[gst-devel] Daily IRC logs

wim.taymans at chello.be wim.taymans at chello.be
Sun May 6 06:27:35 CEST 2001

[06:30] <arik> DEBUG(5644:-1)gst_thread_signal_thread:342: signaling thread
[06:30] <arik> INFO(5646:-1):gst_thread_main_loop:335: gstthread: thread "main_thread" is stopped
[06:30] <omega_> -1 enabled everything (== 0xffffffff)
[06:30] <arik> right
[06:30] <omega_> that indicates that you're going to PAUSED or READY at some point, on a thread
[06:30] <arik> hmm
[06:31] <arik> READY i think
[06:31] <omega_> why?
[06:31] <arik> in play i am going to READY and then PLAY
[06:31] <omega_> it shouldn't stop the thread unless you're switching *away* from playing
[06:32] <arik> hmm
[06:32] <arik> DEBUG(5699:-1)gst_pad_renegotiate_func:1086: negotiation succeeded
[06:32] <arik> DEBUG(5699:-1)gst_pad_renegotiate:1174: pads aggreed on caps :)
[06:41] <arik> i wonder why it's not playing any frame's
[06:41] <arik> a gst_main issue maybe?
[06:42] Nick change: taaz -> taazzzz
[06:42] <omega_> depends on your pipeline structure
[06:42] <arik> right, i'm guessing it's wrong
[06:44] <omega_> without the code I can't tell
[06:44] <arik> omega_: i could send it to you if you have time
[06:45] <arik> it's prob laughable though ;-)
[06:45] <omega_> I can look at it, yeah
[06:45] <arik> k, give me a sec
[06:45] <arik> at this point it's actually similar to gstmediaplay
[06:46] <arik> mail address?
[06:46] <omega_> omega at temple-baptist.com
[06:46] <arik> k
[06:47] <arik> sent
[06:48] <arik> i'm sure i'm just backwards, i don't really understand the pipeline stuff yet
[06:49] <omega_> ooh, the binary works right away, but I'm gonna rebuild anyway <g>
[06:49] <arik> hehe ;-)
[06:49] <omega_> what version of gstreamer do you have installed?
[06:49] <arik> latest anon cvs
[06:50] <omega_> you installed it?
[06:51] <arik> yep
[06:51] <omega_> brave <g>
[06:51] <arik> heh
[06:54] <omega_> takes a media file in argv[1] ?
[06:54] <omega_> any specific type?
[06:54] <arik> nope
[06:55] <arik> what i need is for mediasink to be a generic playback widget
[06:55] <arik> autoplug
[06:55] <omega_> yup
[06:55] <arik> so i can move on to features ;-)
[06:55] <omega_> that's the idea behind libgstplay
[06:56] <omega_> yeah, there's a significant problem with state, the main thread is being set to NULL
[06:56] <arik> heh
[06:56] <arik> that would be bad ;-)
[07:17] <arik> see anything?
[07:18] <omega_> sorry, distracted
[07:18] <arik> s'ok
[08:39] <arik> dum de dum
[08:39] <omega_> ?
[08:51] <arik> hey
[08:51] <omega_> yeah?
[08:51] <arik> nothing, just coding away
[08:51] <omega_> ok, I'm trying to follow the debug statements here...
[08:51] <arik> cool, thanks
[08:51] <arik> it's prob more then one problem
[08:51] <arik> if we can get this working i'll put it in cvs
[09:08] <arik> i'm happy with the general design
[09:10] wtay-zZz (wim at cable-195-162-214-190.upc.chello.be) got netsplit.
[09:10] arik (arik at adsl-64-174-42-2.dsl.snfc21.pacbell.net) got netsplit.
[09:10] omega_ (omega at omegacs.net) got netsplit.
[09:10] omega_ (omega at omegacs.net) returned to #gstreamer.
[09:10] wtay-zZz (wim at cable-195-162-214-190.upc.chello.be) returned to #gstreamer.
[09:10] arik (arik at adsl-64-174-42-2.dsl.snfc21.pacbell.net) returned to #gstreamer.
[09:58] steveb (steveb at node1ee06.a2000.nl) joined #gstreamer.
[09:58] <omega_> arik: you there?
[09:58] <steveb> yo
[09:58] <omega_> yo
[09:58] <arik> omega_: yep
[09:58] <omega_> arik: see mediasink.c:342
[09:59] <arik> k
[09:59] <arik> there
[09:59] <omega_> see what's wrong?
[09:59] <omega_> that's not the only problem, but it was a big one
[09:59] <arik> hmm
[10:00] <arik> heh
[10:00] <arik> yeah i do see the problem
[10:00] <arik> right
[10:00] <arik> ;-)
[10:00] <arik> incorrect translation from sample code
[10:02] <omega_> then in media_sink_load_uri, you remove the disksrc from priv->bin and create a new priv->bin
[10:02] <omega_> where you should simply add it to priv->thread, along with new_element
[10:03] <arik> ah
[10:03] <arik> ok
[10:03] <omega_> and there are some autoplug problems here that are a lot harder to deal with
[10:03] <arik> hmm
[10:04] <arik> see i told you ;-)
[10:04] <arik> this is the problem with copying code ;-)
[10:04] <omega_> right, you see why I wanted to just get what we have working ;-)
[10:04] <arik> hehe
[10:04] <arik> well that doesn't work either but for different reasons
[10:05] <omega_> yeah, but it's more understood
[10:05] <omega_> and I'm about to fall asleep soon
[10:05] <arik> heh
[10:06] <arik> so do i not want to parent the disksource to the bin at all?
[10:06] <arik> or do i want to reparent it later to the thread?
[10:06] <omega_> well, the problem is that the autoplugger is just plain incorrect for incsched1
[10:07] <arik> ah
[10:07] <arik> so i do want to use the colorspace stuff then
[10:07] <omega_> the patch I made is to, after the typefind finishes, remove it from the bin (no change), *not* create a new pipeline, add the disksource and new_element to the priv->thread
[10:07] <omega_> colorspace?
[10:07] <arik> right
[10:07] <arik> same thing i did
[10:07] <omega_> huh?
[10:07] <omega_> you just lost me
[10:08] <arik> sorry
[10:08] <arik> colorspace is where gstmediaplay crashes for me
[10:08] <omega_> hmmm
[10:08] <omega_> we're nowhere close to that yet
[10:08] <arik> it fails a g_assert != NULL
[10:08] <omega_> neat, what line?
[10:08] <arik> not in my code, in the gstmediaplay code, line 146
[10:08] <omega_> checking
[10:09] <omega_> which file?
[10:09] <arik> gstplay.c
[10:09] <arik> sorry, should have said that ;-)
[10:09] <omega_> are you using HEAD or INCSCHED1 ?
[10:09] <omega_> HEAD
[10:10] <arik> HEAD i think
[10:10] <omega_> you definitely want to be using INCSCHED right now
[10:10] <arik> ah
[10:10] <arik> what is that?
[10:10] <arik> a branch
[10:10] <omega_> a branch with significant enhancements
[10:10] <arik> but for what?
[10:10] <arik> oh ok
[10:10] <arik> the branch tag is INCSHED1?
[10:10] <omega_> completely different scheduler system
[10:10] <omega_> BRANCH-INCSCHED1
[10:11] <arik> ah
[10:11] <arik> ok
[10:11] <arik> updating now
[10:12] <omega_> chance are, if it's faulting on colorspace, you don't have Hermes installed
[10:12] <omega_> therefore colorspace doesn't get built
[10:12] Action: steveb whispers "when will the merge be?"
[10:13] <arik> hermes?
[10:13] <omega_> I'm closing on the problem with the mixer example, I should figure that out tomorrow, then I think the merge is appropriate
[10:13] <omega_> hermes is an rgb->rgb colorspace conversion library
[10:13] <omega_> search freshmeat for it
[10:13] <arik> ok
[10:13] <arik> hmm
[10:13] <omega_> then autogen.sh as usual in incsched1
[10:13] <arik> right
[10:14] <arik> i'm just not finding the branch
[10:14] <omega_> cvs update -dP -r BRANCH-INCSCED1
[10:14] <omega_> er, INCSCHED1
[10:14] <arik> right
[10:14] <omega_> that fails?
[10:14] <arik> yep
[10:14] <arik> oh wait
[10:14] <arik> nm
[10:15] <omega_> sp?
[10:15] <arik> hrm
[10:15] <arik> there we go
[10:15] <arik> yeah i think so
[10:15] <arik> got it
[10:15] <arik> now to get hermes
[10:15] <arik> and then if gstmediaplay compiles
[10:15] <arik> i can base my code off it with a plan to merge back at some point
[10:16] <arik> i have a pretty different coding style then that though
[10:17] Action: omega_ is gonna have to write the shell of the dynamic autoplugger tomorrow
[10:17] <omega_> in the meantime, /me needs to sleep
[10:17] <arik> heh
[10:17] <arik> night
[10:17] <omega_> arik: you're off for the summer, right?
[10:18] <arik> omega_: not exactly, i am working
[10:18] <arik> right now
[10:18] <arik> why?
[10:18] <omega_> just wondering
[10:18] <arik> i work at netscape
[10:18] <omega_> so you'll be around tomorrow?
[10:18] <omega_> ah
[10:18] <arik> yeah
[10:18] <omega_> you mean aol
[10:18] <arik> hehe
[10:18] <arik> ;-P
[10:18] Uraeus (cschalle at c224s9h5.upc.chello.no) joined #gstreamer.
[10:18] <omega_> er, aol/time-warner
[10:18] <omega_> yo
[10:18] <Uraeus> hi
[10:18] <arik> aoltw
[10:18] <omega_> just in time for me to leave ;-)
[10:18] <arik> but yeah i'll be around tomorrow
[10:18] <omega_> ok
[10:18] Action: omega_ sleeps
[10:18] <arik> ttyt
[10:19] <omega_> l8r Uraeus
[10:19] <Uraeus> hmm
[10:19] omega_ (omega at omegacs.net) left irc: [x]chat
[10:19] <ajmitch> hey Uraeus ;)
[10:19] <Uraeus> arik: here to make a mozilla plugin for gstreamer?
[10:19] <Uraeus> ajmitch: hi
[10:19] <arik> Uraeus: heh, no not yet, good idea though ;-)
[10:19] <ajmitch> Uraeus: converted to debian yet?
[10:19] <arik> Uraeus: i may just have to do that at some point
[10:19] <Uraeus> ajmitch: never!
