IRC Logs

*******************************************************************
[03:01] <Company> i'm gonna commit this stuff now
[03:02] <Company> oggdemux, vorbisdec and friends
[03:03] <ds-work> woohoo
[03:13] <Company> ?
[03:14] <Company> ah, you meant my comment?
[03:14] <Company> well, there's still lots missing
[03:14] <ds-work> yes
[03:14] <Company> vorbisenc and oggmux
[03:15] <Company> vorbisenc should be renamed vorbisfileenc or something btw
[03:37] hadley ([email protected]) joined #gstreamer.
[03:41] LeRoutier ([email protected]) joined #gstreamer.
[03:42] <LeRoutier> hello
[03:42] <LeRoutier> still someone around ?
[03:42] <LeRoutier> well, i suppose that even if not there, you log traffic when you're idle
[03:43] <ds-work> some do
[03:43] <ds-work> I log traffic in 3 places :o
[03:43] <LeRoutier> hi ds-work | in gstreamer/docs/random/mimetypes : decoder for DivX is :     Decoder: dvdreadsrc, dvdnavsrc
[03:43] <Company> i don't
[03:44] <Company> chat is about talking, not reading up
[03:45] <ds-work> I've been ignoring a lot of the backlog recently, since traffic has gone up
[03:48] <Company> information that should be archived goes into the mailing list, stuff that people need to be reminded of goes into bugzilla :)
[03:50] <ds-work> that leaves a category of stuff that I need to know but doesn't need to be archived
[03:51] <thaytoo> stuff that you need to know is hidden as carefully as possible ;)
[03:52] hadley ([email protected]) left irc: "Leaving"
[03:52] <Company> that needs to be pushed to you
[03:52] <Company> either via constant "ds-work: ping?" or mail directly to you
[03:58] <LeRoutier> i just mailed BBB for the various doc problems i've found as this doc end with "Blame Ronald Bultje <[email protected]> aka BBB for any mistakes in
[03:58] <LeRoutier> this document."
[03:58] KA_afk ([email protected]) got netsplit.
[03:59] <LeRoutier> well, going to bed, it's 4AM. good night guys (and gals if any, which i doubt)
[03:59] LeRoutier ([email protected]) left irc: "Leaving"
[03:59] KA_afk ([email protected]) returned to #gstreamer.
[04:08] thaytoo ([email protected]) left irc: "leaving"
[04:09] <Company> hm, oggmux should be really simple
[04:09] <Company> apart from that you can't seek when muxing oggs
[04:10] <Company> but other than that you put data of buffers from different streams into the ogg stream
[04:10] <Company> which is buffer with lowest timestamp in, data to push out
[04:12] <Company> the only problem is knowing if it's a proper ogg format
[04:20] <C> why is it that i get 1000 errors when all i did was gst_init (&argc, &argv);??
[04:27] <C> nm, forgot to copy a header directory
[04:38] aeyakovenko ([email protected]) joined #gstreamer.
[04:38] <aeyakovenko> how do i remove an element from a bin?
[04:40] <Company> gst_bin_remove
[04:41] <aeyakovenko> cool, thanks
[05:00] harshy ([email protected]) left irc: "I quit for now"
[05:10] Nick change: bluejay|busy -> bluejay
[05:18] harshy ([email protected]) joined #gstreamer.
[05:20] walters ([email protected]) joined #gstreamer.
[05:43] mathrick|sleep ([email protected]) left irc: Remote closed the connection
[05:55] KA_afk ([email protected]) got netsplit.
[05:58] KA_afk ([email protected]) returned to #gstreamer.
[06:02] KA_afk ([email protected]) got netsplit.
[06:02] KA_afk ([email protected]) returned to #gstreamer.
[06:03] KA_afk ([email protected]) got netsplit.
[06:04] KA ([email protected]) joined #gstreamer.
[06:09] KA_afk ([email protected]) got lost in the net-split.
[06:35] mxpxpod ([email protected]) joined #gstreamer.
[06:41] Marsupilami23 ([email protected]) left irc: "I don't wanna grow up, I'm a Toys R Us Kid. There's a million toys at Toys R Us that I can play with!"
[07:05] taf2_ ([email protected]) joined #gstreamer.
