
[20:38] <doml> barf
[20:38] wheels ([email protected]) joined #gstreamer.
[20:38] <doml> i'm more used to the:
[20:38] <doml> nautilus asks gnome-vfs what to do
[20:38] <doml> gnome-vfs tells nautilus
[20:38] <doml> everyone's happy
[20:38] <doml> ...
[20:38] <thomasvs> heh
[20:38] <thomasvs> well, I had never heard of or seen libgsf before
[20:39] <thomasvs> my first thought was - why is this not gnomevfs ?
[20:39] <thomasvs> I'm sure I dind't really understand it too well
[20:39] <thomasvs> btw, would be nice if the docs build got upgraded so the API docs were visible in devhelp
[20:39] <doml> i could tell you what GSF is and why jody and i wrote it
[20:39] <doml> but that might bore you to death
[20:39] <thomasvs> hehe :)
[20:39] <thomasvs> I don't bore easily
[20:40] <thomasvs> but I don't want to waste your time either
[20:40] <doml> i have plenty of time today if you're interested
[20:40] <thomasvs> I'll ask the second time I'm having a gsf/rsvg-related problem that I cant't figure out
[20:40] Action: mathrick will hear it too :)
[20:40] <thomasvs> because right now I'm going to go skating
[20:40] <doml> enjoy yourself
[20:40] <thomasvs> well, if you can give me the five minute condensed version, I can still catch it over dinner :)
[20:40] <doml> i'll commit your patch tonight. gnome 2.5.4 will have it
[20:40] <doml> sure
[20:41] <doml> libgsf serves 2 main purposes:
[20:41] <doml> 1) make anything look like a "FILE"
[20:41] <doml> 2) serve as a compound document abstraction (ZIP, TAR.GZ, MS OLE2)
[20:41] <doml> you want writing to a file to be the same as writing to memory. you want that to be the same as writing to a remote file. or a COM IStream
[20:42] <thomasvs> ah, to figure out files within files ?
[20:42] <doml> gnomevfs is good, but it stops short of what "Office Applications" need
[20:42] <doml> yes. The OpenOffice and MSOffice file formats have multiple files & directories within them
[20:43] <doml> as a side note, you want 0 overhead for reading an XML.GZ file. or to write an XML.BZ2 file
[20:43] <doml> it should all just be "transparent"
[20:44] <taaz> i've got some svg that makes, i guess, the gdk loaders go nuts
[20:44] <taaz> by way of gqview trying to thumbnail
[20:44] <doml> nah... librsvg never goes nuts ;-)
[20:44] <mathrick> doml: offtop slightly, but does it relate to Nautilus and transparent archives handling?
[20:44] <doml> mathrick: jody and i hope that pieces of nautilus will be rewritten to use this
[20:45] <mathrick> doml: ahh, good, it was about only good thing in gmc :>
[20:45] <doml> thomasvs: and gdk-pixbuf loaders are *crack*
[20:45] <taaz> err... maybe it's .swf files... i think the svg bugs have gone away.  by now...
[20:45] <doml> ah... my old swfdec loader. i thought that i had fixed the bugs in that
[20:46] Action: mathrick notes gdk-pixbuf loaders just *love* to barf on huge images
[20:46] <doml> if used properly, it shouldn't be hanging on a large SWF file
[20:47] <mathrick> no, i don't mean huge SWF, i mean just any huge (think 4000x10000) image
[20:47] <doml> yeah, that would probably fail
[20:47] <doml> that would allocate at least (4000 x 10000 x 8 x 4) bytes
[20:47] <mathrick> like, evo actually even hangs with not so huge files (something like 4k x 2k) only
[20:48] <mathrick> s/only/only)/
[20:48] Zeenix ([email protected]) joined #gstreamer.
[20:48] <mathrick> lo Zeenix
[20:48] <mathrick> i'd like loaders did tiling, or at least some kind of bullet proofing against huge files
[20:49] <mathrick> it's annoying to have gThumbnail unable to load really big images
[20:49] <doml> indeed
[20:50] <mathrick> and solution like implementing different image loaders for image browser kinda invalidates gdk-pixbuf loaders in the first place
[20:50] <Zeenix> hi
[20:50] <Zeenix> Ura_gym: good that you joined the gym :)
[20:53] sxpert_ ([email protected]) joined #gstreamer.
[20:56] wheels ([email protected]) left irc: Read error: 60 (Operation timed out)
[21:01] wheels ([email protected]) joined #gstreamer.
[21:01] mathrick ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[21:02] mathrick_ ([email protected]) joined #gstreamer.
[21:02] Nick change: mathrick_ -> mathrick
[21:04] <mathrick> thomasvs: care to set a topic? it looks so blank without one :)
[21:08] sxpert ([email protected]) left irc: Read error: 110 (Connection timed out)
[21:09] somex1 ([email protected]) joined #gstreamer.
[21:12] #gstreamer: mode change '+o thomasvs' by ChanServ!ChanServ@services.
[21:12] Topic changed on #gstreamer by [email protected]: I feel so naked without topic
[21:12] #gstreamer: mode change '-o thomasvs' by [email protected]
[21:13] <thomasvs> is it me, or did we just hit 50 nicks in this channel ???
[21:13] Nick change: Ura_gym -> Uraeus
[21:13] <Uraeus> thomasvs: no, there is no one named Nick here
[21:13] <Company> it's you - without you it'd be 49
[21:14] <thomasvs> Company: hehe
[21:14] nated ([email protected]) joined #gstreamer.