[10:20] Action: ajmitch sighs
[10:20] <Uraeus> arik: yup, can't have you working at netscape without doing some GNOME integration stuff :)
[10:20] <arik> Uraeus: ;-) i am actually doing that atm
[10:20] <arik> GNOME integration that is
[10:21] <Uraeus> arik: ok, so the big question is: when do you, Stuart and Christhopher B. get togheter and move mozilla from xpcom to bonbobo :)
[10:21] <arik> Uraeus: bwahahahahaha ;-)
[10:22] <Uraeus> I take that as a 'not soon' :)
[10:22] <arik> hehe
[10:22] <arik> i'll mention it to pav and blizzard ;-P
[10:22] <ajmitch> Uraeus: when will KDE be moved to bonobo? ;)
[10:23] <Uraeus> ajmitch: actually it is not an impossibility, think Mattias Ettrich for one is interested in the idea
[10:23] <ajmitch> Uraeus: i know ;)
[10:23] <arik> Uraeus: well they don't have a componant model
[10:23] <arik> so it does make some sense
[10:23] <ajmitch> i'd like to see it happen
[10:23] <ajmitch> they don't?
[10:24] <ajmitch> what do they have then?
[10:24] <arik> an embedding tool more or less
[10:24] <Uraeus> but I think GNOME would have to make some sorts of concessions, or many of the KDE hackers will feel they have to swallow to much pride
[10:24] <arik> kparts does interprocess embedding but it isn't a component model
[10:24] <ajmitch> they have KParts & XParts
[10:25] <ajmitch> i think KParts is closer to a component model
[10:25] <arik> closer
[10:25] <ajmitch> how is bonobo different?
[10:25] <ajmitch> what makes a 'true' component model?
[10:26] <arik> linking mostly and actual network-transparent object querying and whatnot
[10:26] <arik> but i'm by no means an expert
[10:27] <ajmitch> yeah, not sure if DCOP works transparently over the network
[10:27] <ajmitch> i know that it can use XMLRPC
[10:27] <arik> don't think so, but i could be wrong
[10:27] <arik> that's pretty cool
[10:27] <Uraeus> ajmitch: i am partially moving away from RH today :)
[10:27] <ajmitch> Uraeus: how?
[10:28] <Uraeus> ajmitch: I am installing the SGI version of redhat
[10:28] <arik> heh
[10:28] <ajmitch> Uraeus: sicko ;)
[10:28] Ow3n (owen at ti34a80-0852.bb.online.no) joined #gstreamer.
[10:28] <ajmitch> hey Ow3n
[10:28] <Ow3n> yo
[10:29] <Uraeus> hi Ow3n
[10:29] <Ow3n> Uraeus: What are you doing inside on a day like this?
[10:29] <Uraeus> waking up
[10:29] <Ow3n> Me too.
[10:30] <Ow3n> Today was going to be a hacking day but looking at the sky I'm having 2nd thoughts.
[10:30] <Uraeus> Ow3n: I think we should move soon if we want to propose GnoStream for GNOME 2
[10:31] Action: ajmitch must look @ gnostream
[10:31] <Ow3n> The problem is there just isn't much to look at yet.
[10:31] <Ow3n> At least not much ccode.
[10:32] <Ow3n> Well, there is quite alot of code actually but it doesn't do much yet :(
[10:32] <ajmitch> oh well, just get some victims^H^H^H^H^H^H^Hcoders to help you
[10:32] <Ow3n> :)
[10:32] <Ow3n> Any volunteers?
[10:32] <Ow3n> <g>
[10:33] <Uraeus> Ow3n: what did Jim Gettys say when you called him?
[10:33] steveb (steveb at node1ee06.a2000.nl) left irc: Read error to steveb[node1ee06.a2000.nl]: EOF from client
[10:34] <ajmitch> hehe, i still gotta get onto a generic player component if it's still neded
[10:34] <Ow3n> He was still in Houston :(
[10:34] <Ow3n> I'll try again on Monday.
[10:34] Action: Uraeus is awfully temptet to say; Houston we have a......no I better not it is overused :)
[10:35] <Ow3n> Doh
[10:36] <ajmitch> so you hope to have GNOME include gnostream and gstreamer? ;)
[10:36] <ajmitch> muhaha, the takeover has started ;)
[10:36] <Ow3n> He-he
[10:37] <Uraeus> ajmitch: the sales pitch is: a soundserver is yesterday news, we need a media server to stay current
[10:37] <Ow3n> ajmitch: Have you got any further with your bonobo stuff?
[10:38] <ajmitch> Ow3n: no, unfortunately uni has taken over :(
[10:39] <ajmitch> i am hoping to try and set aside some time for coding soon ;)
[10:39] <Ow3n> Ah, yes I remember the problem. Unfortunately it doesn't get any better when you get a job.
[10:39] <arik> later all
[10:39] arik (arik at adsl-64-174-42-2.dsl.snfc21.pacbell.net) left irc: leaving
[10:39] <Uraeus> ajmitch: how does your bonobo stuff compare to bonobo-media?
[10:40] <ajmitch> Uraeus: heh, i haven't really done anything yet nor looked at bonobo-mdeia
[10:40] <Ow3n> It's kind of tough using bonobo when there are so few examples.
[10:41] <Uraeus> Ow3n: have you looked at bonobo-media?
[10:41] <Ow3n> I had a look a while ago. Just before GUADEC.
[10:41] <Uraeus> ajmitch: well if you are looking for a project which is easier to fit into a tight schedule so do gstmediaplay desperatly need some loving
[10:41] <Ow3n> The problem is that most of the code using bonobo is just using it to encapsulate widgets.
[10:42] <Uraeus> Ow3n: well now Cactus and wtay has done some integration so maybe there is more of use now
[10:42] <ajmitch> Uraeus: what sorta luvin does it need?
[10:42] steveb (steveb at node1ee06.a2000.nl) joined #gstreamer.
[10:43] <ajmitch> Uraeus: cos i thought that others were working on it or similar projects ;)
[10:43] <Uraeus> ajmitch: well, 1. choice of output plugin to use., 2. try using Xmms plugins if no Gstreamer plugin for format available.,3.maybe add support for Xmms and native effects plugins
[10:44] <ajmitch> playlists?
[10:44] <Uraeus> ajmitch: that too
[10:44] <Uraeus> ajmitch: the only project which really competes AFAIK is hadess stuff, but that is just for music
[10:44] <ajmitch> i should try and use some xmms plugins, how good is the support?
[10:44] <ajmitch> yeah, i want a decent video player to show stuff off
[10:45] <Uraeus> ajmitch: very dodgy for me, mp3 worked and mod works, but none of my others
[10:45] <ajmitch> what about better choice of input, like v4lsrc?
[10:45] <ajmitch> i got a friend who has a supported card, and wants to watch tv on her computer
[10:45] <Uraeus> ajmitch: well, that would be nice, but having a basic working mediaplayer is pri-1 I think
[10:46] <ajmitch> yeah ;)
[10:46] Action: ajmitch sees if gstreamer compiles today
[10:46] <Uraeus> ajmitch: adding the my first two issues and your playlist would really do a lot do improve gstmediaplay
[10:47] <ajmitch> yes, a decent media player would be useful (must be better than noatun!)
[10:47] <ajmitch> we cannot let that arrogant Njaard have better software!! ;)
[10:47] <Uraeus> no, no way :)
[10:48] <Uraeus> actually, if we add the option to choose output plugin people could even test loading a mpeg and outputing it a AVI, that would be a real user eye opener
[10:49] <ajmitch> yeah
[10:50] <Uraeus> well time for me to get some air, sun and most importantly some food :)
[10:50] Nick change: Uraeus -> Ura_out
[10:50] <ajmitch> hehe
[11:23] Nick change: ajmitch -> ajzzzz
[12:43] ajzzzz (ajmitch at p10-max7.dun.ihug.co.nz) left irc: Ping timeout for ajzzzz[p10-max7.dun.ihug.co.nz]
[13:17] Ura_out (cschalle at c224s9h5.upc.chello.no) left irc: syntax error - user imploded
[13:37] Ow3n (owen at ti34a80-0852.bb.online.no) left irc: [x]chat
[14:28] Nick change: maYam_ZzZz -> maYam
[15:24] Nick change: wtay-zZz -> wtay
[17:38] wtay (wim at cable-195-162-214-190.upc.chello.be) got netsplit.
[17:39] wtay (wim at cable-195-162-214-190.upc.chello.be) returned to #gstreamer.
[17:39] wtay (wim at cable-195-162-214-190.upc.chello.be) got netsplit.
[17:39] wtay (wim at cable-195-162-214-190.upc.chello.be) returned to #gstreamer.
[17:39] #gstreamer: mode change '+o wtay' by ChanServ!s at ChanServ
[17:40] #gstreamer: mode change '-o wtay' by wtay!wim at cable-195-162-214-190.upc.chello.be
[17:44] Uraeus (cschalle at c224s9h5.upc.chello.no) joined #gstreamer.
[17:44] <Uraeus> hi
[17:45] <wtay> hello
[17:45] <wtay> I have to go for a while now though...
[17:45] Nick change: wtay -> wtay-away
[17:45] <Uraeus> hmm, and just as I was to brag about running XFS
[17:49] dobey (dobey at joined #gstreamer.
[17:50] <Uraeus> hi dobey
[17:51] <dobey> hi
[17:57] Uraeus (cschalle at c224s9h5.upc.chello.no) left irc: syntax error - user imploded
[18:04] <maYam> hi dobey
[18:04] <dobey> hi love
[18:13] Uraeus (cschalle at c224s9h5.upc.chello.no) joined #gstreamer.
[18:14] Nick change: maYam -> maYam_away
[18:28] Uraeus (cschalle at c224s9h5.upc.chello.no) left irc: Read error to Uraeus[c224s9h5.upc.chello.no]: EOF from client
[18:29] Uraeus (cschalle at c224s9h5.upc.chello.no) joined #gstreamer.
[18:34] <Uraeus> dobey: still around?
[18:37] <dobey> yeah
[18:37] <dobey> sup?
[18:38] <Uraeus> I did: mkfs -t xfs -f /dev/hda, by mistake, and fdisk /dev/hda gives me a blank, am I fucked at reboot?
[18:42] <Uraeus> dobey?
[18:42] <dobey> yes?
[18:42] <Uraeus> you say my question?
[18:42] <Uraeus> err..saw my question?
[18:42] <dobey> i have no idea
[18:43] <Uraeus> ah
[18:43] Action: Uraeus have an extremly bad feeling about this
[18:43] <dobey> hrmm
[18:44] <Uraeus> xfs seems to work well however
[18:58] Uraeus (cschalle at c224s9h5.upc.chello.no) left irc: syntax error - user imploded
[19:12] Zeenix (programmer at joined #gstreamer.
[19:12] <Zeenix> hi there
[19:12] <dobey> hola
[19:16] <Zeenix> gstreamer needs asound.so lib, where will i get that ?