[07:22] taf2 ([email protected]) left irc: Read error: 113 (No route to host)
[07:23] taf2__ ([email protected]) joined #gstreamer.
[07:39] harshy ([email protected]) left irc: Remote closed the connection
[07:42] harshy ([email protected]) joined #gstreamer.
[07:42] taf2_ ([email protected]) left irc: Read error: 110 (Connection timed out)
[07:50] Action: sxpert is back (gone 07:50:48)
[08:04] Nick change: bluejay -> bluejay|sleep
[08:06] mathrick ([email protected]) joined #gstreamer.
[08:06] <mathrick> mornin
[08:15] bilboed ([email protected]) joined #gstreamer.
[08:15] <bilboed> hi all
[08:16] <bilboed> just popping by quickly... Anybody know who is/will be working on the midi plugins ?
[08:17] rr_t3a ([email protected]) joined #gstreamer.
[08:17] rr_t3a ([email protected]) left irc: Client Quit
[08:23] aeyakovenko ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[08:25] ChrisHJW ([email protected]) joined #gstreamer.
[08:38] sublett ([email protected]) joined #gstreamer.
[08:42] mathrick ([email protected]) left irc: Remote closed the connection
[08:55] mathrick ([email protected]) joined #gstreamer.
[09:00] harshy ([email protected]) left irc: "I quit for now"
[09:03] walters ([email protected]) left irc: "out"
[09:04] Nick change: mathrick -> mathrick|Uni
[09:05] harshy ([email protected]) joined #gstreamer.
[09:09] ChrisHJW ([email protected]) left irc: Read error: 110 (Connection timed out)
[09:18] Shoragan ([email protected]) joined #gstreamer.
[09:22] trow ([email protected]) left irc: Remote closed the connection
[09:22] Action: sxpert is away: I'm busy
[09:22] <thomasvs> bilboed: leif morgan johnson
[09:24] harshy ([email protected]) left irc: "I quit for now"
[09:37] Company ([email protected]) left irc: Remote closed the connection
[09:45] Action: sxpert_work is back (gone 16:04:45)
[09:46] sublett ([email protected]) left irc: "I like food, food is good!"
[09:47] harshy ([email protected]) joined #gstreamer.
[09:59] BBB ([email protected]) joined #gstreamer.
[10:02] thomasvs ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:18] thomasvs ([email protected]) joined #gstreamer.
[10:26] dolphy ([email protected]) joined #gstreamer.
[10:29] <dolphy> morning
[10:29] <BBB> howdy
[10:29] Action: BBB fixes random stuff
[10:40] <dolphy> BBB: have you seen my first draft at XOverlay doc ?
[10:41] <thomasvs> taaz: ping ?
[10:43] <BBB> dolphy: not yet, is it online (www)?
[10:43] <BBB> or in CVS?
[10:50] <thomasvs> BBB,dolphy: can you guys run "make clean" from gstreamer/docs just once, so the old html/ files that are renamed aren't synced online anymore ?
[10:51] Nick change: Jara[zZ] -> Jaramir
[10:51] <BBB> sur
[10:51] <BBB> dolphy: can you change one thing please?
[10:51] <BBB> or two
[10:52] <BBB> dolphy: myplugin should ideally be my_filter
[10:52] <BBB> and sinkconnect should be sink_link
[10:52] <BBB> that's just for cinsistency
[10:52] <BBB> thomasvs: done
[10:53] <dolphy> BBB: well an XOverlay is mostly not a filter
[10:53] <dolphy> BBB: my_sink maybe
[10:53] <BBB> that's why it's an example
[10:53] <BBB> just call it my_filter for consistency... I agree it doesn't make much sense, but neither do most of the other examples
[10:53] <BBB> that's why it's examples
[10:54] <dolphy> ok
[10:54] <BBB> one more thing, it might be nice to explain what the purpose of each function is
[10:54] <BBB> it sounds very stupid, but it is probably useful for new people
[10:54] <BBB> but the examples look good and clear, that's very nice :)
[10:55] <dolphy> thx :-)
[10:56] Action: BBB is starting to really like the PWG
[10:56] <BBB> now we need comments from "newbies" so we can adapt and finetune it to their needs
[10:57] Action: BBB doesn't like the term 'newbies'
[10:57] <BBB> but you get the point... ;
[10:57] <BBB> ;)
[10:58] <dolphy> BBB: i ll had a section on upstream events
[10:58] <dolphy> BBB: not so many people implement correctly a handle_src_event method
[10:58] <dolphy> BBB: and that breaks seeking, navigation and so on
[10:58] <BBB> dolphy: please do... that should be in advanced_events.xml
[10:58] <BBB> I want to explain event handling, the use of all events and everything there
[10:58] <BBB> and the difference between src event handling and sink event handing
[10:59] <BBB> +l
[10:59] <BBB> but that's just below tagging on my todo list
[10:59] <BBB> I first want to finish documentation of tagging
[11:01] Action: BBB welcomes all help :)
[11:02] teuf ([email protected]) joined #gstreamer.
