IRC Logs

*******************************************************************
[03:01] bluejay ([email protected]) joined #gstreamer.
[03:01] <bluejay> heh, foser just flamed me for complaining that gst wasn't in portage yet.
[03:02] <foser> i don't flame.. i put one rightfully in it's place
[03:04] Action: Company hands out lollies
[03:04] <Company> gst is in nothing yet because the apps haven't caught up, so nobody needs to care anyway :p
[03:04] <bluejay> yeah. I think I misinterpreted your statement to saying it would be held back until all of gnome 2.6 was ready.
[03:05] <Company> GSTreamer's release schedule isn't part of GNOME
[03:05] <Company> (needs to be set to keep markey, wheels and friends happy) ;)
[03:06] <bluejay> "your"==foser, not Company
[03:06] <Company> yeah, i know
[03:06] <Company> i write "set" instead of "said", so bear with me
[03:06] <BBB> miguel is evil
[03:07] <Company> why?
[03:08] Action: Company assumes BBB meant e Icaza
[03:08] <BBB> yes
[03:08] <foser> dunno what migual was df-ing. but it wasnt gstreamer now was it ?
[03:08] <Company> +d
[03:08] <BBB> he says you need 160 MB GStreamer to play mp3s
[03:08] <BBB> that's theoretically true, but he says it in a way that suggest that our core lib is 160MB binary
[03:08] <Company> well
[03:08] <BBB> gst + gst-plugins + totem + rest is, together, uncompressed, source, 160MB
[03:09] <BBB> that's different
[03:09] <Company> i need 160MB gnome+mono to watch pictures
[03:09] <Company> so what
[03:09] <BBB> heh :)
[03:09] Action: BBB put a reply on p.g.o
[03:09] <foser> the one interesting mono gnome app (the one interesting mono app anyway ;)) also needs gstreamer
[03:10] <Company> lol
[03:11] <Company> comparing xine and gstreamer is like comparing mono and pascal
[03:11] <Company> good to know miguel sometimes doesn't know what he's talking about, too :)
[03:12] <foser> Company: well.. actually it looks like he does know, but he doesnt seem to care much as long as he gets his mp3 player
[03:13] <Company> he should use madplay
[03:14] <foser> mpg123
[03:16] <BBB> mpg321 is better-licensed
[03:22] BBB ([email protected]) left irc: "Client exiting"
[03:27] foser ([email protected]) left irc: "[ I want to believe ]"
[03:37] Misirlou ([email protected]) joined #gstreamer.
[04:09] Nick change: jcsston` -> jcsston
[05:44] spyder482 ([email protected]) joined #gstreamer.
[06:02] kmaraas ([email protected]) left irc: Client Quit
[06:03] Zeenix ([email protected]) left irc: Read error: 110 (Connection timed out)
[06:09] sub_pop ([email protected]) left irc: "Client exiting"
[06:16] sub_pop ([email protected]) joined #gstreamer.
[06:37] psm ([email protected]) left irc: 
[06:43] mathrick|sleep ([email protected]) left irc: Ping timeout: 14400 seconds
[06:55] Misirlou ([email protected]) got netsplit.
[06:55] ds ([email protected]) got netsplit.
[06:59] Misirlou ([email protected]) returned to #gstreamer.
[06:59] ds ([email protected]) returned to #gstreamer.
[06:59] jcsston` ([email protected]) joined #gstreamer.
[07:00] jcsston ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[07:29] mathrick ([email protected]) joined #gstreamer.
[07:38] Nick change: jcsston` -> jcsston
[07:40] thaytan ([email protected]) left irc: Read error: 60 (Operation timed out)
[07:43] Uraeus ([email protected]) joined #gstreamer.
[07:44] thaytan ([email protected]) joined #gstreamer.
[07:45] <Uraeus> morning
[07:54] <steve_b> hi
[07:55] <Uraeus> steve_b: so what is happening in the land of Peter Jackson (aka New Zealand) ?
[07:57] <steve_b> i think we're leaving the poor man alone long enough to start on kong
[07:57] <Uraeus> heh :)
[07:59] ChrisHJW ([email protected]) joined #gstreamer.
[08:07] dennis- ([email protected]) left irc: Read error: 110 (Connection timed out)
[08:34] apoc_ ([email protected]) joined #gstreamer.
[08:41] spyder482 ([email protected]) left irc: "Leaving"
[08:42] ChrisHJW ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[08:50] markey ([email protected]) joined #gstreamer.
[08:51] <markey> morning :)
[08:51] <Uraeus> morning markey
[08:51] <markey> how can I read the number of channels from my spider?
[08:51] apoc ([email protected]) left irc: Read error: 110 (Connection timed out)
[08:51] <markey> I guess it is interleaved, one word left channel, the next right channel?
[08:52] <Uraeus> hmm, since development questions are not my strenght I guess it is time for me to sign off and go to work :)
[08:52] <Uraeus> bye :)
[08:52] <markey> cya
[08:52] Uraeus ([email protected]) left irc: "Client exiting"
[08:55] <markey> oh I should add: I'm talking about the audio data buffer delivered by identity (signal handoff)
[09:05] apoc_ ([email protected]) left irc: "Leaving"
[09:08] <mathrick> markey: afaik, it's part of caps
[09:08] <mathrick> number of channels that is
[09:08] <markey> ah
[09:08] <mathrick> and they're interleaved, but dunno in what way
[09:09] <markey> can you explain to me how to read the channels caps?
[09:09] <markey> I think I know how to read it from a concrete decoder (like mad)
[09:09] <markey> but how to read it from spider?
[09:11] <mathrick> read it from identity, you're connecting it right after spider?
[09:11] thomasvs ([email protected]) joined #gstreamer.
[09:11] <mathrick> yo thomasvs
[09:13] <markey> yes
[09:13] <markey>     gst_element_link_many( m_pFilesrc, spider, pIdentity, m_pAudiosink, NULL );
[09:17] <mathrick> markey: anyway, shouldn't be any different for spider than for mad - i'd try gst_pad_get_caps(gst_element_get_pad (spider, "audio_00"))
[09:17] <markey> thanks
[09:17] <markey> willtry
[09:19] Rotty ([email protected]) joined #gstreamer.
[09:19] thaytan ([email protected]) got netsplit.
[09:19] sxpert ([email protected]) got netsplit.
[09:22] thaytan ([email protected]) returned to #gstreamer.
[09:22] sxpert ([email protected]) returned to #gstreamer.
[09:43] ChrisHJW_log ([email protected]) left irc: Read error: 60 (Operation timed out)
[09:52] Company ([email protected]) left irc: Read error: 60 (Operation timed out)
[10:03] jcsston ([email protected]) left irc: Read error: 60 (Operation timed out)
[10:12] jdahlin ([email protected]) joined #gstreamer.
[10:16] dolphy ([email protected]) joined #gstreamer.
[10:22] sub_pop ([email protected]) left irc: "Client exiting"
[10:25] sub_pop ([email protected]) joined #gstreamer.
[10:40] Nick change: mathrick -> mathrick|Uni
[10:57] <markey> gaah
[10:57] Action: markey been searching for "GstProbs" the last hour in the "stable" API docs
[10:57] <markey> stable == 0.8, I thought..
[10:57] <markey> but GstProps is gone
[10:58] <markey> how can I read this frigging "channels" property from the pad?
[10:59] <jdahlin> what do you want to do?
[11:00] <markey> I need to read the "channels" property from the spider
[11:00] <markey> to determine the number of audio channels used
[11:02] <jdahlin> g_object_get (spider, "channels", &channels, NULL); ?
[11:02] sjoerd ([email protected]) left irc: Remote closed the connection
[11:04] <jdahlin> or no, hmm
[11:05] <markey> mathrick suggested: 
[11:05] <markey> |09:32 < mathrick> markey: anyway, shouldn't be any different for spider than for mad - i'd try
[11:05] <markey> |                  gst_pad_get_caps(gst_element_get_pad (spider, "audio_00"))
[11:05] <markey> but how to read the actual property from the caps?
[11:06] <jdahlin> seems to be a mess
[11:06] <jdahlin> check the size of the caps and, iterate up to that value, get the structure for that value
[11:06] <jdahlin> then check the name of the structure
[11:06] Action: markey wishes it was all written in c++
[11:06] <markey> would make it easier
[11:06] <jdahlin> and finally get the value
[11:07] <markey> ok thx
[11:07] sjoerd ([email protected]) joined #gstreamer.
[11:07] <jdahlin> well, the api could really be improved
[11:07] <markey> absolutely..