[19:17] <Zeenix> i meant libasound.so
[19:22] Zeenix (programmer at left irc: Ping timeout for Zeenix[]
[19:24] Uraeus (cschalle at c224s9h5.upc.chello.no) joined #gstreamer.
[19:25] <Uraeus> hi
[19:58] <Uraeus> wtay-away: plugins to port to gstreamer?: http://mplayer.sourceforge.net/homepage/info.html
[19:59] Nick change: Uraeus -> Ura_away
[20:04] <dobey> sigh
[20:04] <dobey> screw that
[20:40] dobey (dobey at left irc: casa
[20:40] wtay-away (wim at cable-195-162-214-190.upc.chello.be) got netsplit.
[20:40] Ura_away (cschalle at c224s9h5.upc.chello.no) got netsplit.
[20:42] wtay-away (wim at cable-195-162-214-190.upc.chello.be) returned to #gstreamer.
[20:42] Ura_away (cschalle at c224s9h5.upc.chello.no) returned to #gstreamer.
[21:00] Nick change: wtay-away -> wtay-snooker
[21:22] hadess (hadess at pc121-gui14.cable.ntl.com) joined #gstreamer.
[21:28] <hadess> hi "lots of ppl away" gang
[21:32] Nick change: taazzzz -> taaz
[21:32] <hadess> heya taaz
[21:32] <taaz> hey
[22:16] steveb (steveb at node1ee06.a2000.nl) left irc: Ping timeout for steveb[node1ee06.a2000.nl]
[22:34] ajzzzz (ajmitch at p10-max7.dun.ihug.co.nz) joined #gstreamer.
[22:35] Nick change: ajzzzz -> ajmitch
[23:10] Nick change: Ura_away -> Uraeus
[23:10] <Uraeus> hi gang
[23:10] <hadess> wb Uraeus
[23:11] <Uraeus> not much noise here today I see :)
[23:12] <hadess> nah, they're all away =)
[23:13] <Uraeus> hadess do you want to see a screenshot of abiword for MacOS X?
[23:13] <hadess> why not :)
[23:13] <Uraeus> ok, I dcc it to you
[23:15] <Uraeus> hadess: like it :)
[23:15] <hadess> needs some more work i guess <g>
[23:17] <ajmitch> hi Uraeus
[23:17] <ajmitch> hi hadess
[23:17] <hadess> heya ajmitch
[23:17] <Uraeus> hi ajmitch
[23:18] <Uraeus> I'll bet that I am the first one here to run linux with XFS :)
[23:19] <ajmitch> heh, i'm still using ext2
[23:19] <ajmitch> us debian people are so far behind ;)
[23:19] <hadess> i think i'll use reiserfs on my next machine
[23:20] <Uraeus> yeah, debian is kinda the distribution version of 80'ies music
[23:20] <hadess> Uraeus: not quite
[23:20] <Uraeus> hehe
[23:20] <hadess> apt-get install reiser-utils
[23:20] <hadess> and voila =)
[23:21] <Uraeus> well if you want a journalig filesystem designed by vodka drinking russian mafia types I guess debian can provide :)
[23:21] Action: ajmitch sees if gstmediaplay crashes before starting to look at it...
[23:21] <hadess> lol
[23:21] <ajmitch> hmm, segfault...
[23:22] <Uraeus> ajmitch: what did you try to do?
[23:22] <ajmitch> play an ogg file with gstmediaplay ;)
[23:22] <Uraeus> not sure if that works, try a mp3
[23:24] <ajmitch> bah, oggs used to work
[23:24] <ajmitch> nope, crashes on mp3 as well ;)
[23:24] <ajmitch> ** CRITICAL **: file gstplay.c: line 138 (gst_play_init): assertion `priv->video_show != NULL' failed.
[23:25] <Uraeus> hmm
[23:25] <ajmitch> b0rken ;)
[23:26] <hadess> somebody should setup a tinderbox for gstreamer...
[23:26] <hadess> making hourly builds
[23:26] <Uraeus> ajmitch: do you supply the filename on the commandline or try to use the file selector?
[23:26] <ajmitch> yeah
[23:27] <ajmitch> Uraeus: tried with both
[23:27] Action: ajmitch runs cvs update
[23:27] <ajmitch> i recompiled before the cvs update this morning, so the source is about a day & 1/2 old
[23:28] <ajmitch> i remember seeing omega_ talking about some crashes, too....
[23:28] <Uraeus> have you guys seen this: http://mplayer.sourceforge.net/homepage/info.html
[23:28] <Uraeus> I tested it and it is quite nice, probably a good source for plugin code
[23:29] Action: ajmitch looks
[23:29] <Uraeus> have asf support which we lack if I am not mistaken
[23:29] <Uraeus> and have subtitle support for DVD
[23:29] <ajmitch> GPL?
[23:29] <Uraeus> yes, so that might be a problem
[23:29] <Uraeus> since we probably want LGPL
[23:29] <ajmitch> why?
[23:30] <Uraeus> well, cause if GStreamer is going to provide a multimedia framework for GNOME then we would want also commerical apps to be able to build on top
[23:30] <hadess> Uraeus: most of the plugins retain their original copyrights
[23:31] <Uraeus> and all other major libs are LGPL
[23:31] <Uraeus> of course it could be left to the discussion of wether you can have a LGPL framework with GPL plugins
[23:31] <ajmitch> yes, but for plugins?
[23:32] <ajmitch> ask RMS
[23:32] <ajmitch> ;)
[23:32] <hadess> in the end you only link with gstreamer, not the plugins...
[23:32] <Uraeus> hadess: ok so if I make closed source app XXX on top of GStreamer, what happens when someone uses my app to play a mediafile which needs a GPL plugin?
[23:33] <Uraeus> I think there are issues here cause this was the major obstacle for not integrating gimp-print and gnome-print
[23:33] <Uraeus> of course I don't think anyone would really bother to object, except maybe RMS
[23:34] <Uraeus> so I guess what I am saying is that, if possible LGPL or BSD plugins are to be prefered if possible, but if that is not an option: what the heck
[23:35] <ajmitch> Uraeus: you're sounding like Njaard - he wants any plugins but GPL-licensed ones ;)
[23:35] <ajmitch> but that's just cos he hates the GPL ;)
[23:35] <hadess> that's stupid
[23:35] omega_ (omega at omegacs.net) joined #gstreamer.
[23:35] <hadess> yo omega_
[23:35] <omega_> yo
[23:35] <ajmitch> hey omega_
[23:36] <ajmitch> gstmediaplay seems to be crashing a bit here ;)
[23:36] <Uraeus> omega_: and we have decided that allmost all our plugins are illegal
[23:36] <omega_> neat
[23:36] <Uraeus> omega_:http://mplayer.sourceforge.net/homepage/info.html
[23:36] <ajmitch> and Uraeus wants me to hack on gstmediaplay as well :)
[23:37] <Uraeus> omega_: maybe some useable plugin code there (link above)
[23:37] <omega_> yup
[23:38] <Uraeus> ajmitch: well, because it would be great to have a nice gstmediaplay for the 0.2.0 release I had to ask the best hacker here to do it....
[23:38] Action: Uraeus is good at sucking up
[23:38] <hadess> Uraeus: stop brown-nosing
[23:38] <Uraeus> hehe
[23:38] <ajmitch> haha
[23:38] <ajmitch> ajmitch at ajmitch:~/cvs/gnue/geas$ gstmediaplay /music/Legal/Live/Live\ -\ Run\ To\ The\ Water.mp3 
[23:38] <ajmitch> INFO: Initializing GStreamer Core Library
[23:38] <ajmitch> INFO: CPU features: (808029bf) MMX 3DNOW 
[23:38] <ajmitch> ** CRITICAL **: file gstplay.c: line 138 (gst_play_init): assertion `priv->video_show != NULL' failed.
[23:38] <ajmitch> ajmitch at ajmitch:~/cvs/gnue/geas$ 
[23:39] <ajmitch> and it dies...
[23:39] <ajmitch> run from gdb and go insane? ;)
[23:39] <hadess> haha i can't believe you have a "Legal" directory for your mp3s
[23:40] <hadess> what's in the "Illegal" directory then ?
[23:40] Action: hadess gets a pen and paper
[23:40] Action: omega_ is going through is mail
[23:40] <hadess> +h
[23:41] Uraeus (cschalle at c224s9h5.upc.chello.no) left irc: Read error to Uraeus[c224s9h5.upc.chello.no]: Connection reset by peer
[23:41] <ajmitch> hadess: wouldn't you like to know? ;)
[23:41] <hadess> ajmitch: heh
[23:41] <ajmitch> Program received signal SIGSEGV, Segmentation fault.
[23:41] <ajmitch> [Switching to Thread 1024 (LWP 22702)]
[23:41] <ajmitch> 0x4078d728 in gst_static_autoplug_to_render () from /opt/gnome/lib/gst/libgststaticautoplugrender.so
[23:41] <ajmitch> from running in gdb
[23:41] Uraeus (cschalle at c224s9h5.upc.chello.no) joined #gstreamer.
[23:42] <ajmitch> surely we have some fine people who can solve this problem without me thinking? ;)
[23:42] <Uraeus> hmm, ok red-carpet locked up my system, but luckily I run a commercial grade distro with a journalig file system from a renowned Unix vendor :)
[23:42] <hadess> lol
[23:43] <ajmitch> haha
[23:43] <Uraeus> so no problems here :)
[23:43] <hadess> and i run a non-journaling file system from a renouned Linux hacker =)
[23:44] <ajmitch> only time i had to manually fix it was when the power failed during a big (KDE) compile ;)
[23:45] Action: hadess gives a try at making a first working rhythmbox plugin
[23:45] <Uraeus> hadess: you mean the one created by that hacker working for the commercial grade distro?
[23:45] Action: ajmitch tries to compile gstreamer again
[23:46] <hadess> Uraeus: i don't think Remy Card works for Redhat (no, Stephen Tweedie isn't the creator of ex2)
[23:46] <ajmitch> bbl, time to eat food ;)
[23:46] Action: Uraeus hides in shame
[23:46] <hadess> but stephen is a very nice guy =)
[23:47] <hadess> and he is the maintainer with Theodore T'So
[23:49] Action: hadess verifies what he said
[23:49] <Uraeus> hadess: I talked to Arik Devens yesterday, who was a Ximian intern last year and now is a Mozilla intern
[23:49] <hadess> Uraeus: and ?
[23:49] <omega_> he's building a media player
[23:49] <Uraeus> and he is working on GNOME & Mozilla integration
[23:50] <omega_> cool
[23:50] <hadess> omega_: oh, that's the guy who was here yesterday, right ?