[21:15] <Uraeus> nated: : you met Nat Friedman and got Nated?
[21:15] <thomasvs> Uraeus: is it norwegian comedian night or something ?
[21:15] <nated> can anyone help me debug a problem with my alsasink? gst-launch sinesrc ! alsasink, doesn't work, though alsa does in general..
[21:15] <nated> good one Uraeus
[21:16] <mathrick> thomasvs: ok smartass, now you can change it to "Wee, with 50 nicks we need no topic" :>
[21:16] <Uraeus> thomasvs: yup, exhaustion makes my quality control even lower than ususal (yes, I know that is almost impossible, but still :)
[21:16] <ds-work> now we have 51
[21:16] <mathrick> nated: alsasink device="hw:0" ?
[21:16] <thomasvs> somebody set us up a topic
[21:16] <Company> thomasvs: can we require bison >= 1.35 somehow?
[21:17] <mathrick> thomasvs: hehe
[21:17] <Company> /topic the bomb
[21:17] <nated> mathrick, same error "WARNING **: error setting period count minimum (2): Invalid argument"
[21:17] <thomasvs> Company: do you mean, technically, or whether it's a good idea to do so ?
[21:17] <nated> pipeline doesn't want to play
[21:18] <Company> thomasvs: technically - bison 1.28 doesn't compile
[21:18] <Uraeus> thomasvs: think we need sure that Wim only has ALSA installed on his laptop :)
[21:18] <Company> thomasvs: see #127838 and #131452
[21:18] <mathrick> nated: urk, no idea what to do with that, probably it needs some other obscure param tweaked
[21:19] <Company> nated: GStreamer 0.6.x ?
[21:19] <Zeenix> Uraeus: got my email?
[21:19] <ds-work> Company: bug #71615 has a patch for glib disabling global binding
[21:20] <nated> Company, yes...0.6.4
[21:20] <ds-work> Company: no clue why it was never applied
[21:20] <Zeenix> Uraeus: at [email protected]?
[21:20] <nated> running debian..
[21:20] <nated> so i haven't configured much alsa/gstreamer wise...but using mplayer/xmms alsa does work
[21:21] <Company> nated: you can consider alsa broken then - it's working in 0.7
[21:21] <Uraeus> Zeenix: yup
[21:21] <Company> ds-work: probably because there are issues - xine uses RTLD_GLOBAL, too
[21:23] <Company> ds-work: i'd be happy if we implemented the libtool thingy
[21:24] <Company> would get around needing to fix goom
[21:24] <Company> dunno about ffmpeg
[21:24] <ds-work> ffmpeg would be ok, too
[21:24] <Company> it's a noble quest to get 2 media frameworks working in sync anyway ;)
[21:25] <ds-work> what version of libtool do we need to handle it correctly?
[21:25] <Company> ramon said 1.5.3
[21:25] <Company> i say -ENOCLUE
[21:25] dolphy ([email protected]) joined #gstreamer.
[21:26] <dolphy> re
[21:26] <Company> and it'd still not be "correctly" anyway as there might be systems that just don't support it
[21:28] <thomasvs> Company: if you want to add a version check for bison, go ahead
[21:29] <Company> is there an autoconf macro that can do that?
[21:30] <nated> does this line look odd? "Preparing channel: (null) 44100Hz, 2 channels" (the null part)
[21:31] <thomasvs> Company: no, it's "run -v and parse version"
[21:32] <Company> :(
[21:39] <Company> there should be a AC_CHECK_PROG_VERSION somewhere
[21:40] <mathrick> nated: is it error message? if not, I'd say don't bother, (null) most probably means just "some property wasn't explicitly set"
[21:40] <Company> or it means "no format negotiated" which woulkd be bad ;p
[21:41] <nated> mathrick, just part of the output from 'gst-launch sinesrc ! alsasink'
[21:41] <mathrick> nated: ok, but does it work otherwise?
[21:41] <nated> but maybe i'll just take Company's advice, and upgrade to 0.7.x
[21:41] <nated> sure, alsa works, just not with gstreamer
[21:42] <mathrick> nated: argh, you don't get me, anyway just upgrade :)
[21:45] Marsupilami23 ([email protected]) left irc: Remote closed the connection
[21:51] alley_cat ([email protected]) left irc: Read error: 110 (Connection timed out)
[21:51] <dolphy> ds, ds-work: ping ?
[21:54] alley_cat ([email protected]) joined #gstreamer.
[21:56] <mathrick> dolphy: some usability notes about gst-player
[21:56] <mathrick> dolphy: first, there is no way really to restart media, you can only drag slider to beginning
[21:56] Action: thomasvs sees /dev/null being readied
[21:56] <mathrick> thomasvs: wow, by whom?
[21:56] Nick change: sxpert_ -> sxpert
[21:57] <thomasvs> mathrick: you'll find out soon enough :)
[21:58] <Company> mathrick: you're supposed to restart media by dragging the slider (see Rhythmbox and gnome usability list from years ago)
[21:58] <Company> the discussions about the stop button
[21:58] <mathrick> Company: I HATE that way
[21:58] apoc__ ([email protected]) joined #gstreamer.
[21:58] kmaraas ([email protected]) joined #gstreamer.