[11:07] <jdahlin> gst_pad_get (pad, "channels", &channels, NULL);
[11:08] <jdahlin> shouldn't be impossible to write at all
[11:08] <jdahlin> gst_pad_get_caps_value
[11:09] <markey> gst_names_have_the_tendency_to_become_insanely_long
[11:09] <jdahlin> gst_element_factory_get_num_pad_templates ? :)
[11:09] <markey> omg ;)
[11:10] <markey> that's all because of using c + glib
[11:10] <jdahlin> partly
[11:10] <markey> you need to encode type information in the function name
[11:10] <markey> sucks..
[11:10] <markey> with real OO languages you mostly don't have that
[11:11] <markey> </rant>
[11:11] <jdahlin> why don't you use the c++ bindings then?
[11:12] <markey> 1) gst api still changes frequently, bindings are never up to date
[11:12] <markey> 2) another dependency
[11:12] <jdahlin> the api for 0.8 won't change
[11:12] <markey> 3) I can never really judge gst by just using bindings
[11:12] Action: jdahlin tries to get the python bindings up to date
[11:17] kmaraas ([email protected]) joined #gstreamer.
[11:19] <markey> To get the properties of a caps structure the functions gst_caps_get_boolean(), gst_caps_get_fourcc_int(), gst_caps_get_int(), gst_caps_get_string(), gst_caps_get_float(), which all take a property name as an argument. 
[11:19] <markey> ^^ I wonder where this is properly documented 
[11:20] <markey> can't find gst_caps_get_int() API docs
[11:20] <jdahlin> It seems to be outdated
[11:21] <markey> omg
[11:21] <markey> *sigh*
[11:21] <jdahlin> use gst_caps_get_structure()
[11:21] <jdahlin> like I said
[11:21] <jdahlin> maybe I wasn't clear enough
[11:22] sublett ([email protected]) joined #gstreamer.
[11:22] <jdahlin> for (i = 0; i < gst_caps_get_size(); i++)
[11:22] <jdahlin>   structure = gst_caps_get_structure(caps, i);
[11:22] <markey> thx
[11:22] <jdahlin>   if (gst_structure_has_field(structure, "channels"))
[11:23] <jdahlin>     if(!gst_structure_get_int(structure, "channels", &channels))
[11:23] <jdahlin>       *throw error*
[11:26] Action: jdahlin ponders how GstStructure should be represented in python
[11:26] <jdahlin> as a list or as a dictionary
[11:31] sjoerd ([email protected]) left irc: Read error: 113 (No route to host)
[11:32] <dolphy> jdahlin: i restarted buildbot on my home machine
[11:32] <markey> (process:25055): GStreamer-CRITICAL **: file gstpad.c: line 2518 (gst_pad_get_caps): assertion `GST_IS_PAD (pad)' failed
[11:33] <markey> any idea?
[11:34] <markey> maybe the pad name is wrong
[11:34] <markey> audio_00
[11:34] <dolphy> markey: what are you doing ?
[11:34] <markey> doing this:
[11:34] <markey>    GstCaps* caps = gst_pad_get_caps(gst_element_get_pad( pGstEngine->m_pSpider, "audio_00" ) );
[11:35] <jdahlin> dolphy: great
[11:35] <jdahlin> dolphy: I have no idea why it just stopped
[11:35] ChrisHJW_log ([email protected]) joined #gstreamer.
[11:36] <jdahlin> markey: there are no audio_00 pad in the spider
[11:36] <markey> src_00 is correct, no?
[11:36] <jdahlin> I'm not sure what it's called
[11:36] Company ([email protected]) joined #gstreamer.
[11:37] <jdahlin> Company probably knows
[11:37] <markey> I'll try
[11:37] <markey> |Pad Templates:
[11:37] <markey> |  SRC template: 'src_%d'
[11:37] <markey> says gst-inspect
[11:37] <jdahlin> Company: if I want to find out the number of channels on the decoder (if available)
[11:37] <jdahlin> true
[11:38] <jdahlin> Company: and using spider, how can I easily find the correct pad?
[11:45] sjoerd ([email protected]) joined #gstreamer.
[11:46] <dolphy> jdahlin: power failure at home i think :)
[11:47] <jdahlin> ah, that explainsit
[11:47] <dolphy> jdahlin: seems it needs a rm -rf 
[11:47] <dolphy> jdahlin: on source tarbals
[11:47] <jdahlin> dolphy: rm gstreamer/build/po/sv.po should be enough for now
[11:47] <jdahlin> dolphy: I hate conflicts
[11:55] foser ([email protected]) joined #gstreamer.
[11:55] <markey> yay!
[11:55] <markey> now it works
[11:55] <markey> scr_0  <-- correct pad name
[11:55] <markey> err src_0
[12:05] <markey> thanks again jdahlin 
[12:06] <jdahlin> markey: you're welcome
[12:06] <jdahlin> markey: just make sure you're app rocks :)
[12:06] <jdahlin> your
[12:07] <markey> the gst engine will rock, I'm sure :)
[12:07] Shoragan ([email protected]) joined #gstreamer.
[12:08] <jdahlin> what are you writing btw?
[12:09] <markey> amaroK
[12:09] <markey> http://amarok.sf.net
[12:13] <jdahlin> looks quite nifty
[12:13] Action: jdahlin is a gnomer though :)
[12:15] <markey> I almost suspected that ;)
[12:15] <markey> gnomers are pretty rare on this chan in general ;)
[12:15] <jdahlin> heh
[12:15] <jdahlin> I also prefer a iTunes like interface
[12:16] <jdahlin> so we could have lots of "interesting" discussions :)
[12:16] <markey> like rhythmbox / JuK
[12:16] <markey> well we will have a "Collections" module, too
[12:16] <jdahlin> Me and friend wrote a player a couple of months ago: http://www.imendio.com/projects/jamboree/
[12:16] <markey> looks nice, bit like Rhythmbox
[12:17] <jdahlin> or iTunes
[12:17] <markey> yes.. ;)
[12:18] md` ([email protected]) left irc: 
[12:19] Nick change: markey -> markey|afk
[12:31] ain ([email protected]) joined #gstreamer.
[12:36] markey|afk ([email protected]) left irc: "leaving"
[12:57] Nick change: kmaraas -> km_mtg
[12:58] md` ([email protected]) joined #gstreamer.
[13:19] kwm ([email protected]) joined #gstreamer.
[13:24] apoc ([email protected]) joined #gstreamer.
[13:24] <apoc> hi
[13:24] <Company> good morning angels
[13:45] <Company> whoops
[13:45] <Company> do we really have the GPL in COPYING and not the LGPL?
[13:46] <Company> we kinda do
[13:46] <Company> there's a COPYING.LIB
[13:46] <Company> aren't we supposed to be all GPL by now?
[13:50] Action: Company checks if thomasvs reacts to poking
[13:51] <jdahlin> Company: he's away this week
[13:51] <jdahlin> We're not supposed to be all GPL
[13:51] <jdahlin> wouldn't that be quite useless for non-free users of gstreamer?
[13:52] <Company> all LGPL
[13:52] <Company> oops
[13:52] <Company> anyway, the COPYING file contains the GPL
[13:53] <ain> morning
[13:53] <Company> and i think we don't have any GPL code left
[13:53] <jdahlin> not even the utilities?
[13:53] <Company> afaik they'Re LGPL, too
[13:53] <Company> they need to be
[13:53] <jdahlin> it makes sense to make them GPL, no?
[13:53] <Company> or you can't use them with non-GPL-compatible plugins
[13:54] <Company> it would be kinda bad if gst-register refused registering divx plugins ;)
[13:56] <Company> they're having an LGPL header at least
[13:57] <jdahlin> indeed
[14:01] md` ([email protected]) left irc: 
[14:03] <Company> i wrote a mail to the list
[14:05] Shoragan ([email protected]) left irc: "Leaving"
[14:08] ploum ([email protected]) joined #gstreamer.
[14:10] <jdahlin> (process:16443): GLib-GObject-WARNING **: the copy_value() implementation of type `GstIntRange' seems to make use of reserved GValue fields
[14:10] <jdahlin> hmm.
[14:14] <Company> oops
[14:15] md` ([email protected]) joined #gstreamer.
[14:15] <Company> jdahlin: how do you trigger that?
[14:16] <jdahlin> Company: by using g_boxed_copy on a GstStructure
[14:16] <jdahlin> probably wrong
[14:16] Action: jdahlin tries with gst_structure_copy
[14:16] <Company> no
[14:17] <Company> it shouldn't be wrong
[14:18] <Company> i need to discuss that with ds-work before doing any changes
[14:18] <jdahlin> yeah
[14:18] <jdahlin> interesting, it doesn't happen in gdb
[14:18] <jdahlin> I need to do some GstIntRange python mapping love
[14:19] <jdahlin> I do have a question though
[14:20] <jdahlin> http://rafb.net/paste/results/s2028434.html
[14:20] <jdahlin> is that's really required for finding out how many channels used by a pipeline?