[23:50] <Uraeus> and he might make the mozilla gstreamer plugin we been missing
[23:50] <Uraeus> yes
[23:50] <hadess> "Remy Card, Second extended file system designer and developer"
[23:50] <omega_> Uraeus: yeah
[23:51] Action: omega_ is watching an unencrypted DVD on a machine that's too slow
[23:51] <omega_> but it's coming right off the disc
[23:51] <Uraeus> just rub it in hadess, ok you where right for once :)
[23:52] <hadess> Uraeus: but stephen hacks on filesystems in general, and theodore t'so is the e2fsck maintainer
[23:52] <hadess> omega_: coool
[23:53] <hadess> Uraeus: no, i just wanted to make sure i was telling the truth...
[23:57] <omega_> bleagh, one or both ofmy 80gb drives just reached max mount count
[23:57] <hadess> omega_: what would it be that's so slow ?
[23:57] <hadess> huh...
[23:57] <omega_> I'm gonna RMA one of them cause it's making lots of noise, then I can switch filesystems on both of them
[23:57] <omega_> hadess: the fact that it's 80gb
[23:57] <hadess> omega_: no, about the dvd playing
[23:57] <omega_> 500MHz machine trying mpeg2dec and ac3dec
[23:58] <omega_> I need a 525 or maybe 550MHz machine,and this is a laptop....
[23:58] <hadess> hmmm...
[23:58] <omega_> so I'm gonna try to get my desktop (Duron 750) to run it
[23:58] <Uraeus> omega_: get the SGI redhat CD and you get up and running on XFS in no time, no more max mount counts :)
[23:58] <omega_> unfortunately the DVD is a not-very-good TV transfer
[23:58] <omega_> the 80gb drives just have data, not boot
[23:59] <omega_> so I add the third RMA'd drive, newfs with xfs, copy data
[23:59] <omega_> should take "only" 1hr to copy a drive, but that's assuming 25MB/sec sustained on both drives (differnet ATA/100 channels, but the VFS isn't that fast)
[00:00] --- Sun May  6 2001
[00:02] <ajmitch> back
[00:02] <ajmitch> damn, gstreamer still compiling
[00:03] Action: ajmitch uses xmms for music instead
[00:03] <Uraeus> what...mp3.com is f*cking assholes, I enter some data and choose 'no
[00:03] <Uraeus> I dont want info from pfeizer
[00:03] <Uraeus> get a message that I missed a fiedl
[00:03] <Uraeus> err field
[00:03] <Uraeus> clicks ok again and notice that they have also reset the pfeizer field to yes
[00:03] Action: ajmitch turns xmms off & uses stereo
[00:05] greg_ (greg at home.sente.pl) joined #gstreamer.
[00:06] <ajmitch> btw, any ideas on why gstmediaplay segfaulted? ;)
[00:07] <omega_> it probably touched NULL
[00:07] <Uraeus> ajmitch: do you have hermes installed?
[00:07] Action: ajmitch looks
[00:08] <ajmitch> nope
[00:08] <Uraeus> ajmitch: ok gstmediaplay crashed all the time for me when I didn't have that installed
[00:10] <ajmitch> installing...
[00:10] <Uraeus> ajmitch: I can dcc you the source.tar.gz file if you want
[00:10] <Uraeus> ajmitch: remember to make, make install on the colorspace element
[00:10] <ajmitch> nah, apt-get will do thanks ;)
[00:10] <ajmitch> colorspace element?
[00:10] <omega_> make install??
[00:11] <Uraeus> sorry filter not element
[00:11] <Uraeus> omega_: of course?
[00:12] <Uraeus> omega_: why wouldn't I do make install?
[00:12] <omega_> um, if you're not building apps with it, just running stuff that comes in the main package, you do not want to make install
[00:12] <omega_> that just confuses things by having two sets of plugin available
[00:12] <omega_> and wastes a lot of time and disk space
[00:12] <Uraeus> you just run gstreamer from the source catalog?
[00:13] <Uraeus> err directory
[00:13] <omega_> from the builddir, yes
[00:13] <ajmitch> i consistently get this in gdb:
[00:13] <ajmitch> 0x4078d728 in gst_static_autoplug_to_render () from /opt/gnome/lib/gst/libgststaticautoplugrender.so
[00:13] <omega_> if you use autogen, it sets all the options you need to run from builddir
[00:13] <ajmitch> no matter what i try to play (mp3, ogg, mpeg)
[00:13] <omega_> doesn't surprise me, the autoplugger needs a rewrite
[00:13] <ajmitch> blame wtay then? ;)
[00:14] <hadess> heh
[00:14] <omega_> of course
[00:14] <ajmitch> cool, it always helps to be able to blame someone
[00:17] <Uraeus> omega_: heard anything more from PD?
[00:17] <omega_> nope
[00:19] <Uraeus> heh, you scared him silent
[00:20] <omega_> heheh
[00:20] <omega_> ooh, I sustained 21MB/sec copying from drive to drive
[00:20] <ajmitch> wow
[00:20] <omega_> so 42MB/sec over the hpt370 controller
[00:20] <ajmitch> i get about 2MB/sec ;)
[00:20] Action: ajmitch has a crap computer
[00:21] <ajmitch> so arik will be doing a gstmediaplay replacement??
[00:21] <omega_> he's working on a different player, but I dunno if it'll be a replacement
[00:21] <ajmitch> ok
[00:21] <omega_> the goal is to get a library that does media player stuff
[00:21] <ajmitch> so you still want people to hack gstmediaplay and improve it? ;)
[00:21] <omega_> the program is not relevant to that, except that it should use as many features as we can think of
[00:22] <ajmitch> k
[00:22] <omega_> well, not just yet, we need to get autoplug working again
[00:22] <ajmitch> you want libgstplay to be as good as possible
[00:22] <omega_> but then developing a good API and a solid libgstplay.so is critica
[00:22] <ajmitch> ok
[00:22] <omega_> from that, the bonobo component can be 100 lies
[00:22] <omega_> er, lines
[00:23] <ajmitch> depends
[00:23] <omega_> yeah, on whether it's bonobo-media or not
[00:23] <ajmitch> depends on how complicated you want the component UI to be, too
[00:23] <omega_> if it's -media, the gui is someone else's job
[00:23] <ajmitch> but that shouldn't matter much ;)
[00:23] <ajmitch> yeah
[00:32] <ajmitch> how can one stream mp3s, etc from an http server using a nice (gstreamer) GUI program? 
[00:33] <ajmitch> is it something that needs added to gstmediaplay? ;)
[00:33] <omega_> you mean a shoutcast/icecast client?
[00:33] <omega_> that just means a different URI handed to libgstplay
[00:34] <ajmitch> yeah
[00:34] <Uraeus> hmm, worrysome, seems that LGP has failed to give the money they where supposed raise for Gnumatic, i hope Ximian has gotten all the money they where supposed to already
[00:34] <ajmitch> can the current gstmediaplay handle URIs?
[00:34] <omega_> lgp?
[00:34] <omega_> ajmitch: dunno
[00:34] <omega_> doubt it
[00:35] <Uraeus> Linux Global Partners, the main VC company behind Gnumatic and Ximian (they also failed to deliver promised money to MetroLink)
[00:35] <omega_> uh oh
[00:36] <hadess> huh, bad shit
[00:36] <Uraeus> lucklily I think Ximian already has recieved the money they where supposed to, but I am not 100% sure
[00:36] Action: hadess listens to brit pop
[00:42] Action: omega_ listens to norsk pop
[00:42] <omega_> ;-)
[00:42] <hadess> heh
[00:42] <omega_> and write the autoplugcache
[00:43] <Uraeus> omega_: I knew I would get you in touch with your roots in the end ;)
[00:43] <omega_> doh
[00:44] Action: ajmitch sees whether linux will die horribly on testing plex86
[00:44] <hadess> and i'm writing the mp3player component of rhythmbox =)
[00:45] <omega_> static gst pipeline or are you gonna autoplug it?
[00:45] <hadess> and right now, it doesn't look too good :P
[00:45] <omega_> I hope you're gonna support arbitrary media types in rhythmbox...
[00:45] <hadess> i hope to add mime-types definition to see what each component supports
[00:46] <hadess> only audio in any case..
[00:46] <omega_> cause I wanna use it to rip my CDs to vorbis
[00:46] <ajmitch> hehe, i can use konqueror for that ;P
[00:46] <hadess> oh, yeah, that's planned, just a simple prefs to choose
[00:47] <Uraeus> ajmitch: well that app is tainted, use grip while waiting :)
[00:47] <ajmitch> Uraeus: grip crashes consistently ;)
[00:47] <Uraeus> huh? never for me
[00:47] <hadess> damn, audio performance on this box is horrible, it skips constantly
[00:47] <ajmitch> crashes every time here
[00:48] <ajmitch> is grip still maintained?
[00:48] <Uraeus> not sure, been a while since a new release was made, but Mike has been around for a long time so I guess he is just taking a breather
[00:49] <ajmitch> grip hasn't been gstreamer-assimilated yet?
[00:50] <Uraeus> not yet ;)
[00:51] <ajmitch> wow, evolution is taking more memory than X & mozilla combined...
[00:51] <omega_> oooh
[00:52] Action: ajmitch keeps waiting for gstreamer to finish compiling
[00:56] <hadess> "warning: passing arg 1 of `e_tree_memory_callbacks_new' from incompatible pointer type"
[00:56] <hadess> is that the first or the second arg that's the pb ?
[00:57] Action: hadess pokes around
[00:57] <hadess> omega_: /
[00:59] greg_ (greg at home.sente.pl) left irc: [x]chat
[01:21] Uraeus (cschalle at c224s9h5.upc.chello.no) left irc: Client Exiting
[01:21] <omega_> hadess: ?>
[01:22] <hadess> omega_: was about this argument thingy, but i found out, thanks
[01:23] <ajmitch> hmm
[01:26] <omega_> the latest BIOS for the KT7-RAID allows me to finally use both video cards *and* the hpt370 !
[01:27] <hadess> lucky chap :)
[01:28] <ajmitch> hehe
[01:28] <ajmitch> i wish i had that sorta hardware
[01:53] greg_ (greg at home.sente.pl) joined #gstreamer.
[01:56] Action: ajmitch looks at glade in awe & confusion as he sees that you can embed bonobo objetcs, include a GstMediaPLay object ;)
[01:59] <omega_> oooh
[01:59] <ajmitch> yeah, i compiled the stuff in cvs
[02:00] <ajmitch> so i also can put in all these evolution & nautilus components ;)
[02:00] <omega_> neat
[02:00] Action: ajmitch is just playing with glade and looking at gstmediaplay code
[02:01] <omega_> how do you do this?