[21:58] <mathrick> Company: it's the stupidest way to make applications intuitive I've ever seen
[21:59] <mathrick> dolphy: second, after selecting file from recent list in File menu, it gets added to playlist, instead of being opened, this is confusing
[21:59] <mathrick> Company: besides, it's insufficient
[21:59] <Company> yeah, but that was the result of that discussion - i don't care very much anymore
[22:00] <dolphy> mathrick: well i m focusing my attention on totem atm
[22:00] <mathrick> Company: sometimes you just can't achieve what you want w/o hella lot of clicking
[22:00] <mathrick> dolphy: ah, ok, don't bother then
[22:00] <dolphy> mathrick: i would really prefer that you send feedback on gst-totem please
[22:00] <dolphy> mathrick: gst-player will never make it in the desktop
[22:00] <dolphy> mathrick: gst-totem might
[22:00] <mathrick> dolphy: haven't tried it yet
[22:01] <Company> mathrick: i so don't care about that, but i know it's been a usability discussion, so fight with them ;)
[22:02] <mathrick> Company: I do, but it's true I should do that probably on RB-list, instead of #rhythmbox, the only problem being I'm not subscribed to former :)
[22:03] <dolphy> mathrick: well simply get totem's CVS HEAD
[22:03] <dolphy> mathrick: and ./configure --enable-gstreamer
[22:03] <dolphy> mathrick: it's that simple
[22:03] <mathrick> dolphy: hehe :)
[22:03] <mathrick> dolphy: does it require gnome HEAD?
[22:03] <dolphy> mathrick: nope
[22:03] <mathrick> dolphy: ah, then ok, I'll try
[22:03] <dolphy> mathrick: i m running it on Fedora core 1
[22:04] jaiserca ([email protected]) joined #gstreamer.
[22:04] <mathrick> I wonder if 0.99.9 made it to Sid yet 
[22:04] <mathrick> hmm, irrelevant anyway :)
[22:05] <mathrick> thomasvs: look, 54! :)
[22:05] Action: mathrick wonders when will next round number be hit
[22:05] <mathrick> like 64 ;)
[22:06] <dolphy> mathrick: running sid ?
[22:06] <mathrick> dolphy: yep
[22:06] <dolphy> mathrick: seb128 will do a totem with gstreamer package soon
[22:06] <dolphy> mathrick: i told him i ll handle support
[22:06] <mathrick> dolphy: that's nice
[22:07] <mathrick> dolphy: but I'll use CVS anyway, I think you prefer feedback for recent code, don't you ;)
[22:09] doml ([email protected]) left irc: "Client exiting"
[22:09] <dolphy> mathrick: exactly :)
[22:12] <mathrick> dolphy: what is cvsroot for totem? I'm too lazy ;)
[22:15] apoc_ ([email protected]) left irc: Connection timed out
[22:15] <taaz> mathrick: 0.99.9 totem was in sid 4 or 5 days ago
[22:16] <mathrick> taaz: k, thanks
[22:21] <dolphy> mathrick: you have a gnome account ?
[22:21] <mathrick> dolphy: nope
[22:23] <mathrick> hrmpf, it's rather not easy to find out cvsroot on gnome pages, gotta bug maintainer about that
[22:23] Action: mathrick searches again
[22:23] <dolphy> mathrick: hmm i don't know the anoncvs address
[22:24] <mathrick> duh, and link provided by Software map is bogus
[22:24] <mathrick> besides being http:// one of course
[22:24] <dolphy> mathrick: maybe just try :pserver:[email protected]:/cvs/gnome
[22:25] <mathrick> dolphy: yeah, I'll try, but GNOME CVS webpages could use serious usability facelifting
[22:25] thaytan ([email protected]) joined #gstreamer.
[22:25] <Company> grrrrrr
[22:25] <Uraeus> hi thaytan
[22:25] <mathrick> y0 +h4y+4n
[22:25] sublett ([email protected]) joined #gstreamer.
[22:26] Action: Company kicks qtdemux
[22:26] <mathrick> bah
[22:26] <mathrick> mathrick@megumi:~/Dev$ cvs -d:pserver:[email protected]:/cvs/gnome co totem
[22:26] <mathrick> no such user anonymous in CVSROOT/passwd
[22:28] Action: mathrick wonders how long LXR has been disabled now
[22:28] <mathrick> I've never seen it online, and I think I first tried something like 3-4 years ago :\
[22:28] <mathrick> "The GNOME Cross Reference is currently disabled. Look for it to return soon!", hah!
[22:29] <Company> mathrick: try @anoncvs.gnome.org
[22:29] <mathrick> yep, anoncvs works
[22:30] <taaz> mathrick: even a sid totem update today
[22:30] <mathrick> taaz: wow, that's active indeed
[22:33] kwm ([email protected]) left irc: "Leaving"
[22:33] <Uraeus> dolphy: starting totem I get: ** (totem:619): CRITICAL **: file egg-recent-model.c: line 328 (peek_state): assertion `info->states != NULL' failed
[22:33] <Uraeus> dolphy: any idea what is causing that?
[22:34] <Company> Uraeus: your gtk cvs
[22:35] <Uraeus> Company: my gtk is to new?
[22:35] <Company> (at least i have that since i updated to gtk 2.3.x
[22:36] <Uraeus> hmm, totem lists a bunch of those then segfaults for me
[22:37] thaytan ([email protected]) left irc: "See y'all"
[22:37] foser ([email protected]) left irc: "[ I want to believe ]"
[22:37] <Uraeus> thomasvs: who do I talk to in order to get access etc. for setting up a project on fd.o? hp?
[22:38] sjoerd_ ([email protected]) left irc: Remote closed the connection
[22:38] sjoerd ([email protected]) left irc: Remote closed the connection
[22:38] sjoerd ([email protected]) joined #gstreamer.