[11:03] <teuf> morning
[11:03] <BBB> mornin'
[11:03] <BBB> teuf: wanna comment on the PWG too?
[11:04] <teuf> BBB: haven't had time to look much at it yet
[11:04] <teuf> BBB: the only comment for now is that http://www.freedesktop.org/~gstreamer/data/doc/gstreamer/head/pwg/html/section-preface-reading.html should have a link to mathieu's gobject doc
[11:05] <BBB> where's that?
[11:05] thaytan ([email protected]) joined #gstreamer.
[11:06] <teuf> http://www.le-hacker.org/papers/gobject/index.html
[11:06] <BBB> k
[11:08] <teuf> some links on the main page of http://www.freedesktop.org/~gstreamer/ don't work, is that  known problem ?
[11:09] wheels ([email protected]) left irc: "using sirc version 2.211+KSIRC/1.3.9"
[11:12] <BBB> don't think so... please add those to a bugzilla bug, too... we really need to keep track of all those bugs
[11:13] <dolphy> BBB: yesterday i was looking at the seek event stuff some more
[11:13] <teuf> BBB: I'll read the PWG and compares what it says with my shorten plugin code to see if it helps me understand all the code I copied and pasted in this plugin, and to see if the PWG helps me to port the plugin to the new caps system ;)
[11:14] <dolphy> BBB: and it seemed that the video pad of avidemux was not calling the src_handle_event method when receiving it
[11:14] teuf ([email protected]) left #gstreamer.
[11:14] teuf ([email protected]) joined #gstreamer.
[11:14] <thomasvs> teuf: yeah, I'm working on the site, some work and some don't
[11:14] <thomasvs> teuf: feel free to tell me which
[11:14] <dolphy> BBB: might that be possible that the sometimes pad of avidemux has not been set the right event handler?
[11:14] <dolphy> BBB: i ll look at it deeper today
[11:15] <teuf> thomasvs: the two links on the main page at http://www.freedesktop.org/~gstreamer/ 
[11:16] <teuf> thomasvs: on this same page, if you click on gstreamer logo, you are redirected on freedesktop.org or on freedesktop.org/~gstreamer depending on where you click
[11:17] <BBB> dolphy: could be... is it being called?
[11:18] <BBB> dolphy: the event handler needs to be set before calling element_add_pad(), or else the scheduler will mess up
[11:18] <BBB> bit complicated
[11:21] Rotty ([email protected]) joined #gstreamer.
[11:25] <dolphy> BBB: i ll check that
[11:42] <BBB> k
[12:03] Action: BBB wonders about tag writing
[12:07] Action: BBB pings company for tag writign
[12:07] pb_ ([email protected]) joined #gstreamer.
[12:24] sublett ([email protected]) joined #gstreamer.
[12:26] alley_cat ([email protected]) joined #gstreamer.
[12:39] iain ([email protected]) joined #gstreamer.
[12:43] <dolphy> BBB: you sure a chain function should call gst_pad_event_default in the chain function for unhandled events ?
[12:43] <dolphy> BBB: i thought it was better to just gst_pad_push to src pad the data
[12:46] sublett ([email protected]) left irc: "I like food, food is good!"
[13:02] sublett ([email protected]) joined #gstreamer.