[02:01] <ajmitch> hmm, am using glade-gnome from sid
[02:01] <omega_> no
[02:01] <ajmitch> it has an extra tab on the palette called controls
[02:01] <omega_> er, ok
[02:01] <omega_> hmmm
[02:13] <ajmitch> you don't mind me playing with the gstmediaplay UI a bit, do you? ;)
[02:13] <omega_> go for it
[02:13] <omega_> the wider range of ideas we have the better
[02:13] <omega_> since I'd like to make the GUI a project of the GNOME Usability group at some point
[02:14] <ajmitch> ooh, ok
[02:14] <ajmitch> how hard to add a playlist, i wonder? ;)
[02:14] <omega_> depends
[02:14] <omega_> I tend to think a playlist should be a bonobo component with a specific IDL
[02:14] <ajmitch> ok
[02:14] <omega_> because then it can be used for any program
[02:14] <omega_> and it can be significantly upgraded over time
[02:52] greg_ (greg at home.sente.pl) left irc: Read error to greg_[home.sente.pl]: Connection reset by peer
[02:53] Nick change: wtay-snooker -> wtay
[02:53] <wtay> hello
[02:53] <ajmitch> hey wtay
[02:53] <ajmitch> gstmediaplay is having some autoplug pronlems ;)
[02:54] <wtay> hmmm
[02:55] <ajmitch> i'll try and play something and show you the crash
[02:55] arik (arik at adsl-64-174-42-2.dsl.snfc21.pacbell.net) joined #gstreamer.
[02:55] <arik> lo all
[02:55] <ajmitch> hi arik
[02:55] <wtay> yo
[02:56] <ajmitch> Program received signal SIGSEGV, Segmentation fault.
[02:56] <ajmitch> [Switching to Thread 1024 (LWP 21208)]
[02:56] <ajmitch> flxdec_typefind (buf=0x81695a8, private=0x80e8488) at flx_decoder.c:120
[02:56] <ajmitch> 120       if ((data[4] == 0x11 || data[4] == 0x12
[02:56] <ajmitch> hmm, it was an autoplug error before...
[02:57] <ajmitch> [10:22:09] <ajmitch> 0x4078d728 in gst_static_autoplug_to_render () from /opt/gnome/lib/gst/libgststaticautoplugrender.so
[02:57] <wtay> ajmitch: type gstreamer-inspect colorspace
[02:57] <ajmitch> ajmitch at ajmitch:~/cvs/gstreamer/gstplay$ gstreamer-inspect colorspace
[02:57] <ajmitch> INFO(21228:-1): Initializing GStreamer Core Library
[02:57] <ajmitch> INFO(21228:-1): CPU features: (808029bf) MMX 3DNOW 
[02:57] <ajmitch> ** WARNING **: gstplugin: registry needs rebuild: run gstreamer-register
[02:57] <ajmitch> no such element or plugin 'colorspace'
[02:57] <ajmitch> ajmitch at ajmitch:~/cvs/gstreamer/gstplay$ 
[02:58] <wtay> ajmitch: get Hermes and rebuild plugins/filter/colorspace
[02:58] <ajmitch> ugh, it requires that?
[02:58] <wtay> yup
[02:58] <ajmitch> hermes in sid not good enough for ya??
[02:58] <ajmitch> ;)
[02:58] <wtay> ajmitch: fine for me (I use it) :-)
[02:59] <ajmitch> wtay: then why not for me?
[02:59] <arik> so i think i will give up on my own player and just work with gstplay instead
[02:59] <ajmitch> hmm
[02:59] <ajmitch> i installed hermes after i rebuilt once
[02:59] <ajmitch> i should rebuild that dir again
[02:59] <wtay> yes
[02:59] <wtay> arik: gstplay is pretty clean IMO, so...
[03:00] <ajmitch> what will you hack, gstmediaplay, or the library behind it?
[03:00] <arik> wtay: i have some issues with it but it seems like a better idea to rework it then start *another* codebase
[03:00] <arik> gstmediaplay
[03:00] <ajmitch> oh, ok
[03:00] <wtay> arik: yup
[03:01] <arik> wtay: plus mine doesn't work right ;-)
[03:01] Action: ajmitch hacked gstmediaplay a little this morning
[03:01] <wtay> I think error recovery is the most important issue now
[03:02] <arik> who owns gstmediaplay?
[03:02] <wtay> arik: root :-)
[03:02] <ajmitch> i added an open location entry to the menu, is that needed?
[03:02] <arik> wtay: no i mean the codebase ;-)
[03:02] <arik> ajmitch: i was planning on doing that
[03:02] <ajmitch> ok ;)
[03:03] <wtay> arik: you can say that I own it then...
[03:03] <arik> wtay: i would be interested in doing it at some point (although for the moment i am planning on sending patches)
[03:03] <ajmitch> hmm
[03:04] <ajmitch> colorspace plugin installed, it still crashes
[03:04] <wtay> arik: good
[03:04] <wtay> ajmitch: details?
[03:04] <ajmitch> Uraeus has a list of things he wants in it before 0.2.0
[03:04] <ajmitch> wtay: trying to load it in gdb... ;)
[03:05] <arik> course atm gstmediaplay prints out gobs of "cothread: can't switch to NULL cothread!
[03:05] <ajmitch> same thing from gdb
[03:05] <arik> " errors
[03:05] <ajmitch> it does?
[03:05] <wtay> arik: in INCSCHED1?
[03:05] <arik> wtay: *nod*
[03:05] <arik> it plays though which is good
[03:06] <ajmitch> hmm, i am playing with it from HEAD
[03:06] <wtay> arik: hmm, it does? that news..
[03:06] <wtay> s/that/that's
[03:06] <arik> wtay: well it does for me anyway
[03:06] <wtay> cool
[03:06] <wtay> arik: audio/video?
[03:07] Action: wtay cvs updates INCSCHED1
[03:07] <arik> wtay: video at least
[03:07] <wtay> neat..
[03:07] <arik> audio causes it to just segfault for me
[03:08] <wtay> oh
[03:08] <arik> oh wait
[03:08] <arik> nm
[03:08] <arik> i know why that is happening
[03:08] <arik> no
[03:08] <arik> no i don't ;-)
[03:09] <wtay> hmm, what audio are you trying to play?
[03:09] Action: omega_ is building the autoplug-cache
[03:10] <arik> wtay: mp3
[03:10] Action: ajmitch wonders how gstmediaplay can still crash in the file when the flx plugin has been removed ;)
[03:10] <arik> oh and it also seems to lock up once it finishes playing a file
[03:10] <arik> lots of fun problems to solve ;-)
[03:10] <wtay> arik: most of those problems are core related...
[03:11] <arik> wtay: *nod*
[03:11] <wtay> arik: aka not implemented..
[03:11] <arik> wtay: hehe
[03:11] <taaz> you all look idle... someone please make 'make dist' work on a debian machine. arts paths are all goofed. thanks ;)
[03:12] Nick change: taaz -> taaz-away
[03:12] <omega_> pffff
[03:12] <ajmitch> make dist? hmm
[03:12] <wtay> make coffee sounds better...
[03:13] <omega_> wtay: don't you mean `make cigarette` ?
[03:13] <wtay> omega_: that too :-)
[03:13] Action: ajmitch is puzzled
[03:13] <ajmitch> Program received signal SIGSEGV, Segmentation fault.
[03:13] <ajmitch> [Switching to Thread 1024 (LWP 21788)]
[03:13] <ajmitch> flxdec_typefind (buf=0x8169ac0, private=0x80e9dd8) at flx_decoder.c:120
[03:13] <ajmitch> 120       if ((data[4] == 0x11 || data[4] == 0x12
[03:13] <ajmitch> this means it's trying to ask the flxdec plugin what it does, right?
[03:14] <wtay> ajmitch: it asks the flx plugin if it can handle the data in the buffer (probably NULL)
[03:14] <ajmitch> then why does it do this after i rm'ed the flx plugin from /opt/gnome/lib/gst? ;)
[03:14] <wtay> ajmitch: because it's still in the build dir
[03:15] <arik> omega_: lo
[03:15] <omega_> yo
[03:15] <ajmitch> wtay: i'm not running it from the build dir tho, does that matter?
[03:16] <wtay> ajmitch: I think it does...
[03:17] <ajmitch> true, it does (looked with strace)
[03:18] <ajmitch> now to figure out why it crashes still ;)
[03:19] <ajmitch> wee, it crashes elsewhere now!! ;)
[03:19] <ajmitch> 0x40956fe1 in vorbis_typefind (buf=0x816a7e8, private=0x80e83f8) at vorbis.c:85
[03:19] <ajmitch> 85        gulong head = GULONG_FROM_BE (*((gulong *)GST_BUFFER_DATA (buf)));
[03:19] <ajmitch> (trying to play an ogg)
[03:19] <omega_> ajmitch: maybe crashing in a different thread
[03:19] <ajmitch> yeah
[03:20] <wtay> ajmitch: weird.. never seen that one...
[03:21] <ajmitch> did the same when trying to play an mp3
[03:22] <ajmitch> no recent changes to HEAD that could have broken this?
[03:23] <wtay> ajmitch: nope, works fine here
[03:23] <ajmitch> hmm, any ideas? ;)
[03:24] <ajmitch> i'm sure its not my changes that did anythingm since they are totally unrelated ;)
[03:24] <wtay> no idea..
[03:26] <wtay> ajmitch: what did you change?
[03:26] <ajmitch> wtay: only added a callback for popping up an 'open location' window, and the code to pop that up
[03:28] Nick change: maYam_away -> maYam_sleep
[03:28] <ajmitch> wtay: being called away soon? ;)
[03:28] <wtay> ajmitch: most probably :-)
[03:29] <maYam_sleep> oh.. wta-ay...
[03:29] <maYam_sleep> nah ;)
[03:30] <wtay> maYam_sleep: comming! just a few more seconds...
[03:36] <omega_> ick
[03:36] <omega_> mplayer make has a serious bug
[03:36] <ajmitch> hehe
[03:37] <omega_> rather, X11 makedepend does
[03:39] <arik> weird weird weird
[03:39] <arik> is there something strange about how gstmediaplay handles its glade file or do i not remember how this works?
[03:40] <arik> cause i added some menu items and i don't see them
[03:40] <omega_> hmmm
[03:40] <ajmitch> no, it works fine for me
[03:40] <ajmitch> hey, gstmediaplay works (badly) if i rerun gstreamer-register?