[14:20] hallibaby ([email protected]) joined #gstreamer.
[14:20] hallibaby ([email protected]) left irc: Client Quit
[14:21] <Company> pretty much
[14:21] <Company> though you should use get_negotiated_caps
[14:21] <jdahlin> awkward
[14:21] <Company> and negotiated caps have only 12 structure
[14:21] <Company> and negotiated caps have only 1 structure
[14:21] <Company> yeah, that's what you get by using abstractions
[14:21] <jdahlin> okay, but I still need to iterate over all elements and all pads?
[14:22] <Company> uh, no
[14:22] <Company> just ask the element you want to know the number of channels from
[14:22] <Company> the audiosink for example
[14:22] <jdahlin> and how can I easily find that out if I use spider?
[14:23] <jdahlin> just using the sink?
[14:23] <Company> yeah
[14:23] <Company> you connect spider to something - ask that something
[14:24] <Company> it could even be that you don't get any channels
[14:24] <Company> some soundcards support mp3 decoding
[14:25] <jdahlin> and, is it enough to check the sinkpad for the audiosink?
[14:25] <jdahlin> err srcpad
[14:25] <Company> sinkpad
[14:25] <Company> yeah, checking the sinkpad of the audiosink should be enough
[14:56] <foser> what is exactly needed to get the gnome gst mixer to work/build ?
[14:57] walters ([email protected]) joined #gstreamer.
[14:59] <Company> foser: gst-plugins - or at least the contents of gstreamer-interfaces.pc
[14:59] <Company> and all the gnome stuff ;)
[14:59] <Company> oh, walters works for redhat now?
[15:00] <foser> hmm i got that.. 'all the gnome stuff' means gnome-media ?
[15:01] <Company> yeah
[15:01] <Company> well, it's part of gnome mixer ;)
[15:02] <Company> the mixer is part of gnome media
[15:02] <walters> Company: yeah
[15:02] <Company> i'm shuffeling my words nicely today
[15:02] <Company> walters: what do you work on?
[15:04] <walters> Company: desktop stuff, although i've kind of been pulled temporarily for selinux bits
[15:06] <Company> good to know
[15:07] <Company> though it's kinda bad Gnome is losing the "student input" and moving into the professional/commercial direction
[15:14] markey ([email protected]) joined #gstreamer.
[15:15] <foser> Company: but it seems the volume-control won't build if you don't have an element implementing the interface (?)
[15:15] <markey> re
[15:16] <Company> foser: didn't know about that (and if so, that's a bug)
[15:17] <foser> i'll see if i can reproduce
[15:18] <Company> (the configure checks for gst in gnome-media seem broken to me anyway
[15:18] <Company> )
[15:19] Action: Company updates to be sure
[15:19] <Company> hey, markey is german
[15:19] Action: markey hides
[15:19] <Company> you only need to hide from iain
[15:19] <Company> markey: from where?
[15:19] <markey> dortmund
[15:19] <markey> u?
[15:20] <Company> near hamburg
[15:21] <markey> Company, what would be the easiest to use a software volume mixer?
[15:21] <markey> just fading down the output volume by a percentage
[15:21] Action: ain 'll get over the german issue...hopefully
[15:21] <Company> use the volume element
[15:21] <markey> ah yes thx
[15:21] Nick change: ain -> iain
[15:21] <Company> poor iain
[15:21] <Company> poor germany
[15:22] <Company> any gnome-media maintainer here?
[15:22] <Company> s/maintainer/hacker/
[15:22] Action: Company pokes thaytan
[15:24] Action: iain sets about filling in another job application form and wonders why he feels like he's selling out
[15:25] mathrick|Uni ([email protected]) left irc: Remote closed the connection
[15:31] <iain> "What have been your Major Achievements in rcent years and why do you consider them to be significant?"
[15:31] <iain> I learned to walk and talk at the age of 2...
[15:32] <foser> psycho crap..
[15:33] <iain> I hate talking about myself so I find these questions really hard
[15:33] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[15:33] <iain> "What would you consider to be the Main Strengths which you would bring to the post?"
[15:35] <Company> "I consider holding myserlf back from removing Germany from the face of the earth to be a mayor achievement. I think it's important for the continuing success of open source software."
[15:35] <Company> "My main strength is that I can hold myself back from eradicating Germany real well."
[15:35] <Company> i hate such questions
[15:36] <iain> "If working for a major news agency, I will not spread malicious rumours about german men"
[15:37] <jdahlin> Company: the GstIntRange warning, should I file a bug?
[15:37] <Company> jdahlin: yeah, that's a good idea
[15:38] <jdahlin> okay, will do
[15:54] alley_cat ([email protected]) left irc: Read error: 60 (Operation timed out)
[15:58] foser ([email protected]) left irc: "[ I want to believe ]"
[15:59] smoke ([email protected]) joined #gstreamer.
[16:01] BBB ([email protected]) joined #gstreamer.
[16:01] <BBB> mornin'
[16:01] <BBB> dolphy, thomasvs: are you there?
[16:02] <Company> * Company checks if thomasvs reacts to poking
[16:02] <Company> <jdahlin> Company: he's away this week
[16:02] <BBB> :(
[16:02] <jdahlin> dolphy is here though
[16:02] <jdahlin> but not right now
[16:02] <BBB> heh :)
[16:03] <BBB> jdahlin: were you able to reproduce my sloppy MPEG playback issue with my provided sample file?
[16:03] <BBB> (part of ...)
[16:03] ploum ([email protected]) left irc: "Leaving"
[16:03] <jdahlin> BBB, no, but I've been able to reproduce it with other mpeg files
[16:03] <BBB> ok, that's good enough
[16:03] <BBB> so do we have a plan to solve it?
[16:04] <BBB> (as in: if we want people to use totem-gst, we need to give them a reason to do so)
[16:04] <BBB> and ask dolphy if he has opinions on my totem CPU usage issues
[16:05] <markey> hmm where there some changes to GstClock?
[16:05] <markey> my code stopped working
[16:05] <markey> getting bogus time values
[16:05] <BBB> (divx5/mp3, 25fps,768x576 taking 100% CPU on totem whereas it only takes +/- 70% on gst-launch commandlines)
[16:05] <BBB> since when?
[16:06] <Company> there weren'T any changes
[16:06] <markey> 0.7.3
[16:06] <Company> oh
[16:06] <Company> yeah, a lot :)
[16:06] <BBB> when did you commit your gstclock updates, company?
[16:06] <BBB> wasn't that 0.7.3 -> 0.7.4? :)
[16:06] <markey> how to get the current time positio?
[16:06] <markey> +n
[16:06] <BBB> gst_element_get_time(), or query the sink element
[16:06] <BBB> I think
[16:06] <markey> thx
[16:07] <BBB> totem does the latter
[16:07] <jdahlin> BBB, it seems to be working if I'm using another scheduler, eg basicgthread
[16:07] <Company> query the sink element
[16:07] <jdahlin> and, people with smp systems has mentioned opt is crashing randomly
[16:07] <BBB> GstFormat fmt = GST_FORMAT_TIME; gint64 value; gst_element_query (sink, GST_QUERY_POSITION, &fmt, &value); // value is now the time in ns
[16:07] <jdahlin> that needs to be investigated though.
[16:07] <BBB> I noticed
[16:07] <markey> BBB: thanks
[16:07] <BBB> jdahlin: I know it works using basicgthread
[16:08] <BBB> that was my reason to suggest to switch to basicgthread
[16:08] <BBB> my *main* reason
[16:08] <BBB> not my only reason
[16:08] sjoerd ([email protected]) left irc: Read error: 113 (No route to host)
[16:09] <jdahlin> BBB, I think it's bad idea to switch during stable releases
[16:09] <jdahlin> Since we don't have good enough regression tests yet
[16:10] <BBB> yup, I know
[16:10] <BBB> we should switch as soon as 0.8.x is branched off, though
[16:10] <BBB> opt is dead (mho)
[16:10] <BBB> we can switch to ds/company's new scheduler when that's written ;)
[16:10] <BBB> nobody wants to maintain opt, nobody understands it fully and large parts are unimplemented... it's not worth it
[16:10] <BBB> (mho)
[16:11] <Company> that's not the problem
[16:11] <BBB> it's the consequence of the problem ;)
[16:11] <Company> the problem is that opt does something that schedulers shouldn'T ever do: it keeps buffers
[16:11] <BBB> so does basic
[16:11] sjoerd ([email protected]) joined #gstreamer.
[16:11] <BBB> and it's unrelated to the issue that opt is unmaintained
[16:12] <Company> basic does keep buffers?