[13:16] <BBB> gst_pad_event_default() does that, plus some more
[13:27] <dolphy> BBB: well yeah but isn't it better to get people to push to the src pad
[13:27] <BBB> no
[13:27] <dolphy> BBB: because if one has multiple src pad
[13:27] <dolphy> BBB: they will use that method which will do it wrong
[13:27] <BBB> gst_pad_event_default() takes care of all that
[13:29] <dolphy> well it's fine for tee
[13:29] <dolphy> but not for a N-to-1 element with a tuner interface for example
[13:30] <dolphy> s/N-to-1/1-to-N/
[13:30] Action: BBB is getting weird build errors
[13:30] <BBB> N-1 is explained in the appropriate section
[13:30] <BBB> and we'll explain there why you can't use gst_pad_event_default()
[13:30] <BBB> but in general, people *should* use it
[13:30] <BBB> that's why we use it in the example
[13:31] <BBB> *examples
[13:32] Action: BBB has tag docs ready
[13:32] Action: BBB continues with events
[13:32] bilboed ([email protected]) left #gstreamer.
[13:33] <dolphy> BBB: i m doing it now
[13:33] <dolphy> BBB: almost finished
[13:34] <BBB> oh, perfect
[13:34] Action: BBB looks what else needs doing
[13:34] <BBB> will you document the tuner interface?
[13:34] <BBB> I'll do another interface then
[13:34] <BBB> colorbalance, apparently
[13:35] <BBB> actually, I'll do tuner, since there's some TV-specific bits in it
[13:35] Action: BBB wants to get that clear
[13:35] Action: thomasvs needs to buy an xslt book
[13:41] <dolphy> BBB: i feel more confortable doing colorbalance than tuner
[13:42] <BBB> ok
[13:42] <BBB> I'll do tuner
[13:42] <BBB> colorbalance is easy ;)
[13:42] <dolphy> yup
[13:42] <dolphy> :)
[13:43] markey ([email protected]) joined #gstreamer.
[13:49] <dolphy> http://www.freedesktop.org/~gstreamer/data/doc/gstreamer/head/pwg/html/chapter-advanced-events.html
[13:50] <dolphy> BBB: check it there
[13:52] <BBB> your code doesn't have a gray background?
[13:54] <BBB> you probably want to explain events through bytestream, too
[13:54] <BBB> (gst_bytestream_get_status())
[13:55] <dolphy> ah i don't know anything about those :)
[13:55] <BBB> anywya, looks good... I asume you'll also explain upstream events and the meaning of eahc event here?
[13:55] <dolphy> feel free to complete 
[13:55] <BBB> oh, ok, I'll add those when you're finishe
[13:55] <BBB> +d
[13:55] <dolphy> well i don't know all events :)
[13:55] <BBB> doesn't matter
[13:55] <dolphy> but i wanted to be sure that PWG contains something telling to people how to keep the upstream event working :)
[13:56] <BBB> make a list of all events (<sect2 ..>) and explain the ones you know
[13:56] <BBB> leave the rest empty
[13:56] <dolphy> ok
[13:56] <BBB> I left some items WRITEME too
[13:56] <BBB> (use WRITEME everywhere so we know which parts need writing ;) )
[13:57] <dolphy> yup
[13:58] <dolphy> i ll maybe also add a section telling about elements with 1-to-N that should not use gst_pad_event_default
[13:58] <dolphy> the PWG is getting quite nice and complete now :)
[13:58] <dolphy> that's a good improvement :)
[13:59] markey ([email protected]) got netsplit.
[13:59] mathrick|Uni ([email protected]) got netsplit.
[13:59] ds-gromit ([email protected]) got netsplit.
[13:59] herzi ([email protected]) got netsplit.
[13:59] bluejay|sleep ([email protected]) got netsplit.
[13:59] taaz ([email protected]) got netsplit.
[13:59] <BBB> dolphy: no! not there
[13:59] <BBB> dolphy: that should *really* be in the part about 1-to-N elements
[14:00] <BBB> event sections explain about events... not about corner case elements that need special care
[14:00] <BBB> that should be in the section on that corner case element type
[14:00] <BBB> ;)
[14:00] markey ([email protected]) returned to #gstreamer.