[03:41] <omega_> makes sense
[03:41] <arik> ajmitch: hmm
[03:41] <omega_> the registry handling needs a rework
[03:41] <arik> ajmitch: not for me
[03:41] <arik> oh i bet i know what is happeneing
[03:41] <arik> it is using the installed glade file not the local one
[03:41] <omega_> yup
[03:41] <ajmitch> oh, ok
[03:41] <omega_> thus: don't install gstreamer ;-)
[03:41] <ajmitch> i installed the glade file ;)
[03:41] <omega_> hack your makefile to use the builddir
[03:42] <arik> ah much better
[03:42] <omega_> -I...gstroot/ ...gstroot/gst/libgst.la
[03:42] <ajmitch> ah...
[03:42] <omega_> libtool gcc
[03:42] <arik> omega_: heh
[03:42] <ajmitch> gstmediaplay works now
[03:42] Action: ajmitch listens to DC Talk - Mind's Eye
[03:42] <ajmitch> sounds good ;)
[03:43] <ajmitch> arik: what are you hacking into the glade file?
[03:43] <ajmitch> i'll send you my changes if you want, so you can take a look
[03:43] <arik> ajmitch: i am adding support to present the movie at double and fullscreen sizes
[03:44] <ajmitch> arik: ok, i've added an open location dialog, and for some reason, a playlist window (not used)
[03:44] <arik> ajmitch: i have some ideas about the best way to do playlist
[03:44] <omega_> are you both hacking on gstmediaplay?
[03:44] <ajmitch> seems like it
[03:44] <omega_> arik: taaz claims to have some really good ideas on playlists too
[03:44] <arik> omega_: i decided it was prob best not to start a new project
[03:44] <omega_> heh
[03:44] <ajmitch> Uraeus told me some stuff needed done ;)
[03:45] chillywilly (baumannd at d120.as29.nwbl0.wi.voyager.net) joined #gstreamer.
[03:45] <ajmitch> hey chillywilly
[03:45] <chillywilly> hey
[03:45] <omega_> yeah, the player needs a lot of work still
[03:45] <arik> omega_: i mentioned to wtay that i would be intrested in maintaining it at some point
[03:45] <omega_> cool
[03:45] <arik> i wonder how libglade lets you do construct_menus_with_data
[03:45] <ajmitch> heh
[03:46] <arik> ah there we go
[03:46] <arik> neeto
[03:46] <ajmitch> time to hack libglade? ;)
[03:46] <omega_> ajmitch: down, boy
[03:46] <ajmitch> ;)
[03:46] <omega_> ajmitch: good doggie ;-)
[03:46] <arik> ajmitch: nah, it's in there
[03:46] Action: ajmitch starts gnawing ankles
[03:46] Action: omega_ runs
[03:46] Action: omega_ runs faster
[03:47] Action: chillywilly kicks ajmitch
[03:47] Action: omega_ watches the channel degrade before his eyes
[03:47] <chillywilly> damn dogs
[03:47] Action: omega_ participates ;-(
[03:47] <ajmitch> omega_: who let chillywilly in? ;)
[03:47] <omega_> lilo
[03:47] <chillywilly> yeah blame it on him
[03:48] <ajmitch> hey, can i post a nice big stack trace? ;)
[03:48] <omega_> um
[03:48] <ajmitch> only about 35 lines
[03:48] <omega_> only post what's relevant, to the first gst_ functon
[03:48] <omega_> eek
[03:48] <omega_> only the top is usually relevant
[03:49] <ajmitch> ***** GStreamer ERROR ***** in file gstbin.c at gst_bin_remove:215
[03:49] <ajmitch> Element: /main_thread/main_bin          Object: /typefind
[03:49] <ajmitch> Error: no such element in bin
[03:49] <ajmitch> ***** attempting to stack trace.... *****
[03:49] <ajmitch> #0  0x4065cb47 in fork () from /lib/libc.so.6
[03:49] <ajmitch> #1  0x400965ee in fork () from /lib/libpthread.so.0
[03:49] <ajmitch> #2  0x400776ca in g_on_error_stack_trace () from /usr/lib/libglib-1.2.so.0
[03:49] <ajmitch> #3  0x40029853 in gst_default_error_handler (file=0x4003f8c2 "gstbin.c", 
[03:49] <ajmitch> #4  0x4001fe33 in gst_bin_remove (bin=0x80e9700, element=0x81fa440)
[03:49] <ajmitch> #5  0x405b35f1 in gst_play_typefind (bin=0x80e9700, element=0x81fa318)
[03:49] <ajmitch> #6  0x405b3853 in gst_play_set_uri (play=0x80e9280, 
[03:49] <ajmitch> that should be enough
[03:49] <ajmitch> ;)
[03:49] <ajmitch> trying to select another song to play
[03:49] <omega_> um, neat
[03:49] <omega_> it means that the player isn't using the typefind stuff very well
[03:49] <ajmitch> not enough info, i bet ;)
[03:49] <omega_> it's plenty
[03:49] <ajmitch> ok
[03:50] <omega_> to tell that the code will be tossed shortly <g>
[03:50] <ajmitch> so we just have to hack the player to pieces? ;)
[03:50] <omega_> well, no.  work on other stuff until dynamic autoplug exists ;-)
[03:50] <ajmitch> lol
[03:50] <ajmitch> yeah, i was playing an ogg then tried to load an mp3 ;)
[03:51] <ajmitch> hey, xmms can play both, why can't this? ;)
[03:51] <omega_> cause xmms is lame
[03:51] <chillywilly> oooh, thems fighting words
[03:51] <chillywilly> you gonna take that omega_?
[03:51] <hadess> heh
[03:52] <omega_> um, /me thwacks ajmitch
[03:52] Action: ajmitch ducked
[03:52] Action: omega_ kicks ajmitch
[03:52] <ajmitch> hmm
[03:52] <omega_> hehehe
[03:52] <ajmitch> sure you want to treat users like this? ;)
[03:52] <omega_> ah
[03:52] <omega_> sure ;-)
[03:52] <hadess> ajmitch: you're not a user, you're a developer
[03:53] <chillywilly> blah your just a dog remember
[03:53] Action: hadess kicks ajmitch
[03:53] <ajmitch> hadess: i am ;)
[03:53] <ajmitch> i am?
[03:53] <omega_> heh
[03:53] Action: ajmitch looks at his desktop for signs of development
[03:53] <hadess> heh
[03:53] <chillywilly> little ankle biter ;)
[03:53] <ajmitch> hmm, xchat, mozilla, evolution, xmms
[03:53] <ajmitch> wait - emacs, glade ;)
[03:54] <arik> hmm
[03:54] Action: omega_ wonders how many more hints ajmitch needs than emacs and glade ;-)
[03:54] <ajmitch> hey, emacs is useful for more than development
[03:54] <omega_> yes, but if you look carefully, you'll see a file ending in a .c open
[03:54] <arik> needing a gstplayprivate in callbacks.c bad
[03:54] <chillywilly> emacs can whipe your ass for you too
[03:54] <ajmitch> chuck a kernel with it and you have an OS; )
[03:54] <omega_> that makes it a development OS, er, editor <g>
[03:54] <arik> have better idea
[03:55] Action: omega_ hands ajmitch 80MB more for emacs
[03:55] <chillywilly> vim is for real men
[03:55] Action: ajmitch gets that 80MB by rm-rf'ing gstreamer ;)
[03:55] <wtay> arik: callbacks.c is bad allthogether...
[03:55] <arik> wtay: well sort of
[03:56] <arik> wtay: it makes sense to have it if it was actually being used
[03:56] <hadess> wb wtay
[03:56] <omega_> ajmitch: only 80? and I meant RAM
[03:56] <arik> wtay: at the moment most of what it would do it isn't doing
[03:56] <ajmitch> don't diss emacs
[03:56] <wtay> hadess: dude :)
[03:56] <arik> wtay: you could easily fold it into gstmediaplay but it might make more sense to pull stuff out of gstmediaplay and into callbacks
[03:57] <wtay> arik: or make it more modular, like menu handling, toolbar handling etc in different files..
[03:57] Action: ajmitch has to leave for a bit, bbl
[03:57] <arik> wtay: that's another way to go yeah
[03:58] Nick change: ajmitch -> ajbusy
[03:58] Action: wtay has to sleep..
[03:58] <hadess> huh, 4am...
[03:59] Action: omega_ is building the foundation of the dynamic autoplugger
[03:59] <wtay> cool
[03:59] <wtay> omega_: did you check out the gstoss stuff i did?
[03:59] <omega_> not yet
[04:01] Action: hadess is making some new l33t button widget
[04:01] Action: wtay created an mp3 of the sound of the game he was playing in x64...
[04:01] <omega_> x64?
[04:01] <wtay> vice
[04:01] <omega_> huh?
[04:01] <hadess> c64 ?
[04:02] <wtay> C64 emulator :-)
[04:02] <arik> hadess: heh
[04:02] <hadess> arik: heh, i was right =)
[04:02] <omega_> with gstoss ?
[04:02] <wtay> yup
[04:02] <omega_> wow
[04:02] <omega_> announce to the list
[04:02] <wtay> realplayer doesn't work
[04:02] <omega_> -devel
[04:02] <arik> later all
[04:02] arik (arik at adsl-64-174-42-2.dsl.snfc21.pacbell.net) left irc: leaving
[04:02] <chillywilly> realplayer is evil
[04:03] <wtay> it looks like it does strange things (maybe exec something with a custom env)
[04:03] <omega_> yeah, probably
[04:03] Action: hadess needs to finish the widget before going to bed
[04:03] <omega_> chillywilly: you don't know the half of it
[04:03] <chillywilly> well I'm going home
[04:03] <wtay> although I think audiooss.c can handle realplay (maybe an older version, I dunno)
[04:04] <omega_> hmmm, something to look ito
[04:04] <chillywilly> catcha later
[04:04] <wtay> chillywilly: cya
[04:04] chillywilly (baumannd at d120.as29.nwbl0.wi.voyager.net) left irc: later
[04:04] <omega_> l8r
[04:04] <wtay> omega_: I didn't try audiooss because it needs NAS or something
[04:04] <omega_> oooh, autoplugcache works
[04:04] <omega_> yeah, didn't build for me
[04:04] <wtay> hmm
[04:05] <omega_> autoplugcache is cool ;-)
[04:05] <wtay> coool
[04:07] <omega_> now we need to get object destruction to work....
[04:07] <omega_> but we can leave that until we move to gobject
[04:07] <omega_> we need to get autoplug rebuilt ASAP, then 0.2.0
[04:08] <omega_> maybe by next weekend ?
[04:08] <wtay> maybe I should try to look into the problems I still have with INCSCHED1
[04:08] <wtay> try to understand how INCSCHED1 works...
[04:09] <omega_> yeah
[04:09] <omega_> the mail I sent should be sufficient
[04:09] <wtay> I think I understand the thing, I just need to get familiar with the code
[04:09] <omega_> ok
[04:10] <wtay> omega_: you think you have enough info to create an autoplugger?