[16:12] <BBB> bufpens
[16:12] <BBB> sure
[16:12] <Company> nope
[16:12] alley_cat ([email protected]) joined #gstreamer.
[16:12] <Company> it only puts a buffer in there and then switches cothreads to immediately process it
[16:13] <BBB> hm, yes, that's a bufpen, right?
[16:13] <BBB> I guess your definition of "keeps a buffer" is different from mine
[16:13] <BBB> ;)
[16:13] <BBB> that'sa why you hack schedulers and I write plugins
[16:13] <Company> well, opt puts loads of buffers in a bufpen
[16:13] <BBB> I know :P
[16:13] <Company> so there might be some left when an iteration is done
[16:14] <Company> basic never leaves any buffers in bufpens after an iteration is done
[16:22] <BBB> ok, didn't know that
[16:22] <BBB> prooves my ignorance about schedulers
[16:22] <BBB> ;)
[16:22] <BBB> I'm off, gotta go experiments today
[16:22] <BBB> I'll try to be online some part of the day
[16:22] <BBB> oh, I'm working on gst-rec again :)
[16:22] <BBB> maybe I'll finally get to a release by now
[16:23] <BBB> ;)
[16:23] <BBB> bbl!
[16:23] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[16:24] <markey> time position code works fine now
[16:25] <markey> hmm mad plugin makes noises sometimes when seeking
[16:36] yippi ([email protected]) left irc: Remote closed the connection
[16:46] ChrisHJW ([email protected]) joined #gstreamer.
[16:56] <markey> what to do in the eos signal callback?
[16:56] <markey> my code apparently does the wrong thing
[16:57] <Company> register an idle callback
[16:57] <markey> pardon?
[16:58] <markey> you mean do nothing?
[16:58] spyder482 ([email protected]) joined #gstreamer.
[17:03] <Company> no, i mean register a handler that is run in an idle callback
[17:04] <jdahlin> eek
[17:04] <jdahlin> gstvalue is so broken wrt to boxed types :(
[17:06] BBB ([email protected]) joined #gstreamer.
[17:06] <Company> jdahlin: that's probably because we didn't know that we may not use the second pointer
[17:07] Uraeus ([email protected]) joined #gstreamer.
[17:07] <jdahlin> Company: well, you're using the G_TYPE_BOXED GType, but you're not following the boxed rules
[17:07] <jdahlin> Company: boxed types should only be created by glib itself, by g_boxed_register_static
[17:08] <Uraeus> hi
[17:08] <BBB> jdahlin: that's crap, external libs may create boxed types too
[17:08] <jdahlin> hello Uraeus
[17:08] <BBB> hi Uraeus
[17:08] <jdahlin> BBB, yes, but they should use g_boxed_register_static
[17:08] <BBB> jdahlin: why else is it part of the API?
[17:08] Nick change: bilboed_ -> bilboed
[17:08] <BBB> jdahlin: oh, right :)
[17:09] <jdahlin> BBB, gstreamer does some nasty internal stuff to mimic GBoxed types
[17:09] Action: BBB missed the point of the sentence
[17:09] <jdahlin> maybe we can move away to another type, maybe G_TYPE_OBJECT
[17:10] <jdahlin> but I can't change anything in there without breaking stuff
[17:10] <Company> G_TYPE_VALUE?
[17:10] <jdahlin> hmm, that might work
[17:11] <jdahlin> but we still can't change it if we want to retain backward compatibility
[17:11] <Company> i'd consider it a bug and fix it
[17:11] <jdahlin> It's a bug
[17:11] <jdahlin> but is it worth fixing it?
[17:11] <BBB> I'd call it an ABI change
[17:12] <BBB> even though we don't change the layout of structures
[17:12] <BBB> or similar
[17:12] <BBB> don't touch it, just don't
[17:12] <BBB> 0.8.x is out
[17:12] <BBB> let's maintain it nicely
[17:12] <BBB> swing stuff may go into 0.9.x
[17:12] <jdahlin> Can we I add stuff?
[17:12] <jdahlin> s/we //
[17:12] <BBB> :)
[17:12] <BBB> I guess
[17:12] <Company> just fix it if it's broken
[17:13] <jdahlin> so, I can register some new, less crippled types :)
[17:13] <BBB> if it's API/ABI compatible and makes sense ;)
[17:13] <jdahlin> Company: Not if it breaks the ABI
[17:13] <Company> i'm not gonna get anal about ABI as long as it doesn't break our apps
[17:13] Action: BBB thinks company has a point
[17:13] <jdahlin> Well, we might not know that we broken stuff until we release the next version
[17:13] <BBB> if you can convince me that it doesn't break stuff (that's really very easy ;) ) I'm fine with it, I guess
[17:13] <jdahlin> and waits 6 months for people to use it
[17:14] <Company> if an app relied on that, i'd consider that app broken :p
[17:14] <Company> relying on undocumented side effects is broken
[17:15] <BBB> that's not the point, and you know it
[17:16] <Company> if we're that picky about ABI, i'll leave maintaining stable to the people interested
[17:16] <BBB> we've said that we will go for an ABI-stable series called "0.8.x" in GNOME-2.6
[17:16] <jdahlin> the point is that we already released 0.8 and told people go ahead use it, it's stable
[17:16] <Company> it is stable
[17:16] Action: BBB points company to miguel's words... let's try to convince him that gstreamer is actually worth it
[17:16] <Company> but it's not like we don't change it
[17:17] <jdahlin> we can't change existing APIs or break API if we want people to use it
[17:17] <jdahlin> it's not like in the good 'ol days :-)
[17:18] <Company> we don't break an API
[17:18] <BBB> Company: btw, do you mind if I change alsasink/alsasrc to derive from gstalsamixer instead of gstalsa? that'd make them similar to gstoss and make alsa useful for me in gstrec
[17:18] <BBB> we break ABI in a way
[17:18] <Company> no problem
[17:18] <BBB> it just doesn't change struct layout
[17:18] <BBB> :)
[17:19] <BBB> Company: thnx, will do tonight (US time)
[17:19] <Company> we break ABI the moment we change one line of code if you claim that
[17:19] <BBB> Company: :) maybe... ohwell, I'd say: go for it, and make sure that totem, RB and such apps don't break
[17:19] <BBB> then I'm fine with it
[17:19] <BBB> ok?
[17:20] <BBB> that sounds like a very light requirement to me, and a very reasonable one
[17:21] <BBB> no incompatible API changes, no ABI struct layout changes, and RB/totem should keep working
[17:21] Action: jdahlin shrugs
[17:21] thomasvs ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[17:21] <BBB> poor thomasvs, he died the moment I said that... bad bad me ;)
[17:21] <BBB> jdahlin: feel free to add stuff to it, I'm just trying to keep company into maintaining 0.8.x for a while :)
[17:21] yippi ([email protected]) joined #gstreamer.
[17:22] <BBB> I don't want to be responsible for maintaining 0.8.x myself
[17:22] <BBB> *alone
[17:23] <jdahlin> You'll have me and thomas for a while, that's for sure
[17:23] <BBB> (let me try saying that again: I will maintain 0.8.x - but not alone)
[17:23] <BBB> cool! :) but I want ds and company, too
[17:23] <Uraeus> and I be singing in the hallways :)
[17:23] <BBB> file bug reports for non-playing media that play with other mediaplayers
[17:23] <BBB> that'd be helpful
[17:24] <BBB> like the mp3s from wheels/markey (I forgot who of them filed the bug)
[17:24] <markey> I blame wheels
[17:25] <Uraeus> ok, I got lot of bad mp3's which gives me errors in rb, I bugzilla them when I get back from the gym
[17:25] Nick change: Uraeus -> Ura_gym
[17:26] <dolphy> BBB: what gst-launch line ?
[17:26] <dolphy> BBB: gst-plugins/examples/gstplay/player takes the same cpu usage than gst-totem
[17:26] <dolphy> BBB: so it's not a matter of GUI
[17:26] <dolphy> BBB: it's just the pipeline
[17:27] <dolphy> BBB: but if you run a lightweight pipeline in gst-launch for sure it will take less cpu
[17:27] <BBB> gst-launch filesrc location=file.avi ! avidemux .video_00 ! { queue ! ffdec_mpeg4 ! ffcolorspace ! ximagesink } avidemux0.audio_00 ! { queue ! mad ! osssink }
[17:27] <dolphy> ok that's a lightweight pipeline
[17:27] <dolphy> no audio convert
[17:27] <dolphy> no videobalance
[17:27] <dolphy> no scaler
[17:27] <BBB> scaler doesn't work
[17:27] <BBB> did I mention that yet?