[14:05] taaz ([email protected]) got lost in the net-split.
[14:05] bluejay|sleep ([email protected]) got lost in the net-split.
[14:05] herzi ([email protected]) got lost in the net-split.
[14:05] ds-gromit ([email protected]) got lost in the net-split.
[14:05] mathrick|Uni ([email protected]) got lost in the net-split.
[14:05] Nick change: KA -> KA_afk
[14:06] <BBB> gr
[14:06] <BBB> how do I do a global search-and-replace in vim?
[14:06] <BBB> s/bla1/bla2/g only does one line
[14:06] Action: BBB forgot the trick
[14:11] <alley_cat> BBB: %s/...
[14:13] <dolphy> BBB: :1,$s/search/replage/g
[14:13] herzi ([email protected]) joined #gstreamer.
[14:13] taaz ([email protected]) joined #gstreamer.
[14:14] ds-gromit ([email protected]) joined #gstreamer.
[14:14] bluejay|sleep ([email protected]) joined #gstreamer.
[14:17] mathrick|Uni ([email protected]) joined #gstreamer.
[14:20] <BBB> dolphy, alley_cat: both thanks
[14:27] <bitshifter> which element would I pick if I wanted an audio volume normaliser of some sorts? volume?
[14:28] Action: BBB knows nothing of audio
[14:28] <BBB> volume just adapts volume
[14:29] <BBB> it doesn't do anything pretty - normalizing sounds very fancy so I guess volume doesn't do that
[14:31] <bitshifter> well, maybe I'm using the wrong term
[14:32] <bitshifter> basically I want some filter that makes sure that the output volume is effectively the same even if the input has different 'volume levels'. Something like replay gain, only that it tries to find the replay gain automatically
[14:39] mathrick|Uni ([email protected]) left irc: Remote closed the connection
[14:44] <dolphy> weee
[14:44] <dolphy> seeking kind of works again
[14:45] <markey> does it?
[14:45] <markey> yay!
[14:45] <BBB> wheeee
[14:45] <markey> I've never got it work
[14:45] Action: markey got already quite frustrated with GST
[14:46] <BBB> :(
[14:46] <markey> well, nothing really works like it should :/
[14:46] <BBB> notihng? that's not true
[14:46] <markey> ok, basic playback works fine
[14:46] <markey> that's about it
[14:46] <BBB> I use gstreamer HEAD in the company that I work for ;)
[14:46] <BBB> that works pretty well
[14:46] <teuf> transcoding audio also works fine
[14:47] <BBB> video transcoding works too, mostly
[14:47] <BBB> most MPEG stuff works fine
[14:47] <markey> each slightly advanced feature that I try fails ;)
[14:47] <BBB> (everything is MPEG here)
[14:47] <markey> like seeking
[14:47] <BBB> right, advanced!
[14:47] Action: teuf never tried gstreamer with video
[14:47] <BBB> seeking in MPEG works
[14:47] <BBB> and dolphy apparently just fixed AVI
[14:47] <markey> for instance, identity.. won't work
[14:47] <BBB> :?
[14:47] <markey> I _need_ to get access to the audio samples
[14:47] <markey> for our visualisations
[14:49] <dolphy> BBB: seeking gets delayed because of queue
[14:49] <dolphy> BBB: what about sending seek event closer to spider 
[14:49] <dolphy> BBB: instead of the sink element
[14:51] <BBB> *you* chose to send it to sink elements
[14:51] <BBB> I'm perfectly fine with sending it into the src pads of spider directly
[14:52] <BBB> markey: identity works perfectly fine here
[14:52] <BBB> I use it for all sort of purposes in MPEG Data
[14:52] <BBB> (you don't want to see my apps, I do all sort of cruel gruesome evil stuff with identities)
[14:52] <markey> well, here it doesn't. you saw my code.. iDunno what's wrong with it ;)
[14:54] <dolphy> BBB: ok
[14:56] <dolphy> BBB: when receiving a seek event
[14:56] <dolphy> BBB: avidemux produces a flush event downstream ?