[04:10] <omega_> the incsched parts, yeah
[04:10] <omega_> update in gst/autoplug/ tomorrow
[04:11] <wtay> hmm, is autoplugcache the cache for previously autoplugged components or the buffer cache?
[04:11] <omega_> no, it's a buffer cache
[04:11] <wtay> ok
[04:11] <omega_> the test code constructs the pipeline, iterates for a while
[04:11] <omega_> then hits 'reset' on the cache, and plays again
[04:12] <omega_> the file plays the first few seconds, then cleanly starts over at reset
[04:12] <omega_> now I'm gonna do limitted autoplug
[04:12] <omega_> run the typefind element, check for mp3 or vorbis, play
[04:12] <wtay> cool
[04:12] <omega_> we may need to rethink the typefind API a little though
[04:12] <wtay> possibly yes
[04:12] <omega_> problem: if you get the first buffer
[04:12] <omega_> all typefinds fail
[04:13] <omega_> you want to send the second buffer, but there's no context for the typefind to know anything about it
[04:13] <omega_> so we can punt and simply construct a larger buffer
[04:13] <omega_> or rev the API (hard)
[04:13] <omega_> I'd tend to punt
[04:13] <omega_> since it's so light-weight
[04:13] <wtay> I was also thinking about sending the typefind function a different offset of the buffer
[04:13] <omega_> have a limit of N buffers, or M bytes, before giving up
[04:14] <omega_> but if it has no sate..
[04:14] <omega_> er, state
[04:14] <omega_> consider if the mp3 frame header spans the first two buffers
[04:15] <wtay> let's say the typefind function takes buffers until it gets N bytes and then run them through the different typefind functions
[04:15] <omega_> what is N?
[04:15] <omega_> what if N is 4 bytes, and we wait for 1MB
[04:15] <omega_> um
[04:15] <wtay> well
[04:15] <omega_> something odd is happening
[04:16] <omega_> I run -regsiter, then -launch, no probs
[04:16] <omega_> run -launch again and it says to rebuild the regsitry
[04:16] <wtay> where are you runnning -launch from?
[04:16] <omega_> gst/autoplug/
[04:16] <omega_> it shouldn't matter
[04:17] <wtay> doesn't lauch create an .xml file, which touches the dir modify time?
[04:17] <omega_> oh
[04:17] <omega_> ick
[04:17] <omega_> btw, someone mentioned the nodiratime mount option
[04:17] <omega_> which would hose the time checks
[04:18] <wtay> depends on what time we check for, atime is quite useless in general
[04:18] <omega_> oh, right, duh
[04:18] <omega_> mtime
[04:19] <wtay> we should seriously think about something to locate plugins by URI..
[04:19] <omega_> yes
[04:19] <wtay> like file: -> disksrc
[04:20] <omega_> maybe have a source registry
[04:20] <omega_> source plugins register what they can handle
[04:20] <omega_> hadess: you around?
[04:20] <wtay> I was thinking about a regexp syntax to describe the type, instead of the extension...
[04:20] <omega_> hmmmm
[04:21] <hadess> omega_: si ?
[04:21] <omega_> the 'type' ?
[04:21] <omega_> hadess: you have that gnomevfssrc code?
[04:21] <wtay> hmm, doesn't work too wel in this case..
[04:21] <omega_> if you're not gonna have time to work on it, can you send it to me?
[04:21] <hadess> no, i removed it last time i updated gstreamer :/
[04:21] <omega_> grrr
[04:21] Nick change: ajbusy -> ajmitch
[04:21] <hadess> but i will work on it this WE, i need to finish this widget first though
[04:22] <omega_> ok
[04:22] <wtay> gst_elementfactory_register_uri()
[04:22] <omega_> so, would we then have a regexp library dependency?
[04:22] <omega_> or does glib have something rudimentary?
[04:23] <wtay> hmmm, I don't think so
[04:23] <omega_> not that I can see
[04:24] <wtay> mabe regexp is a bit too much after all
[04:24] <omega_> yeah, though I duno what the best method would be
[04:25] <wtay> we can't really do anything with the extensions really
[04:25] <wtay> btw how are going to use those?
[04:25] <omega_> dunnpo
[04:25] <omega_> do we need to?
[04:25] <wtay> we have them
[04:25] <wtay> could speed up typefind
[04:25] <omega_> maybe
[04:26] <omega_> wtay: where's the mime/type of a GstCaps?
[04:26] <omega_> ->name ?
[04:26] <wtay> ->id
[04:27] <wtay> use gst_type_id_to_mime etc
[04:27] <omega_> um, er
[04:27] <omega_> ok
[04:27] <wtay> we don't keep the mime types around, they always are converted to the id
[04:27] <omega_> right
[04:27] <omega_> that function doesn't exist
[04:28] <wtay> gst_type_find_by_id()->mime
[04:28] <omega_> there' gst_caps_get_mime
[04:28] <wtay> ok, that's the function you need
[04:29] <omega_> yup, that's it
[04:29] <wtay> internally does _find_by_id()->mime
[04:29] <omega_> bingo
[04:29] <omega_> yeah
[04:29] <ajmitch> hmm
[04:29] <omega_> btw, I had an idea for gst_buffer_append
[04:29] <omega_> it can look at the parent buffer of the two buffers
[04:30] <ajmitch> so what parts of gstreamer aren't broken right now that we can play with? ;)
[04:30] <omega_> if they're the same, and the offset shows they're contiguous...
[04:30] <omega_> it unrefs both and creates a new subbuffer to span where they both were ;-)
[04:30] <omega_> ajmitch: um, colorized debug ;-)
[04:30] <wtay> omega_: hehe
[04:30] <ajmitch> hehe
[04:30] <wtay> ajmitch: aasink :-)
[04:30] <omega_> wtay: this works great for typefind
[04:31] <wtay> omega_: yeah, could speed up mp3 parsing too
[04:31] <omega_> it simply appends the new buffers, and if they're contiguous, no copies occur
[04:31] <omega_> wtay: true
[04:31] <ajmitch> wtay: i should try that out, what's the gstreamer-launch format for playing an mpeg to an aasink? ;)
[04:31] <omega_> gstreamer-launch disksrc location=abc.mpg ! mpeg2dec ! aasink
[04:31] <omega_> driver=3 puts it in the current terminal, else it creates a window
[04:32] <omega_> abc.mpg better be an elementary stream
[04:32] <omega_> else disksrc ! mpegNparse video_00! mpeg2dec
[04:32] <wtay>  gstreamer-launch disksrc location=abc.mpg ! mpeg2parse video_00! mpeg2dec ! aasink
[04:32] <omega_> now that we have delayed connections <g>
[04:32] <ajmitch> it's an mpeg1 video
[04:32] <ajmitch> mpeg2parse handles that?
[04:32] <omega_> mpeg1parse instead
[04:32] <omega_> mpeg2dec handles mpeg1
[04:33] <wtay>  gstreamer-launch disksrc location=abc.mpg ! mpeg1parse video_00! mp1videoparse ! mpeg_play ! aasink
[04:33] <omega_> wtay: mpeg2dec can play mpeg1 videos
[04:33] <wtay> omega_: doesn't accept the mpeg1 caps yet...
[04:33] <omega_> eh?
[04:33] <omega_> trying
[04:33] <omega_> ah
[04:33] <omega_> trivially fixable
[04:34] <wtay> yup
[04:34] <omega_> ajmitch: line 106 of plugins/mpeg2/mpeg2dec/gstmpeg2dec.c
[04:34] <omega_> change from GST_PROPS_INT (2) to
[04:34] <omega_> GST_PROPS_LIST ( GST_PROPS_INT (1), GST_PROPS_INT (2) )
[04:35] <wtay> or GST_PROPS_INT_RANGE (1,2)
[04:35] <omega_> true
[04:35] <ajmitch> omega_: wouldn't it be easier for you or wtay to commit the fix? ;)
[04:35] <omega_> but that doesn't scale to MPEG 1, 2, 4
[04:35] <omega_> I can do it
[04:35] <wtay> true :)
[04:36] <omega_> hmmmm
[04:36] Action: omega_ has other problems
[04:36] <omega_> but I'll commit that anyway
[04:37] <ajmitch> well, we won't go into you other problems ;)
[04:37] <omega_> ah, right ;-)
[04:37] <wtay> omega_: probs?
[04:37] <omega_> committed
[04:37] <omega_> ** CRITICAL **: file gstpad.c: line 1390 (gst_pad_pull): assertion `peer != NULL' failed.
[04:37] <ajmitch> hmm, problems, committed - people could get the wrong impression of you omega_ ;)
[04:37] <omega_> we need to have gst-specific assertions
[04:37] <omega_> ajmitch: who's to say that's the wrong impression?
[04:38] Action: omega_ laughs maniacally
[04:38] Action: ajmitch wonders why insanity & coding go so well together
[04:38] <wtay> hmm, HEAD goes into full rebuild mode
[04:38] Action: omega_ points to ajmitch
[04:38] <omega_> wtay: ?
[04:39] <wtay> omega_: a chnage to configure.in I didn't have yet...
[04:39] <omega_> ah, mpeg2dec:sink isn't connected for some reason
[04:40] Action: ajmitch is bored
[04:40] <wtay> omega_: looks like it also need the systemstream=false propety..
[04:41] <omega_> eh?
[04:41] <ajmitch> i've been told to keep my hands off gstmediaplay, i guess i should try and hack something of my own now ;)
[04:41] <omega_> eh?
[04:41] <omega_> no, just assume that the media handling is 'b0rken'
[04:41] <omega_> wtay: which needs systemstream=false ?
[04:41] <omega_> mpeg2dec has it
[04:42] <wtay> omega_: but disksrc will not provide it..
[04:42] <omega_> so? mpeg1parse is supposed to
[04:42] <omega_> and it does
[04:42] <wtay> you're right...
[04:43] <omega_> uh?
[04:43] <wtay> in INCSCHED1?
[04:43] <omega_>         mpegversion: Integer range: 1 - 2
[04:43] <omega_>         systemstream: Boolean: FALSE
[04:43] <omega_> yes
[04:43] <omega_> oops
[04:43] <wtay> heh
[04:43] <omega_> I put int_range(int(),int())
[04:44] <wtay> ouch
[04:44] <omega_> still fails
[04:44] <omega_> it managed to survive that, believe it or not
[04:44] <omega_> sure you didn't put some code in for that 'case' /
[04:44] <omega_> ?
[04:45] <wtay> ?
[04:45] <omega_> since it's varargs now
[04:45] <omega_> maybe not
[04:45] <omega_> anyway, it still fails
[04:45] <wtay> It silently ignores..