[17:27] <BBB> totem always gives me a subwindow of the size of the original video
[17:27] <BBB> it doesn't scale
[17:27] <jdahlin> http://lists.gnome.org/archives/desktop-devel-list/2003-August/msg00066.html
[17:28] <BBB> audioconvert takes no noticeable CPU
[17:28] <BBB> I'm not sure about videobalance
[17:28] <BBB> point is: try to figure out which element takes so much CPu
[17:28] <BBB> I'm asking you because you know the gstplay pipeline better than me
[17:28] <BBB> if we know that, I can try to optimize it
[17:28] <BBB> ok?
[17:28] km_mtg ([email protected]) left irc: Client Quit
[17:29] thomasvs ([email protected]) joined #gstreamer.
[17:31] <BBB> becuase a 70% CPU is acceptable for a divx/mp3 AVI full-PAL
[17:31] <BBB> but 100% isn't
[17:31] <Company> jdahlin: doing what havoc wants in that mail is totally impossible
[17:31] <BBB> not on a 1GHz machine
[17:31] <BBB> especially not if xine/mplayer don't have that issue
[17:32] hallibaby ([email protected]) joined #gstreamer.
[17:32] hallibaby ([email protected]) left irc: Client Quit
[17:33] trow ([email protected]) joined #gstreamer.
[17:34] <dolphy> BBB: here xine takes the same cpu than gst-totem
[17:34] <dolphy> BBB: so i consider it not too bad
[17:34] <dolphy> BBB: here scaler works since the beginning
[17:34] <dolphy> BBB: you always had something strange on your setup
[17:34] <dolphy> BBB: with this 16x16 window
[17:35] <dolphy> BBB: i never got that
[17:35] <dolphy> anyway i m off for some meeting
[17:35] <BBB> I don't have that, that was related to something wrong with my CVS checkout of switch
[17:35] <BBB> I don't have that anymore
[17:35] <BBB> but totem + ximagesink doesn't scale here
[17:35] <BBB> I'm sorry
[17:36] <dolphy> weird
[17:36] <dolphy> and with gst-plugins/examples/gstplay/player ?
[17:36] <dolphy> you tried
[17:36] <dolphy> ?
[17:36] <dolphy> resizing the toplevel xwindow created
[17:37] <BBB> not yet, will try tonight
[17:37] <BBB> you use xvimagesink, don't you?
[17:37] <BBB> go to the meeting and test afterwards, ok?
[17:38] <BBB> and ximagesink still crashes if you resize it, try to reproduce that too
[17:38] <jdahlin> I've noticed crashes while trying to resize
[17:38] <dolphy> i m always using ximagesink
[17:38] <jdahlin> not sure if it's the latest version
[17:38] <BBB> (if you resize it "fanatically"
[17:38] <BBB> see the bugzilla report
[17:38] <BBB> jdahlin: exactly
[17:40] <BBB> why does it cost me so much effort to get a simple "hey, I can reproduce this" or "no, I cannot" from you guys if I email the list, file bugs on bugzilla etc.?
[17:41] <jdahlin> BBB, maybe because people assume things are broken
[17:41] <jdahlin> BBB, I can't resize, it's probably well know (etc)
[17:41] <dolphy> i just tried once more here
[17:41] <dolphy> and it works
[17:41] <jdahlin> known
[17:41] <dolphy> BBB: send me a shot of your totem not scaling
[17:41] <BBB> dolphy: will do tonight
[17:42] <dolphy> BBB: i haven't been able to run totem since it requires gtk 2.3
[17:42] <BBB> oh
[17:42] <dolphy> BBB: i hope i will get up to speed soon
[17:42] Action: BBB kills himself
[17:42] <dolphy> BBB: but still it should work
[17:42] <dolphy> BBB: it was working before
[17:42] <BBB> no it doesn't
[17:42] <dolphy> BBB: and gst-plugins/examples/gstplay keeps working
[17:43] <BBB> and since you're the only one saying it does but you never actually tested it recently, I'm quite sure it really just doesn't
[17:43] <BBB> it might be a totem bug
[17:43] <BBB> but still
[17:43] <BBB> bug == bug
[17:43] <BBB> and instead of saying that it doesn't, next time *PLEASE* just try to actually reproduce it, I've said this over and over again
[17:44] <BBB> if gst-player works, that's fine with me
[17:44] <BBB> but gnome people use totem
[17:44] <BBB> so I actually test totem
[17:47] Action: BBB goes back to labwork
[17:47] <BBB> I'll try to file some more bugs tonight
[17:48] <BBB> bbl
[17:48] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[17:49] <alley_cat> totem has never scaled for me ever with gstreamer
[17:50] foser ([email protected]) joined #gstreamer.
[18:05] bilboed ([email protected]) left irc: Remote closed the connection
[18:08] yippi ([email protected]) left irc: Remote closed the connection
[18:09] <markey> volume mixer works wonderfully
[18:09] yippi ([email protected]) joined #gstreamer.
[18:13] jonathang ([email protected]) joined #gstreamer.
[18:14] mxpxpod ([email protected]) joined #gstreamer.
[18:14] <iain> totem works here
[18:14] <iain> scales fine
[18:15] <iain> but mov and asf files only play audio and pausing them gives a hard error
[18:15] <iain> and mpeg's seem to get out of sync on seeking
[18:21] <iain> and avis do the same as mov and asf too
[18:24] <jonathang> the links for gst-player on the home page are invalid
[18:25] <jonathang> they point to "http://gstreamer.freedesktop.org/src/gstreamer/gst-player-0.8.0.tar.{gz,bz2}
[18:25] <jonathang> while they should point to "http://gstreamer.freedesktop.org/src/gst-player/gst-player-0.8.0.tar.{gz,bz2}"
[18:26] <Company> have we even released gst-player 0.8 yet?
[18:26] <jonathang> it's announced
[18:26] <jonathang> and i've just downloaded the tarball
[18:27] Action: jdahlin looks in topic
[18:27] Action: Company hides
[18:30] <jdahlin> jonathang: thanks for spotting it, should be fixed in a couple of seconds
[18:30] <jonathang> jdahlin: ok
[18:32] snikkt ([email protected]) left irc: Remote closed the connection
[18:50] jonathang ([email protected]) left irc: "Leaving"
[18:59] <markey> oh
[19:00] <markey> my engine will only play files of the first type I'm playing after loading
[19:00] <markey> say when I play an mp3 first, I can't play .mod afterwards
[19:00] <markey> I guess I need to reset the spider or something?
[19:00] <jdahlin> You have to recreate the pipeline
[19:00] <markey> ah..
[19:00] <markey> unref and recreate?
[19:01] <jdahlin> recreate, I think
[19:01] <jdahlin> I'm no spider guru
[19:01] <Company> yeah
[19:01] <markey> maybe there's some spider reset feature, that would rock ;)
[19:01] <Company> spider can't cope with reusing
[19:01] <benow> what happened to httpsrc?
[19:01] <markey> spider.reconfigure() ;)
[19:01] <jdahlin> maybe it should be enough by removing and inserting spider
[19:01] <Company> benow: use gnomevfssrc
[19:01] <jdahlin> instead of the whole pipeline
[19:01] <benow> Company: thx
[19:01] <Company> yeah
[19:02] <Company> removing and reinserting spider should work
[19:02] <jdahlin> but, it's a bit more complex
[19:02] <markey> Company: thx, I'll try
[19:02] <jdahlin> since you'll have to handle the links by yourself
[19:02] <Company> and once we have a working autoplugger... :/
[19:02] <markey> isn't spider the same as autoplugging?
[19:02] <Company> yes
[19:02] <Company> but it's not the same as "working"
[19:02] <Company> ;)
[19:02] <markey> heh
[19:03] <markey> so far the spider worked nicely
[19:03] <markey> can't complain
[19:05] <markey> after removing elements, I have to re-link I assume?
[19:05] <markey> guess so
[19:06] <Company> yeah
[19:06] <Company> you should probably unlink, too
[19:06] Action: Company doesn't know if that happens automagically - it should, but who knows ;)
[19:06] ChrisHJW ([email protected]) left irc: 
[19:07] <markey> ok
[19:07] <jdahlin> woho, I got a little bit further on my nsf plugin :)
[19:08] <Company> nsf?
[19:08] <jdahlin> nintendo sound file
[19:08] <jdahlin> like sid
[19:08] <Company> ah
[19:08] <jdahlin> I have a little stupid question
[19:08] <jdahlin> write(audiofd, GST_BUFFER_DATA(buffer), bufferSize);
[19:08] <jdahlin> What can I replace that with?
[19:08] <jdahlin> (using oss directly right now)
[19:08] <jdahlin> gst_pad_push() ?