[22:39] 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 
[22:44] <dolphy> iain_: helloooooooo :)
[22:44] foser ([email protected]) joined #gstreamer.
[22:45] foser ([email protected]) left irc: Client Quit
[22:49] Action: dolphy starts debugging optimalscheduler :-/
[22:49] <Uraeus> dolphy: you are using i686 glibc?
[22:50] <dolphy> Uraeus: hmm dunno
[22:50] <dolphy> Uraeus: why ?
[22:50] <Uraeus> dolphy: you will not trigger the bug unless you do
[22:50] <Uraeus> err sorry
[22:50] <dolphy> Uraeus: which bug
[22:50] <Uraeus> I was thinking about basicomega scheduler
[22:50] <dolphy> Uraeus: my bug is something you probably never heard about
[22:51] <Uraeus> dolphy: a love bug?
[22:51] <dolphy> Uraeus: EOS not working
[22:51] <dolphy> Uraeus: quite important one :)
[22:51] Action: Uraeus was trying to make a stupid joke
[22:52] foser ([email protected]) joined #gstreamer.
[22:53] <Company> ds-work: qtdemux hangs with http://int1.fp.sandpiper.net/sonypde/spider-man_2-tsr_fs2.zip
[22:55] <Uraeus> monty going question beserk in #gnome :)
[22:56] Action: sxpert sees no monty
[22:56] <sxpert> Uraeus: which #gnome ?
[22:57] <Uraeus> xiphmont is Ogg monty (irc.gnome.org)
[22:57] <wheels> Uraeus: Well, he's usually on freenode too.
[22:58] <Uraeus> wheels: yeah, but he usually doesn't ask 10 000 compilation questions ;)
[22:58] <wheels> Uraeus: Ah.  :-)
[22:58] <wheels> Uraeus: Yeah, he's a reltively clever fellow that I think could handle a GNOME build.  :-)
[22:58] <Uraeus> wheels: think he covered fontconfig compilation to every detail now :)
[22:59] <wheels> Actually a lot of the Xiph guys are on Mac, actually.
[22:59] mxpxpod ([email protected]) joined #gstreamer.
[23:00] Action: wheels used to hang out in #vorbis in my days of being enlightened in the ways of the Ogg/Vorbis standards.
[23:00] <Uraeus> I guess after hacking on multimedia codecs for so long he has lost the ability to understand baser stuff like automagic :)
[23:01] <wheels> Uraeus: Well, or the trauma is making him forget.  ;-)
[23:01] jcsston ([email protected]) joined #gstreamer.
[23:01] Action: wheels puts autotools in the category of "necessary evil"
[23:02] <mxpxpod> is anyone here that has worked on gstmm?
[23:02] <Uraeus> mxpxpod: nope, that was the work of sienap and the guy who maintain libsig2 (forgot the name atm)
[23:03] <mxpxpod> martin schulz?
[23:03] <Uraeus> correct
[23:03] <mxpxpod> :)
[23:03] <Uraeus> sienap started it, but martin did most of the work I think
[23:03] <mxpxpod> Uraeus: him and murray
[23:03] <Uraeus> not sure how complete they got before martin was lost to libsig2 :)
[23:04] <Uraeus> not sure murray ever did that much really, he did some stuff for sure, but martin was the guy who had regular commits over a longer period
[23:04] <mxpxpod> heh
[23:04] <mxpxpod> ah, ok
[23:04] <mxpxpod> well, I just updated the .defs files for 0.7.4 and I'm trying to figure out some of this stuff
[23:04] <mxpxpod> it's a mess, though
[23:06] <mxpxpod> a guy I know and I would like to start working towards a gstmm release sometime :)
[23:06] <Uraeus> mxpxpod: yes, my impression is that it was never completed, and I guess picking up half finished projects are harder than something which is just bitrotten
[23:06] <Uraeus> mxpxpod: that would be great!
[23:07] <mxpxpod> Uraeus: but I gotta figure out how all this crap works :(
[23:07] <mxpxpod> including gstreamer
[23:07] <mxpxpod> not that gstreamer is crap
[23:07] <mxpxpod> just that the c++ bindings are a mess
[23:09] <Uraeus> hey cool! my penndrive popped up on my desktop automagically for the first time today :) 
[23:09] Action: mathrick whistles "messy binding for a messy language" song quietly ;P
[23:09] <sxpert> lol
[23:09] <mxpxpod> mathrick: :P
[23:09] <sxpert> mxpxpod: uh... C++ *is* crap
[23:10] <sxpert> mxpxpod: you can't expect muhc
[23:10] <sxpert> much even
[23:10] <mxpxpod> sxpert: now now... at least it's not VB
[23:10] <Company> ooops, i wrapped the wrong structure in xine
[23:10] <Company> ok, i'll do that tomorrow :/
[23:10] <Company> gnight
[23:10] Company ([email protected]) left irc: Remote closed the connection
[23:11] <sxpert> mxpxpod: uh... you're comparing jalapeno peppers to shit...
[23:11] <mxpxpod> sxpert: heh, so?
[23:11] <sxpert> mxpxpod: you can't do that 'D
[23:11] <mathrick> sxpert: btw, I need to take back my words about emacs CVS devs... it's actually Savannah that gives two cvs links, one being for webcvs, and them being pretty close to each other ;)
[23:11] <sxpert> mathrick: ah, he
[23:12] <sxpert> mathrick: lemme guess, you had the web site CVS ?