[04:46] <omega_> it never connects mpeg2dec to mpeg1parse
[04:46] <omega_> does mpeg1parse set the SOMETIMES flag?
[04:46] <omega_> yup
[04:46] <omega_> oh
[04:47] <wtay> mm
[04:47] <omega_> hm, thought it might be video_0
[04:47] <omega_> no, it's video_1
[04:47] <omega_> nope
[04:47] <wtay> dunno what's wrong...
[04:47] <omega_> grrr
[04:47] <omega_> I'll look at it later
[04:47] <omega_> else I'll get side-tracked from the dynamic autoplugger again
[04:47] <wtay> it used to work with a queue
[04:47] <wtay> oooooooooooh
[04:47] <omega_> ?
[04:47] <wtay> dynamic pads are not in INCSCHED
[04:47] <omega_> uh?
[04:47] <wtay> for -launch
[04:48] <omega_> oh
[04:48] <ajmitch> wtay: btw, gstmediaplay stopped crashing after a gstreamer-register
[04:48] <omega_> that means that our handling of the registry is hosed
[04:48] <ajmitch> ;)
[04:48] <ajmitch> sorry for the great news; )
[04:48] <wtay> I still doubt that...
[04:49] <omega_> ?
[04:50] <wtay> ./gstreamer-launch disksrc location=/opt/data/armageddon1.mpg ! mpeg1parse video_00! queue ! { mp1videoparse ! mpeg_play ! aasink }
[04:50] <wtay> works in HEAD
[04:50] <wtay> doesn't work in INCSCHED though
[04:51] <omega_> ok, well, I'll deal later
[04:52] <wtay> aasink driver=2 is sooo much faster then the default one...
[04:52] <omega_> heh
[04:53] <omega_> I don't see that much difference between that and driver=1
[04:54] <wtay> whith driver=2 I can rescale to fullscreen without frame drops
[04:54] <ajmitch> where can us mortals see what is planned for 0.2.0 and later releases...?
[04:54] <omega_> the same place us 
[04:54] <omega_> er, us immortals can't
[04:55] <ajmitch> so us mortals have to ask you immortals when you have time?
[04:55] <omega_> yeah, time to write a list
[04:55] <ajmitch> isn't Uraeus doing something like that?
[04:55] <omega_> he offered to
[04:56] <omega_> as soon as he shows up again we can start
[04:56] <wtay> he's not as immortal, so he'll have to ask us... <g>
[04:56] <ajmitch> well, it's probably early morning over there, he could be sleeping ;)
[04:56] <wtay> naah only 5am :)
[04:56] <omega_> yup
[04:56] <omega_> only
[04:56] <ajmitch> immortals like wtay don't need sleep ;)
[04:57] <omega_> darn, that makes me a mortal ;-(
[04:57] <wtay> hmm, me too :(
[04:57] <omega_> and for proof, just see -cvs
[04:57] <ajmitch> hmm, maybe i should do like last week and stay up all night on the ocmputer; )
[04:57] <omega_> ;-(
[04:58] <omega_> ajmitch: so just how much did you OC your ocmputer, anyway?
[04:58] Action: wtay wants bash tab completion for gst plugins in -launch
[04:58] <ajmitch> ;)
[04:58] <omega_> I'm running a Duro 500 at 750...
[04:58] <omega_> wtay: good idea, tcsh had a way to define your own argument tab completion
[04:58] <omega_> bash should have it too
[04:58] <wtay> oh cool
[04:59] <omega_> it's a readline thing though, as used in bash2
[04:59] Action: wtay is reading man bash
[04:59] Action: omega_ is reading man bash
[05:00] <wtay> Programmable Completion
[05:00] <wtay>        When word completion is attempted for  an  argument  to  a
[05:00] <omega_> is that bash or tcsh ?
[05:00] <wtay> bash
[05:00] <omega_> what ver?
[05:01] <wtay> 2.05
[05:01] Action: omega_ has 2.04.11
[05:01] <ajmitch> bah, you're so far behind ;)
[05:01] <omega_> but it seems to have it
[05:01] <omega_> just not in the man page
[05:01] <wtay> oh?
[05:01] <omega_> ok, I'll look into that tonight ;-)
[05:02] Action: omega_ reads gnome-help-browser bash
[05:02] <wtay> you'll need a compspec
[05:02] <omega_> yup
[05:02] <omega_> have to construct it from gstreamer-inspec
[05:03] <omega_> it would be so cool if it would complete element arg names, and understand filename args, etc.
[05:03] <omega_> even pad names ;-)
[05:04] Action: omega_ sees the bash/readline people hating us for eternity when we tell them that their programmable completion is not sufficient
[05:04] <wtay> I think it's possible..
[05:05] <omega_> ok, I'll try it
[05:06] <wtay> care if I send the bash manual about Programmable Completion?
[05:06] <omega_> huh?
[05:06] Action: ajmitch has to look into how good gstreamer handles buffering
[05:06] <wtay> omega_: since you don't have it...?
[05:06] <omega_> buffering or queueing?
[05:07] <omega_> wtay: I have the info pages
[05:07] <wtay> ok
[05:07] <ajmitch> not sure, i want to encode data coming in the line in jack on the sound card
[05:07] <ajmitch> if the cpu is slow, then the encoding will go slower ;)
[05:07] <ajmitch> i have a slow cpu ;)
[05:07] <omega_> yeah, and stuff will back up
[05:08] <ajmitch> up to what limit?
[05:08] <omega_> whatever you set as the max size of the queue
[05:08] <ajmitch> it uses tmp files, or ram?
[05:08] <omega_> ram
[05:08] <omega_> see the wiki for an idea of a plugin that uses both
[05:08] <ajmitch> hmm, i might have to add more swap ;)
[05:09] <ajmitch> only got 384MB ram & 270MB swap ;)
[05:09] <omega_> or implement the idea on the wiki <g>
[05:09] <omega_> only
[05:09] Action: wtay has to sleep now..
[05:09] <omega_> ok, l8r
[05:09] <wtay> cya all
[05:09] <ajmitch> how's the glib 2.0 port going? ;)
[05:09] Nick change: wtay -> wtay-zZ
[05:10] <omega_> happening probably in a week
[05:10] <ajmitch> ah, cool
[05:11] <ajmitch> then it will all be dependent on an unstable library only available from cvs?
[05:11] <omega_> on, we will do everything we can to make sure that we have a shim to run it on gtk
[05:11] <ajmitch> oh
[05:11] <omega_> although I'm told that such a thing exists already, so I'll bug Tim Janik about that
[05:11] <ajmitch> doesn't worry me, i have glib installed from cvs; )
[05:20] taaz-away (dlehn at left irc: Ping timeout for taaz-away[]
[05:42] chillywilly (baumannd at d127.as9.nwbl0.wi.voyager.net) joined #gstreamer.
[05:56] taaz-away (dlehn at joined #gstreamer.
[05:56] Nick change: taaz-away -> taaz
[05:56] <ajmitch> taaz?
[05:56] <taaz> ajmitch?
[05:56] <chillywilly> ?
[05:56] <omega_> omega_?
[05:56] <chillywilly> chillywilly?
[05:57] <chillywilly> lol
[05:57] Action: omega_ is working on advanced commandline completion
[05:57] <chillywilly> ooooooh
[05:57] <taaz> omega_: you really need to merge that incsched stuff soon.. you're doing normal devel in that branch now it looks like
[05:57] <chillywilly> aaaaaaah
[05:57] <omega_> taaz: yeah, I've pinned down the offending code
[05:57] <ajmitch> taaz: .debs?
[05:57] <omega_> just need to analyze and fix it now
[05:58] <taaz> ajmitch: do you want to try them?
[05:58] <ajmitch> taaz: umm
[05:59] <ajmitch> taaz: i got gstreamer compiled just fine, but others are missing out on a wonderful experience
[05:59] <taaz> i was going to rebuild from a 'make dist' but that's busted...  arts not working due to hardcoded omega-paths or something
[05:59] <omega_> taaz: just turn off arts for now
[05:59] <omega_> it doesn't do much anyway
[05:59] <omega_> don't let something that trivial block you
[05:59] <taaz> oh?  how do i turn off a feature?
[05:59] <omega_> checking
[06:00] <taaz> this is 'make dist' failing... i don't know why it's even trying to compile it.
[06:00] <ajmitch> taaz: you an official maintainer yet?
[06:00] <omega_> taaz: there's already a check for artsc in configure, won't that work?
[06:00] <omega_> do you have arts installed?
[06:00] <chillywilly> probably a configure switch?
[06:00] <taaz> ajmitch: no
[06:01] <taaz> its confusing... it has built just fine.  but now make dist is failing due to mcopidl or something not finding artsflow.idl.
[06:02] <ajmitch> ah, kde stuff eh? ;)
[06:02] <taaz> i wouldn't know... its broken.  someone else fix it please ;)
[06:02] <ajmitch> ;)
[06:03] Action: taaz is sick of dealing with auto* crapola
[06:03] <omega_> why don't you just remove arts from the plugins/Makefile.am file in your working copy for now
[06:03] <omega_> I'll figure it out later
[06:03] Action: omega_ doesn't want to dive into aRts right now
[06:03] <chillywilly> mcop, what the hell is this mcop crap...inventing your own CORBA is retarded
[06:03] <omega_> chillywilly: tell that to stw, everyone here agrees with you
[06:04] <ajmitch> chillywilly: designed for simplicity and speed, but not for interoperability
[06:04] <chillywilly> I am sure M$ uses the same excuse when they "embrace and extend"
[06:07] Action: chillywilly has no love for KDE ppl
[06:07] Action: ajmitch has noticed
[06:14] <taaz> where's a good spot to put debs anyway?  gst.net/debian/?  gst.net/releases/debian/?
[06:15] <omega_> gst.net/releases/0.1.1/debs/
[06:15] <ajmitch> then you can be an official maintainer ;)
[06:19] <taaz> hmm... it's not really 0.1.1 though.  more like 0.1.2-minus-one-week/debs/ ;)
[06:19] <ajmitch> taaz: you gone thru the maintainer process yet?
[06:20] <taaz> um... i answered that already ;)  no
[06:20] <ajmitch> doh
[06:20] <taaz> been kinda busy... i'll get that done at some point
[06:21] <omega_> ooooh, this is gonna be so cool!
[06:21] Action: omega_ is prototyping in shell, but will code it in C eventually
[06:22] <taaz> ?
[06:22] <omega_> gstreamer-launch <tab>
[06:26] <omega_> ooooh
[06:26] <omega_> ./gstreamer-launch <tab> <tab>
[06:26] <ajmitch> oooh?
[06:26] <omega_> Display all 132 possibilities?

More information about the gstreamer-devel mailing list