[19:13] <Company> yeah
[19:13] <markey> isn't sid rather c64 music than nintendo? :)
[19:14] <jdahlin> it is, but nsf is a similar format
[19:19] <jdahlin> My introduction to gst-plugins :-)
[19:20] <benow> sweet. /me is pulling down BeNOW stream via java wrapped gstreamer, woot.  
[19:20] Gatak ([email protected]) joined #gstreamer.
[19:21] <Gatak> Hello
[19:21] <jdahlin> tjenare
[19:21] <benow> the original 'BeNOW now broadcasting with gstreamer' mp3s around anywhere?
[19:21] <Gatak> Could someone help me with gst-plugins. The compilation fails with this line: ld: cannot find -lgstinterfaces-0.8collect2: ld returned 1 exit status
[19:22] <Gatak> jdahlin: :)
[19:22] <jdahlin> Gatak: during what stage of the complication?
[19:23] <Company> benow: http://gstreamer.freedesktop.org/media/
[19:23] jcsston ([email protected]) joined #gstreamer.
[19:23] <Gatak> Well, actually I am using Gentoo and the gst-plugins ebuild for it. But it seem to be during installationa phase
[19:24] <jdahlin> Report it to the maintainer for gst-plugins in gentoo
[19:26] <Ura_gym> that is foser I think, he is often in here
[19:26] Nick change: Ura_gym -> Uraeus
[19:26] <Uraeus> jcsston: did Many get anywhere with mingw?
[19:30] jcsston` ([email protected]) joined #gstreamer.
[19:35] <Gatak> jdahlin: It is already reported. Thanks. http://bugs.gentoo.org/show_bug.cgi?id=45412
[19:36] <Company> jdahlin: i changed the GstValues to not be subtypes of boxed anymore
[19:37] <Company> jdahlin: GST_TYPE_INT_RANGE is derived from G_TYPE_INT now (just because i need to derive from something) and that works fine, too
[19:37] jcsston ([email protected]) left irc: Read error: 60 (Operation timed out)
[19:38] <jdahlin> Company: maybe have G_TYPE_INT and G_TYPE_DOUBLE when it's appropriate?
[19:38] <jdahlin> Company: G_TYPE_STRING for GstFourcc would probably make sense
[19:38] <Company> jdahlin: fourcc is an int, so i've used int
[19:39] <Company> jdahlin: but it's not really right to derive them from there either
[19:39] <jdahlin> ok
[19:39] <jdahlin> Company: I'm not sure how to do it cleanly either
[19:40] <Company> nobody should rely on the hierarchy in that case anyway
[19:41] <jdahlin> right
[19:41] <Company> that measn calling g_boxed_copy on a GST_VALUE_INT_RANGE is wrong
[19:41] <jdahlin> True, since it's not really a boxed type.
[19:42] <Company> those types should probably be fundamental types, but i'm not sure there
[19:42] Gatak ([email protected]) left #gstreamer.
[19:43] <jdahlin> I'm not sure if non-glib libraries are allowed to register fundamental types
[19:43] <Company> is there a way to throttle cpu usage of certain processes?
[19:43] <Company> they are
[19:43] <Company> they wouldn't export an API for it otherwise ;)
[19:43] <jdahlin> true
[19:46] <Uraeus> jcsston`: do you have any screenshot of gstreamer on windows?
[19:51] Uraeus ([email protected]) got netsplit.
[19:51] spyder482 ([email protected]) got netsplit.
[19:51] thaytan ([email protected]) got netsplit.
[19:51] sxpert ([email protected]) got netsplit.
[19:51] dolphy ([email protected]) got netsplit.
[19:51] ds ([email protected]) got netsplit.
[19:51] Misirlou ([email protected]) got netsplit.
[19:51] markey ([email protected]) got netsplit.
[19:51] bitshifter ([email protected]) got netsplit.
[19:51] jcsston` ([email protected]) got netsplit.
[19:51] foser ([email protected]) got netsplit.
[19:51] sjoerd ([email protected]) got netsplit.
[19:51] kwm ([email protected]) got netsplit.
[19:51] NFusi0n ([email protected]) got netsplit.
[19:51] hyriand ([email protected]) got netsplit.
[19:51] ds-work ([email protected]) got netsplit.
[19:51] tromey ([email protected]) got netsplit.
[19:51] Jara[zZ] ([email protected]) got netsplit.
[19:51] sri ([[email protected]) got netsplit.
[19:51] grub_booter ([email protected]) got netsplit.
[19:51] trow ([email protected]) got netsplit.
[19:51] iain ([email protected]) got netsplit.
[19:51] Company ([email protected]) got netsplit.
[19:51] sublett ([email protected]) got netsplit.
[19:51] bluejay ([email protected]) got netsplit.
[19:51] benow ([email protected]) got netsplit.
[19:51] danb ([email protected]) got netsplit.
[19:51] mxpxpod ([email protected]) got netsplit.
[19:51] ChrisHJW_log ([email protected]) got netsplit.
[19:51] sxpert_work ([email protected]) got netsplit.
[19:51] aldug ([email protected]) got netsplit.
[19:51] sub_pop ([email protected]) got netsplit.
[19:51] jdahlin ([email protected]) got netsplit.
[19:51] Rotty ([email protected]) got netsplit.
[19:51] Marsupilami23 ([email protected]) got netsplit.
[19:51] joem ([email protected]) got netsplit.
[19:51] wtay_ ([email protected]) got netsplit.
[19:51] shawarma ([email protected]) got netsplit.
[19:51] yippi ([email protected]) got netsplit.
[19:51] smoke ([email protected]) got netsplit.
[19:51] walters ([email protected]) got netsplit.
[19:51] md` ([email protected]) got netsplit.
[19:51] apoc ([email protected]) got netsplit.
[19:51] steve_b ([email protected]) got netsplit.
[19:51] jimmy_dean ([email protected]) got netsplit.
[19:51] taaz ([email protected]) got netsplit.
[19:51] dilinger ([email protected]) got netsplit.
[19:57] wtay_ ([email protected]) got lost in the net-split.
[19:57] dilinger ([email protected]) got lost in the net-split.
[19:57] taaz ([email protected]) got lost in the net-split.
[19:57] shawarma ([email protected]) got lost in the net-split.
[19:57] ds ([email protected]) got lost in the net-split.
[19:57] jimmy_dean ([email protected]) got lost in the net-split.
[19:57] aldug ([email protected]) got lost in the net-split.
[19:57] sxpert_work ([email protected]) got lost in the net-split.
[19:57] sri ([[email protected]) got lost in the net-split.
[19:57] grub_booter ([email protected]) got lost in the net-split.
[19:57] Jara[zZ] ([email protected]) got lost in the net-split.
[19:57] danb ([email protected]) got lost in the net-split.
[19:57] tromey ([email protected]) got lost in the net-split.
[19:57] ds-work ([email protected]) got lost in the net-split.
[19:57] joem ([email protected]) got lost in the net-split.
[19:57] bitshifter ([email protected]) got lost in the net-split.
[19:57] hyriand ([email protected]) got lost in the net-split.
[19:57] Marsupilami23 ([email protected]) got lost in the net-split.
[19:57] benow ([email protected]) got lost in the net-split.
[19:57] NFusi0n ([email protected]) got lost in the net-split.
[19:57] steve_b ([email protected]) got lost in the net-split.
[19:57] sxpert ([email protected]) got lost in the net-split.
[19:57] bluejay ([email protected]) got lost in the net-split.
[19:57] Misirlou ([email protected]) got lost in the net-split.
[19:57] thaytan ([email protected]) got lost in the net-split.
[19:57] Rotty ([email protected]) got lost in the net-split.
[19:57] jdahlin ([email protected]) got lost in the net-split.
[19:57] dolphy ([email protected]) got lost in the net-split.
[19:57] sub_pop ([email protected]) got lost in the net-split.
[19:57] sublett ([email protected]) got lost in the net-split.
[19:57] ChrisHJW_log ([email protected]) got lost in the net-split.
[19:57] Company ([email protected]) got lost in the net-split.
[19:57] iain ([email protected]) got lost in the net-split.
[19:57] kwm ([email protected]) got lost in the net-split.
[19:57] apoc ([email protected]) got lost in the net-split.
[19:57] md` ([email protected]) got lost in the net-split.
[19:57] walters ([email protected]) got lost in the net-split.
[19:57] markey ([email protected]) got lost in the net-split.
[19:57] smoke ([email protected]) got lost in the net-split.
[19:57] sjoerd ([email protected]) got lost in the net-split.
[19:57] spyder482 ([email protected]) got lost in the net-split.
[19:57] Uraeus ([email protected]) got lost in the net-split.