[14:56] <dolphy> BBB: because seeking works
[14:56] <BBB> yes
[14:56] <BBB> all elements do that
[14:56] <dolphy> BBB: but queues are still feeding video/audio during some seconds before it really skips
[14:56] <BBB> it's to flush cached contents in BP-frame decoders like divx
[14:57] <BBB> so they don't take 'old' frames for B/P frame buildup
[14:57] <BBB> queues should flsuh
[14:57] <BBB> ...
[14:57] <BBB> unless some elements don't handle flushes correctly themselves
[14:58] <dolphy> BBB: ok i ll debug that too then
[14:58] <dolphy> BBB: getting closer ... :)
[15:07] <dolphy> BBB: found...
[15:07] <dolphy> BBB: when seeking the source element is generating a DISCONT event
[15:07] <dolphy> BBB: but no flush
[15:07] <dolphy> BBB: that is the reason i guess ?
[15:11] <BBB> the discont event should contain a flush flag
[15:11] <BBB> and that should be set after a seek
[15:11] <BBB> if avidemux doesn't do that, it's a bug
[15:12] <BBB> (these sort of bugs are one of the reasons why we need to document all events individually)
[15:19] markey ([email protected]) left irc: "leaving"
[15:37] Nick change: KA_afk -> KA
[15:47] <dolphy> BBB: well filesrc is not working like that
[15:47] <dolphy> BBB: it sends a discont
[15:48] <dolphy> BBB: and then a flush
[15:48] <BBB> then filesrc is wrong
[15:48] <dolphy> BBB: not a discont flagged with flush
[15:49] Action: BBB doubts what PWG part to do next
[15:51] <dolphy> BBB: both gnomevfssrc and filesrc are just sending discont with no flush flags
[15:51] Action: dolphy checks the default pad event handling for discont behaviour
[15:53] mxpxpod ([email protected]) left irc: "Guns kill people as much as spoons made Rosie O'Donnell fat.."
[15:54] <dolphy> BBB: why does flush exists then if seek should generate a discont event with a flush flag ?
[15:54] <BBB> because you might want a flush that does not change the clock
[15:54] <BBB> or you might want a discont without flush
[15:54] <BBB> it's all allowed
[15:54] taf2__ ([email protected]) left irc: Read error: 113 (No route to host)
[15:54] <BBB> flush without discont is useful for filesink, for example
[15:54] Rotty ([email protected]) left irc: Remote closed the connection
[15:54] <BBB> it calls fdatasync() or fflush()
[15:55] <dolphy> BBB: i think you are wrong here there's absolutely no flush flag for discont event
[15:55] <dolphy> BBB: aren't you mixing with the seek event flush flag
[15:55] <BBB> discont without the flush flag is useful for internet streams where you lost a few bytes
[15:55] <BBB> eh?
[15:55] <BBB> there should be
[15:55] <BBB> grmbl
[15:55] Action: BBB is going crazy
[15:56] <dolphy> BBB: and anyway queue is checking for GST_EVENT_TYPE being GST_EVENT_FLUSH
[15:56] <dolphy> BBB: not DISCONT event with a flush flag
[15:56] <BBB> ok, I'm probably just completely lost
[15:56] <dolphy> BBB: so i guess you are talking about a new feature not implemented yet :)
[15:56] <BBB> I guess :p
[15:56] <BBB> ok, then let avidemux send a flush event just before the discont event
[15:56] <BBB> or after
[15:56] <dolphy> BBB: ok i ll first fix the actual implementation
[15:56] <BBB> I think before
[15:57] <dolphy> BBB: it's already implemented in that direction but there's a bug
[15:57] <dolphy> BBB: only the discont event is sent
[15:57] <dolphy> BBB: i m checking why
[15:57] <dolphy> BBB: and then the whole thing is fixed
[15:57] <dolphy> BBB: but your idea is interesting btw
[15:57] <dolphy> BBB: adding FLUSH flags to discont
[15:58] <dolphy> BBB: and fix queue so that it flushes on those flagged discont events
[15:58] Action: dolphy starts to understand gstreamer internals better and better everyday :)
[16:01] <teuf> is it possible to start playing a dvd with specific audio and subtitle tracks using gst-launch ?