[23:12] <mxpxpod> man, last post to the gstmm list was in june of last year
[23:12] <mathrick> sxpert: yep, you have 10 points now, do you want to continue? ;)
[23:13] <sxpert> mathrick: ok, I cash them.. you owe me a beer ;D
[23:14] <mathrick> sxpert: ok, current exchange rate for EUR is 1/MAXINT ;)
[23:14] <sxpert> mathrick: the zloty isn't worth much last time I checked
[23:15] <mathrick> sxpert: I'm talking about you points, who told anything about zlotys?
[23:15] <sxpert> hehe
[23:15] <mathrick> s/you/your/ even
[23:16] <mxpxpod> ok, instead of using gst_object_destroy(), what should I use?
[23:17] <mathrick> mxpxpod: wild guess: g_object_destroy ?
[23:17] <mxpxpod> mathrick: ah, ok
[23:17] <mathrick> mxpxpod: after all, it's GObject
[23:17] <mxpxpod> mathrick: I forget minor details like that :)
[23:18] <mathrick> mxpxpod: hehe :)
[23:18] <mathrick> mxpxpod: but gst_object_destroy(), is it in gstmm code?
[23:18] <mxpxpod> yeah
[23:19] <mathrick> mxpxpod: man, that means gst-from-before-glib-2.0 I think O_o
[23:19] <mxpxpod> :(
[23:19] <mxpxpod> um, g_object_destroy isn't found
[23:21] <mathrick> hmmpf, I wonder if it shouldn't be plain g_object_unref() instead
[23:21] <mxpxpod> that happens once in that func already
[23:22] <mxpxpod> so I'll just comment out that object_destroy line
[23:23] <mathrick> mxpxpod: ::destroy is emited on refcount==0 automatically
[23:23] <mxpxpod> mathrick: ah, cool
[23:23] <mathrick> mxpxpod: so if it's in gst_object_unref(), just delete it
[23:24] <mxpxpod> I'm gonna have to email the list and figure out what all is going on with Gst::Object
[23:24] <sxpert> ROTFLMAO. intel adding 64 bit instructions to IA32 and labelling them "innovation"... http://developer.intel.com/technology/64bitextensions/
[23:24] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[23:25] <mathrick> mxpxpod: like, what exactly you don't know? If it's just hierarchy matter, all that there's to it is in docs I think
[23:26] <mxpxpod> no, that's not it
[23:29] <Uraeus> night
[23:29] Uraeus ([email protected]) left irc: "Client exiting"
[23:31] <mathrick> sxpert: it's still tad better than MS "inventing" hardlinks after "a year of intensive engineering efforts", and calling them symlinks for that matter ;)
[23:32] <sxpert> yeah
[23:32] <sxpert> lol
[23:32] <sxpert> mathrick: and patenting the concept ?
[23:33] <mathrick> sxpert: no, I think that just ripping idea out of their own product is enough to satisfy a beast ;)
[23:33] <sxpert> lol
[23:34] <mathrick> sxpert: FYI, "their product" == Xenix
[23:34] <sxpert> that belongs to SCO now :D
[23:43] <mxpxpod> is GST_SEEK_METHOD_SHIFT going to change anytime soon?
[23:43] <ds-work> not until 0.9
[23:44] <mxpxpod> ok, cool
[23:44] <matju> sxpert: note how it's labelled "innovation" but they don't claim it's theirs. it's just that they think "innovation" sounds good so they use it whenever they can.
[23:44] <mxpxpod> because the enums.pl script for generating the enums definitions can't pick up GstSeekType's enum values
[23:45] <matju> sxpert: if anyone assumes "innovation" implies "intel" anyway, the marketing dept has won.
[23:45] nated ([email protected]) left irc: "Client exiting"
[23:45] <sxpert> matju: that's what it would appear to the average CTO moron
[23:48] Nick change: harshyWork -> harshy
[23:49] <dolphy> ds-work: i m reverting my change
[23:49] <ds-work> ok, cool
[23:49] <dolphy> ds-work: and adding GST_PAD_IS_NEGOTIATING macro
[23:50] <ds-work> ok
[23:50] <wheels> matju: Actually they did innovate -- and it mostly flopped.
[23:50] <wheels> matju: The ia64 is a really cool architecture; it's just not very commercially viable.
[23:51] thaytoo ([email protected]) joined #gstreamer.
[23:51] zaheerm ([email protected]) joined #gstreamer.
[23:51] <zaheerm> hi guys
[23:51] <mathrick> was ia64 actually manufactured / sold in real life?
[23:51] <zaheerm> has anyone played with the jack plugin in gst?
[23:52] <mathrick> lo zaheerm, wb thaytoo
[23:52] <zaheerm> i tried helloworld.c edited to use a jackbin and a jacksink
[23:52] <zaheerm> when used with 0.7.4 i get a segfault
[23:52] <ds-work> mathrick: of course.  There are Linux ports
[23:53] <ds-work> mathrick: GStreamer runs on ia64
[23:53] <dolphy> ds-work: have you seen the bug i reported for caps nego race
[23:53] <mathrick> ds-work: I'm not interested in ports, I mean, are they being sold to real customers?
[23:53] <zaheerm> and with 0.6.4, the pipeline runs but....* (process:20739): WARNING **: element play_audio performs a yield, please fix the element
[23:53] <wheels> mathrick: sure -- but it's only targeted at servers
[23:54] <sxpert> wheels: it's nice on paper. the realization however, sucked balls
[23:54] <mathrick> wheels: k, but AFAIK, itanium 1 was never widely deployed?