[19:57] trow ([email protected]) got lost in the net-split.
[19:57] foser ([email protected]) got lost in the net-split.
[19:57] yippi ([email protected]) got lost in the net-split.
[19:57] mxpxpod ([email protected]) got lost in the net-split.
[19:57] jcsston` ([email protected]) got lost in the net-split.
[19:57] sub_pop ([email protected]) joined #gstreamer.
[19:58] lilo ([email protected]) got netsplit.
[19:58] Gatak ([email protected]) joined #gstreamer.
[19:58] ds ([email protected]) joined #gstreamer.
[19:58] Misirlou ([email protected]) joined #gstreamer.
[19:58] jcsston` ([email protected]) joined #gstreamer.
[19:58] mxpxpod ([email protected]) joined #gstreamer.
[19:58] foser ([email protected]) joined #gstreamer.
[19:58] trow ([email protected]) joined #gstreamer.
[19:58] sjoerd ([email protected]) joined #gstreamer.
[19:58] markey ([email protected]) joined #gstreamer.
[19:58] kwm ([email protected]) joined #gstreamer.
[19:58] iain ([email protected]) joined #gstreamer.
[19:58] Company ([email protected]) joined #gstreamer.
[19:58] ChrisHJW_log ([email protected]) joined #gstreamer.
[19:58] sublett ([email protected]) joined #gstreamer.
[19:58] bluejay ([email protected]) joined #gstreamer.
[19:58] NFusi0n ([email protected]) joined #gstreamer.
[19:58] benow ([email protected]) joined #gstreamer.
[19:58] hyriand ([email protected]) joined #gstreamer.
[19:58] bitshifter ([email protected]) joined #gstreamer.
[19:58] ds-work ([email protected]) joined #gstreamer.
[19:58] tromey ([email protected]) joined #gstreamer.
[19:58] aldug ([email protected]) joined #gstreamer.
[19:58] sxpert_work ([email protected]) joined #gstreamer.
[19:58] sri ([[email protected]) joined #gstreamer.
[19:58] grub_booter ([email protected]) joined #gstreamer.
[19:58] Jara[zZ] ([email protected]) joined #gstreamer.
[19:58] danb ([email protected]) joined #gstreamer.
[19:58] jdahlin ([email protected]) joined #gstreamer.
[19:58] Rotty ([email protected]) joined #gstreamer.
[19:58] Marsupilami23 ([email protected]) joined #gstreamer.
[19:58] joem ([email protected]) joined #gstreamer.
[19:58] wtay_ ([email protected]) joined #gstreamer.
[19:58] shawarma ([email protected]) joined #gstreamer.
[19:58] yippi ([email protected]) joined #gstreamer.
[19:58] smoke ([email protected]) joined #gstreamer.
[19:58] walters ([email protected]) joined #gstreamer.
[19:58] md` ([email protected]) joined #gstreamer.
[19:58] apoc ([email protected]) joined #gstreamer.
[19:58] steve_b ([email protected]) joined #gstreamer.
[19:58] jimmy_dean ([email protected]) joined #gstreamer.
[19:58] dilinger ([email protected]) joined #gstreamer.
[19:58] taaz ([email protected]) joined #gstreamer.
[20:00] wtay_ ([email protected]) left irc: Read error: 113 (No route to host)
[20:00] sxpert ([email protected]) joined #gstreamer.
[20:00] Uraeus ([email protected]) joined #gstreamer.
[20:00] spyder482 ([email protected]) joined #gstreamer.
[20:00] thaytan ([email protected]) joined #gstreamer.
[20:01] wtay_ ([email protected]) joined #gstreamer.
[20:01] lilo ([email protected]) returned to #gstreamer.
[20:02] <Gatak> Tried with Libtool 1.5.2-r5 and now the ebuild for gst-plugins works.
[20:02] <taaz> jdahlin: have you been trying python gstplay stuff on a smp or hyperthreaded machine?
[20:02] <jdahlin> taaz: nope, I do not have access to one
[20:02] <jdahlin> taaz: try using the basicgthread scheduler
[20:06] benow ([email protected]) got netsplit.
[20:06] bluejay ([email protected]) got netsplit.
[20:06] iain ([email protected]) got netsplit.
[20:06] trow ([email protected]) got netsplit.
[20:06] danb ([email protected]) got netsplit.
[20:06] Company ([email protected]) got netsplit.
[20:06] sublett ([email protected]) got netsplit.
[20:07] Gatak ([email protected]) left #gstreamer.
[20:08] jdahlin ([email protected]) left irc: "Leaving"
[20:10] trow ([email protected]) returned to #gstreamer.
[20:10] iain ([email protected]) returned to #gstreamer.
[20:10] Company ([email protected]) returned to #gstreamer.
[20:10] sublett ([email protected]) returned to #gstreamer.
[20:10] bluejay ([email protected]) returned to #gstreamer.
[20:10] benow ([email protected]) returned to #gstreamer.
[20:10] danb ([email protected]) returned to #gstreamer.
[20:13] yippi ([email protected]) left irc: "Client exiting"
[20:13] lupusBE ([email protected]) joined #gstreamer.
[20:37] <Uraeus> walters: I think the RB seek bar should be moved a little more away from the volume button. I have to stop myself from time to time to drag the seeker trying to adjust volume. 
[20:40] <ds-work> !
[20:40] benow ([email protected]) left irc: "Guiding the meefalo to glorious meef free living"
[20:40] <Uraeus> hi ds-work
[20:41] dolphy ([email protected]) joined #gstreamer.
[20:41] <Uraeus> hi dolphy
[20:48] jcsston` ([email protected]) left irc: Read error: 60 (Operation timed out)
[20:51] yippi ([email protected]) joined #gstreamer.
[20:55] spyder482 ([email protected]) left irc: Read error: 110 (Connection timed out)
[20:59] mathrick ([email protected]) joined #gstreamer.
[20:59] <mathrick> ho
[21:12] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[21:25] teuf ([email protected]) joined #gstreamer.
[21:38] <markey> doh, mikmod plugin can't seek?
[21:38] <markey> we need a modplug plugin
[21:39] <ds-work> you mean, like, say, modplug?
[21:39] <markey> does it exist?
[21:40] <ds-work> yes
[21:40] <markey> cool :)
[21:40] <Company> that's because mikmod sucks :/
[21:41] <Company> i need shell help
[21:41] <Company> $FILE = "file.mp3"
[21:41] <Company> i want it without the ".mp3"
[21:41] <Company> how do i write that in bash?
[21:41] <ds-work> $(basename $FILE .mp3)
[21:42] <teuf> "${1%shn}flac" converts a trailing shn to flac
[21:42] <ds-work> that will strip off directories, too
[21:43] <Company> thx
[21:45] <markey> when I try to set the pipeline state to GST_STATE_NULL in my eos callback, it all crashes
[21:45] <markey> used to work IIRC
[21:46] Action: markey puzzled regarding eos
[21:46] <Company> that's why i said you should register an idle handler that does that
[21:46] <Company> doing stuff from the pipeline from callbacks directly is tricky
[21:47] <markey> well I didn't really grok it ;)
[21:50] <Company> i dunno what the equivalent of g_idle_add is in KDE land
[21:53] <markey> Adds a function to be called whenever there are no higher priority events pending to the default main loop.
[21:53] <markey> ah yes
[21:55] <markey> I wonder why eos doesn't automatically set the pipeline to STATE_READY
[21:55] <markey> it's a bit weird isn't it
[21:56] <Company> it shouldn'T even set it to paused in an ideal world
[21:56] <Company> sometimes elements want to seek back on EOS
[21:57] <Company> currently this doesn't work
[22:04] spyder482 ([email protected]) joined #gstreamer.
[22:06] <Company> it's interesting that modern video container formats don't carry resyncing-after-error information
[22:07] <ds-work> that's because they're not meant for error-prone streaming
[22:07] <ds-work> like MPEG is
[22:07] <Company> isn't MPEG4 just QT?
[22:09] <Company> and ASF was supposed to do streaming, too
[22:10] <ds-work> MPEG4 can't be used in the same way that MPEG program streams can
[22:11] <mathrick> ASF doesn't include syncing info on each keyframe?
[22:11] <mathrick> I thought it did
[22:11] <Company> it doesn't
[22:11] <Company> unless i've missed something
[22:12] <mathrick> hmm, ASF was designed for streaming, that's strange
[22:12] <Company> the difference between keyframe and not keyframe is 1 bit
[22:12] <mathrick> and what about content?
[22:12] <Company> asf does some interesting other things
[22:13] <Company> but i wouldn'T call them clever
[22:13] <mathrick> like?