[23:54] <wheels> mathrick: (or well, it's not targeted at things other than normal workstations might be a better way of putting it.)
[23:54] <wheels> mathrick: Well, that's just because nobody bought them because they were slow, high power consumption and expensive.  :-)
[23:55] <mathrick> wheels: k, now it agrees with what I've heard about Itanium :)
[23:55] <wheels> sxpert: Not really -- it's a really cool architecture from a design standpoint; the problem is that to really take advantage of it implies throwing away most of the assumptions that modern compilers have been built around.
[23:55] <mxpxpod> in order of operations... does << come before or after +?
[23:55] <zaheerm> any1 know the status of jack support in gst?...and what may be required to fix it...
[23:55] <sxpert> wheels: I know, HP *donated* a cluster to the uni, but they forgot to tell us about the size of the cooling plant necessary... 200000 eur later...
[23:55] <ds-work> dolphy: you're interpreting the backtrace incorrectly
[23:56] <wheels> mathrick: Well, Itanium II has pretty good performance, but the other things still apply.
[23:56] <sxpert> wheels: as I said, nice on paper...
[23:56] <dolphy> ds-work: any fix ?
[23:57] <mathrick> mxpxpod: after, and use google Luke :)
[23:57] <mxpxpod> mathrick: :P
[23:57] <sxpert> wheels: also, the instruction set has too many limitations...
[23:57] <dolphy> ds-work: i don't really understand how the caps renego works...
[23:57] <wheels> mathrick: the problem is that Intel is a still fundamentally a low end player; nobody wanted to buy something really expensive from the guys that were driving most of the world's PCs...
[23:57] <dolphy> ds-work: if everything is done from the thread that called gst_pad_try_set_caps
[23:57] <mxpxpod> mathrick: so, 1 << GST_SEEK_METHOD_SHIFT + 0 should work, right?
[23:57] <wheels> sxpert: Some might have once said that about, say, RISC.  ;-)
[23:57] <mxpxpod> mathrick: instead of having it all ()'d
[23:57] <dolphy> ds-work: or if every element will negotiate in it's own thread
[23:57] <ds-work> dolphy: if everything happens from the same, thread, it should be a problem
[23:58] <sxpert> wheels: referring to the condition upon which you can put multiple instruction in one
[23:58] <dolphy> ds-work: ohhhhhh!!!
[23:58] <dolphy> ds-work: think i know how that's possible
[23:58] <mathrick> mxpxpod: it should do, yes
[23:58] <mxpxpod> mathrick: ok, thanks
[23:58] <dolphy> ds-work: one call is done through the ximagesink thread because of Configureevent arriving from the chain function xevents poll
[23:59] <wheels> sxpert: Well, that's not bad -- it's just one of the first instruction sets that assumes that one will not be writing assembly by hand pretty much ever.  At the same time it alientates modern compilers...
[23:59] <dolphy> ds-work: and another call might overlap from the gtk thread
[23:59] <dolphy> ds-work: when calling gst_x_overlay_expose
[23:59] <ds-work> dolphy: that's a problem :)
[23:59] <dolphy> ds-work: indeed...
[23:59] <dolphy> ds-work: i ll try to fix that tomorrow
[23:59] <dolphy> night
[23:59] <mathrick> nite dolphy
[23:59] <wheels> sxpert: But bringing this back to the original context -- ia64 is a much more "innovative" architecture than x86_64 -- in fact ia64 is one of the more interesting ones to have come along in the last few years; it's just not as marketable as x86_64.
[00:00] --- Wed Feb 18 2004
[00:00] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[00:00] <sxpert> wheels: considering how bad compilers still are with respect to that crap, our researchers pretty much write the calculation functions in assembly
[00:00] Company ([email protected]) joined #gstreamer.
[00:00] <sxpert> wheels: yeah, nice on paper ;D
[00:01] <zaheerm> i will assume that its broken...
[00:01] <sxpert> wheels: then, it fails all reality checks
[00:02] <wheels> sxpert: But "reality" and "innovative" aren't connected terms, really.  "reality" and "useful" might do better.
[00:02] sxpert_work ([email protected]) left irc: Read error: 110 (Connection timed out)
[00:02] <sxpert> yeah, that's my point ;D
[00:03] <mathrick> zaheerm: AFAIK jimmy_dean have used it some, don't know how much tho
[00:03] <wheels> sxpert: But that's completely different from where this started or what I was saying.  :-P
[00:03] <sxpert> ia64 doesn't do anything that x86-64 does in a simpler way...
[00:03] <mathrick> sxpert: now, you're mistaken
[00:03] <sxpert> uh, I think I messed up that last sentence... 
[00:03] <sxpert> time for bed
[00:03] <wheels> sxpert: Sure it does -- because it doesn't inherit 20 years of stupid x86 hacks...
[00:04] <mathrick> sxpert: anything that continues crappy x86 certainly *isn't* simple
[00:04] <sxpert> mathrick: agreed, long live the PPC :D
[00:04] ChrisHJW ([email protected]) left irc: Connection timed out
[00:04] <sxpert> mathrick: last I checked, opteron in ia32 mode "emulates"...
[00:05] <zaheerm> whats the gst-launch syntax for using a custom bin such as jackbin?
[00:05] <zaheerm> is it jackbin.{ blah ! blah ! blah } ?
[00:06] <Company> ( ) instead of { }
[00:06] Company ([email protected]) left irc: Remote closed the connection
[00:09] <matju> wheels: back
[00:09] <matju> wheels: btw, isn't 64-bit ext tech unrelated to IA-64 ?
[00:09] <wheels> sxpert: No, Opteron has the ia32 instruction set in hardware.
[00:10] <matju> wheels: because IA-64 is not compatible with IA-32 (nor IA-x86 in general)
[00:10] <wheels> matju: right
[00:11] <wheels> matju: ia64 (Itanium) does do emulation, whereas the majority of the Opteron's native instruction set is still the standard ia32 instruction set.
[00:11] <matju> sxpert: whatever "emulates" means... it can be argued that any processor since the pentium (or even earlier) doesn't really understand the ia32 set anymore and has an embedded translator to emulate stuff.
[00:12] Action: ds-work hugs PowerPC.  Only had to add 3 instructions for full 64-bit user space
[00:13] <wheels> ds-work: Well, that's basically what happened with x86_64 too.  Or well, Sparc or PA-RISC for that matter.
[00:18] Action: mathrick has no clue what PA-RISC really is, except seeing its name here and there couple of times
[00:19] <wheels> mathrick: processor produced by HP for quite a while -- a lot of the HPUX machines use PA-RISC processors.
[00:19] <ds-work> wheels: x86-64 added a *lot* of instructions and/or instruction modes
[00:20] <wheels> ds-work: Well, for one it updated to the latest Intel instruction sets including i.e. SSE2.
[00:20] ibsteveog ([email protected]) left irc: "quitting"
[00:21] <mathrick> wheels: yeah, I kind of vaguely associate it with HP, but that's all. But then, I have no clue about world outside of x86 altogether :>
[00:21] thomasvs ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[00:21] Action: mathrick wonders where non-x86 people get from
[00:22] Action: wheels does mostly porting of stuff to 64 bit linux at work.
[00:22] <mathrick> from my experience, there's no such person, except handful of Mac users that have no clue about hardware anyway
[00:23] markey ([email protected]) left irc: "nite"
[00:23] <mathrick> I wish I had more contact with other than x86 archs
[00:23] <wheels> Plus actually my first Linux box was an Alpha.  :-)
[00:24] <mathrick> heh, it's probably only processor ever that made it into news :)
[00:24] Action: mathrick remembers hearing about blazing speed 300(150?)MHZ Alpha in evening news ;)
[00:25] <matju> mathrick: 90% of the time i use PC's, but for the rest, i've been exposed to PPC/Mac, Sparc, Arm, HP-PA, SGI/mips, ... and 6809 =)
[00:25] Action: wheels had a 166 MHz system with 128 MB of RAM in ~1996-97...
[00:25] <mathrick> hmm, MIPS is probably arch that's most mysterious to me
[00:26] <wheels> mathrick: it's essentially the same as powerpc
[00:26] <wheels> mathrick: the instruction sets are almost identical
[00:26] <mathrick> no clue who makes it (it's some kind of open spec?)
[00:26] <mathrick> wheels: really?
[00:26] <matju> mathrick: afaik, MIPS is used in SGI, Nintendo64, possibly Playstation too, not sure
[00:26] <wheels> matju: yeah, plus a lot of random crap...
[00:26] <mathrick> wheels: but PPC is fixed endianness?
[00:27] <wheels> mathrick: well, it used to be a separate company, but then it was bought by SGI.
[00:27] <mathrick> MIPS switched endainness is funky
[00:27] <mathrick> wheels: huh? I thought PPC was IBM's doing
[00:27] <mathrick> what is that endianness thingy useful for?
[00:28] <wheels> mathrick: Well, IBM and Moterola, but PPC and MIPs are both pretty standard RISC.
[00:28] <wheels> mathrick: PowerPC has a few more instructions (and granted the last time I was really looking into this was about 5 years ago, but I doubt they've changed radically since.)
[00:28] <mathrick> wheels: is there anything like standard RISC?
[00:29] <mathrick> wheels: i know there's common RISC model, but is there standard actual implementation?
[00:29] <wheels> mathrick: no -- there may have once been, but no longer.
[00:30] <wheels> mathrick: (like when it was a research concept there was probably a reference implementation in a lab or something, but I'm just speculating there)
[00:30] <pb_> mathrick: like most modern risc machines, powerpc does support both endiannesses.  but almost everybody runs it big endian, presumably in deference to motorola.
[00:30] <mathrick> wheels: k
[00:30] <mathrick> I don't get that choosable endianness
[00:31] <mathrick> why would anybody use something like that?
[00:31] <pb_> from the chip vendor's point of view, if you choose one fixed endianness then you are guaranteed to offend a certain fraction of the people who would otherwise be potential customers.
[00:32] <pb_> providing switchable endianness costs virtually no hardware, and it allows you to make friends with both camps.
[00:32] <mathrick> really? is that of sole political reason?
[00:32] <mathrick> nuts
[00:33] <mathrick> and I'd have thought it would be rather costly hw wise to support both
[00:33] Action: wheels notes that he's not even really a hardware guy, but I'm getting sort of sucked into it at work.  :-)
[00:34] <wheels> mathrick: yeah, getting those AND and OR gates working properly with both endianesses is rough.  :-)
[00:34] <thaytoo> mathrick: not really, they have a single element to switch data as it comes in from the RAM
[00:34] <thaytoo> and then most of the chip operates in a fixed endianness
[00:35] <thaytoo> I can't remember which one though :)
[00:36] <mathrick> thaytoo: ahh, right, it's just simple Jedi mind trick ;)
[00:36] <mathrick> thaytoo: I should have guessed :)
[00:40] thomasvs ([email protected]) joined #gstreamer.