[22:13] Action: mathrick MSDNs for ASF specs
[22:13] <Company> they have a 1:1 asf packet/udp packet mapping
[22:13] thaytan ([email protected]) left irc: No route to host
[22:14] <Company> and each asf packet identifies itself with a number
[22:14] <Company> so you can easily "resync" after dropped packets
[22:16] <Company> the best format is hands down ogg
[22:16] <Company> though ogg has 2 other problems
[22:17] Action: mathrick wonders why OOo doesn't do that nifty toc-is-hypertext thing
[22:17] <Company> 1) syncing for muxing requires knowledge about the formats, just doing demux => mux doesn't work
[22:17] Blaasvis ([email protected]) joined #gstreamer.
[22:17] <Company> 2) you require the beginning of the file/stream
[22:17] Action: Blaasvis is not happy... 
[22:17] <Company> i guess that makes mp3 the best format
[22:17] Action: Company gives Blaasvis a cookie
[22:17] <Blaasvis> i want to release totem for arch linux 
[22:17] <Company> better?
[22:18] <teuf> Company: I'm really really unhappy too then
[22:18] <mathrick> huh, that sounds pretty dumb, especially 2)
[22:18] Action: mathrick too
[22:18] <Blaasvis> with gstreamer enabled, but it plays my DVD's very slow
[22:18] Action: Blaasvis pets Company 
[22:18] Action: Company goes shopping
[22:18] <Blaasvis> much better :)
[22:18] <mathrick> Blaasvis: like, 1fps?
[22:19] <Blaasvis> yeah, it goes shock shock... no sound at all
[22:19] <mathrick> hmm, slow-mpeg-playback strikes back?
[22:20] <mathrick> i thought ronald fix^H^H^Hhacked around it?
[22:20] <Blaasvis> i was planning on beating up ronald aka beelzebub  
[22:20] <ds-work> try --gst-scheduler=basicgthread
[22:20] <Blaasvis> but i didn't seen him for a while in our dutch chat channel :)
[22:21] <Blaasvis> i will 
[22:21] <Blaasvis> ds-work, gstreamer it self or the plugins 
[22:21] <ds-work> Blaasvis: totem
[22:22] <Blaasvis> aha... 
[22:22] sub_pop ([email protected]) left irc: Client Quit
[22:26] <mathrick> Company: hmm, maybe syncing was left out to be implemented in per-payload way?
[22:27] <mathrick> looks like M4S2 has resyc flang, dunno what it does though ;)
[22:27] <Blaasvis> ds-work, doesn't work with totem :(
[22:27] je4d ([email protected]) joined #gstreamer.
[22:29] <Blaasvis> any ide's left ?
[22:29] <Blaasvis> idea's 
[22:35] <ds-work> it works here. 'totem --gst-scheduler=basicgthread'
[22:36] <Blaasvis> i thought a compile flag 
[22:40] teuf ([email protected]) left irc: "Client exiting"
[22:43] <Blaasvis> well that didn't help :(
[22:44] markey ([email protected]) left irc: "leaving"
[22:53] <Blaasvis> well thx for the idea... anyway
[22:53] Action: Blaasvis is getting some sleep 
[22:53] Blaasvis ([email protected]) left irc: "sleeping"
[22:58] Zeenix ([email protected]) joined #gstreamer.
[22:58] <Zeenix> hi
[22:58] <mathrick> yo Zeenix
[22:58] jdahlin ([email protected]) joined #gstreamer.
[23:00] <apoc> hey Zeenix
[23:01] <Uraeus> night guys
[23:01] Uraeus ([email protected]) left irc: "Client exiting"
[23:02] <Company> ds-work: is changing an argument of a function to const considered API breakage?
[23:03] <ds-work> Company: it depends on the situation
[23:03] <ds-work> Company: if it was always interpreted as const, then no
[23:03] <Company> gst_util_dump_mem
[23:03] <Company> the first argument should be const
[23:04] <ds-work> yes, make it const
[23:04] <Company> k
[23:05] <ds-work> especially since I assume it's fixing a bug elsewhere
[23:05] bilboed ([email protected]) joined #gstreamer.
[23:10] Zeenix ([email protected]) left irc: Remote closed the connection
[23:29] NFusi0n ([email protected]) left irc: Read error: 110 (Connection timed out)
[23:31] <ds-work> what is Martin Soto's nick?
[23:31] <Company> i don't think he's on irc
[23:32] <ds-work> that would explain why I can't remember it
[23:37] jdahlin ([email protected]) left irc: "adios"
[23:41] alley_cat ([email protected]) left irc: "May the Source be with you!"
[23:48] lupusBE ([email protected]) left irc: "Bezig met verlaten"
[23:51] walters ([email protected]) left irc: Client Quit
[00:00] --- Tue Mar 23 2004
[00:03] beerboy ([email protected]) joined #gstreamer.
[00:03] <beerboy> What's gstreamer's most recent version?
[00:04] beerboy ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[00:05] kwm ([email protected]) left irc: "sleeping"
[00:06] <ds-work> 0.8.0
[00:10] <Company> 2 minutes too late
[00:15] yippi ([email protected]) left irc: "Client exiting"
[00:16] <foser> should gstreamer install interface headers even when it installs none of them , so it can use later added interface based plugins ?
[00:18] <ds-work> foser: er, wha?
[00:19] <foser> ds-work: if i install gst-plugins without any optional plugins, i seem to miss some needed headers..
[00:19] <ds-work> all headers are installed unconditionally, iirc
[00:19] <ds-work> except one, which I fixed the other day
[00:21] thaytoo ([email protected]) joined #gstreamer.
[00:21] <foser> which was ?
[00:22] <ds-work> xoverlay
[00:22] <foser> hmm that might be the one yep
[00:23] <foser> ds-work: when exactly is it important to have that one ? when building any xoverlay plugin ?
[00:23] <Company> s/plugin/app/ unfortunately
[00:24] <Company> and i think gst-player and totem need it
[00:24] <foser> right.. yes
[00:24] <foser> are you aware of any specific libtool version being needed for this release ?
[00:25] <ds-work> for the release, libtool is not needed or used
[00:26] <thaytoo> good morning
[00:26] <foser> hmm right.. must be the gentoo magic then..
[00:30] <Company> morning thaythan
[00:32] sublett ([email protected]) left irc: Read error: 60 (Operation timed out)
[00:37] Shoragan ([email protected]) joined #gstreamer.
[00:40] <Company> if i want to check if the LSB is set, do i need to do (byte & 1) or (byte & 128) ?
[00:41] Action: ds-work comments that 1 is "less" than 128, so 1 would be the bit of lesser significance
[00:42] <Company> it's a yes or no question, so i always ask
[00:42] <Company> i always get those wrong
[00:42] <ds-work> 1
[00:52] <Company> i did that right then
[00:53] <iain> hmmm, once I get marlin saving again, I'm so cutting out samples from this sam 'n' max trailer
[00:53] <iain> "Why...wont...you...just...DIE!"
[00:54] <Company> poor iain
[00:54] <Company> but great it works again
[00:57] <iain> yeah, saving will just be a matter of replacing the f2i element with audioconvert ! interleave
[00:57] <iain> assuming they work ;p
[00:57] jcsston ([email protected]) joined #gstreamer.
[00:57] ct_ ([email protected]) joined #gstreamer.
[00:58] Action: ct_ is away: out for a bit
[01:30] Shoragan ([email protected]) left irc: "Leaving"
[01:30] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[01:47] iain ([email protected]) left irc: "!"
[01:47] Rotty ([email protected]) left irc: Remote closed the connection
[01:59] <mathrick> nite
[02:01] Nick change: mathrick -> mathrick|sleep
[02:14] harshy ([email protected]) joined #gstreamer.
[02:20] foser ([email protected]) left irc: "[ I want to believe ]"
[02:25] lilo ([email protected]) left irc: "Lost terminal"
[02:26] lilo ([email protected]) joined #gstreamer.
[02:29] Misirlou ([email protected]) left irc: "asdf"
[02:29] Misirlou ([email protected]) joined #gstreamer.
[02:34] Action: ds-work wonders why rb lost its ability to seek
[02:34] <Company> in general or only for mp3?
[02:35] smoke ([email protected]) got netsplit.
[02:35] <ds-work> ogg
[02:38] Action: ct_ is back (gone 01:39:06)
[02:40] smoke ([email protected]) got lost in the net-split.
[02:45] <Company> seeks fine here
[02:45] <Company> both ogg and mp3
[02:56] <ds-work> mp3 works
[02:56] <ds-work> odd
[02:57] BOFH ([email protected]) joined #gstreamer.
[02:58] <BOFH> anyone know how to fix this: There is no element present to handle the stream's mime type application/x-id3.
[02:59] <thaytoo> install the mad plugin


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