[00:41] yippi ([email protected]) left irc: "Client exiting"
[00:43] <ds-work> for PowerPC, it's entirely in software
[00:43] <ds-work> so essentially, to get a little-endian powerpc, you use lwx instead of lw instructions
[00:44] <ds-work> except, for some silly reason, they added a hardware swap as well
[00:44] yippi ([email protected]) joined #gstreamer.
[00:44] <ds-work> so lw means "load using the hardware endianness", and lwx means "load and byteswap"
[00:48] matju ([email protected]) left #gstreamer ("Client exiting").
[00:56] <thaytoo> ds-work: they probably had a spare instruction ID
[00:56] <ds-work> PowerPC has lots of spare instruction encodings
[00:56] <ds-work> about 35% free, I think
[00:57] <mathrick> yep, instruction set that doesn't have -x ended instruction is not h4x0r enough :)
[00:57] <thaytoo> and it makes sense when it costs next to nothing in hardwar, and you know that most of the rest of the world is little endian
[00:57] <mathrick> s/instruction/\&s/
[00:58] <thaytoo> woot, I have commit access again
[01:00] ct_ ([email protected]) joined #gstreamer.
[01:03] sublett ([email protected]) left irc: "I like food, food is good!"
[01:04] thomasvs ([email protected]) left irc: Read error: 110 (Connection timed out)
[01:06] thomasvs ([email protected]) joined #gstreamer.
[01:07] bitpoet ([email protected]) left irc: Remote closed the connection
[01:13] steveb__ ([email protected]) joined #gstreamer.
[01:13] mxpxpod ([email protected]) left irc: "Guns kill people as much as spoons made Rosie O'Donnell fat.."
[01:13] steveb_ ([email protected]) left irc: Read error: 60 (Operation timed out)
[01:16] <bitshifter> just out of curiosity: why is gstreamer-0.7 dependent on libxml2? I couldn't find a single #include <libxml/foo> in the gstreamer source tree
[01:16] <bitshifter> well, the gstreamer/ module at least
[01:16] Shoragan ([email protected]) joined #gstreamer.
[01:18] yippi ([email protected]) left irc: "Client exiting"
[01:20] zaheerm ([email protected]) left #gstreamer.
[01:20] <mathrick> nite
[01:21] Nick change: mathrick -> mathrick|sleep
[01:21] mxpxpod ([email protected]) joined #gstreamer.
[01:21] <mxpxpod> what's the replacement for gst_element_factory_new?
[01:23] <ds-work> gst_element_register()
[01:23] <mxpxpod> ds-work: so, what's the point of the factory then?
[01:24] <ds-work> it's the factory code that creates elements from their names
[01:25] <mxpxpod> so how do I create a new one?
[01:26] <ds-work> why would you need to?
[01:27] <ds-work> there's no GstElementFactory object
[01:27] <ds-work> er, actually, there is
[01:27] <mxpxpod> I was gonna say....
[01:27] <ds-work> but you never have to deal with it
[01:28] <mxpxpod> hmm
[01:28] <mxpxpod> the reason I ask is because I'm working on gstmm (again) and I'm trying to figure out how the constructor should work
[01:30] <ds-work> it's not useful to just create GstElementFactory objects
[01:31] <ds-work> it's essentially an internal object that is a link between registry pools and element classes
[01:31] <mxpxpod> but if I can't create an ElementFactory, how can I return it in the find method?
[01:32] <ds-work> the function returns an object
[01:34] <mxpxpod> ok
[01:34] <mxpxpod> ds-work: thanks
[01:41] <mxpxpod> ds-work: is GstBufferPool around anymore?
[01:42] foser ([email protected]) left irc: "[ I want to believe ]"
[01:51] <ds-work> mxpxpod: no
[01:51] <mxpxpod> ds-work: thanks
[02:00] <mxpxpod> dude
[02:00] <mxpxpod> the gstmm wrappers I just fixed work!
[02:00] Shoragan ([email protected]) left irc: "Leaving"
[02:01] <mxpxpod> time to go home
[02:01] <jimmy_dean> mathrick|sleep: I'm still planning on eventually getting a silly equalizer into gstreamer...I've just been delayed lately due to school :(
[02:10] ChrisHJW ([email protected]) joined #gstreamer.
[02:12] pb_ (~pb@2002:5160:45ef:0:240:95ff:fe30:aa84) left irc: "Client exiting"
[02:20] steveb_ ([email protected]) joined #gstreamer.
[02:20] Rotty ([email protected]) left irc: Remote closed the connection
[02:33] smoke ([email protected]) got netsplit.
[02:36] iain_ ([email protected]) left irc: "SO MUCH LOVE SO LITTLE TIME"
[02:38] steveb__ ([email protected]) left irc: Read error: 110 (Connection timed out)
[02:39] smoke ([email protected]) got lost in the net-split.
[02:44] mxpxpod ([email protected]) left irc: "Guns kill people as much as spoons made Rosie O'Donnell fat.."
[02:44] steveb_ ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[02:45] steveb_ ([email protected]) joined #gstreamer.
[02:51] <ds-work> loaded 150 plugins with 502 features
[02:51] <ds-work> I've never seen over 500 features before
[02:51] ct__ ([email protected]) joined #gstreamer.
[02:52] ChrisHJW ([email protected]) left irc: Connection timed out
[02:59] ct_ ([email protected]) left irc: Read error: 60 (Operation timed out)


-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click