[16:04] <BBB> yes, but I don't know how ;)
[16:05] <teuf> ok, I should try to use that instead of mplayer to play dvds then ;)
[16:06] <BBB> *g*
[16:06] <BBB> if it works
[16:06] <BBB> ask thaytan
[16:20] <dolphy> BBB: src->need_flush = GST_EVENT_SEEK_FLAGS(event) & GST_SEEK_FLAG_FLUSH;
[16:20] <dolphy> is that the correct way to check if the event has the seek flag ?
[16:23] <BBB> does the macro GST_EVENT_SEEK_HAS_FLAG() exist?
[16:24] <dolphy> #define GST_EVENT_SEEK_FLAGS(event)		(GST_EVENT_SEEK_TYPE(event) & GST_SEEK_FLAGS_MASK)
[16:24] <dolphy> is that similar ?
[16:24] <dolphy> oops
[16:24] <dolphy> forget that :)
[16:25] <dolphy> BBB: nope no such thing
[16:25] <BBB> then just do what you said
[16:25] <BBB> or add the macro if you wish
[16:30] <dolphy> grmbl
[16:30] <dolphy> found the issue
[16:44] <dolphy> BBB: when sending discont filesrc sets ->need_flush to FALSE
[16:44] <dolphy> BBB: so only discont is sent
[16:44] <dolphy> BBB: no flush
[16:54] <alley_cat> if i use GST_DEBUG in a sink, what do i need to add to the gst-launch commandline to get those debug messages?
[16:56] <dolphy> alley_cat: if you want to see only them you have to use GST_CAT_DEBUG iirc
[16:56] <dolphy> alley_cat: and then in gst-launch --gst-debug=GST_MYCAT:5
[17:04] iain ([email protected]) left irc: "leaving"
[17:04] <alley_cat> hmm, i don't get this
[17:04] iain ([email protected]) joined #gstreamer.
[17:07] <dolphy> alley_cat: you can use categories for debug messages
[17:07] <dolphy> alley_cat: so in your element you send a debug message and you stick it in a category
[17:07] <dolphy> alley_cat: existing one or a new one
[17:07] <dolphy> alley_cat: then when running your pipeline with gst-launch you can tell gst-launch to just show you the debug messages of a specific category
[17:08] <alley_cat> GST_CAT_DEBUG(NAS,"nassink: init");
[17:08] <dolphy> alley_cat: with --gst-debug=DEBUG_CATEGORY_NAME:DEBUG_LEVEL
[17:08] <alley_cat> that's what i use
[17:08] <dolphy> alley_cat: so then --gst-debug=NAS:5 should do it
[17:08] <alley_cat> ahhh, thanks
[17:10] <dolphy> yw
[17:13] Nick change: KA -> KA_afk
[17:15] markey ([email protected]) joined #gstreamer.
[17:17] markey ([email protected]) left irc: Client Quit
[17:26] Uraeus ([email protected]) joined #gstreamer.
[17:30] <BBB> bbl
[17:30] <Uraeus> bye :)
[17:30] BBB ([email protected]) left irc: "Client exiting"
[17:31] <Uraeus> dolphy, thomasvs: El Raval, is that downtown?
[17:32] <thomasvs> it's close to the ramblas
[17:32] <thomasvs> considered to be the dangerous part of town
[17:33] <Uraeus> thomasvs: well I am looking for a place for me and Wim to stay and the appartments in that area is the ones with a nice price tag
[17:35] <Uraeus> thomasvs: what about Barri Gotic?
[17:38] Action: alley_cat wishes nas was better documented
[17:39] <Uraeus> thomasvs: http://www.apartmentsramblas.com/barcelona_apartment_rent/index.asp?apart=3&entrada=06/02/2004&salida=12/02/2004&pers=2#
[17:39] <Uraeus> thomasvs: what do you think of that appartment?
[17:39] <iain> the ramblas is the dangerous part of town? weird, thats the only bit of barcelona we were in :)
[17:43] melay ([email protected]) joined #gstreamer.
[17:43] <melay> hi, is it possible to download real video streams (which realplayer can open)?
[17:45] <thomasvs> Uraeus: looks nice - what odes it cost ?
[17:46] <thomasvs> Uraeus: it looks like it's taken though
[17:46] <Uraeus> thomasvs: afaik so is it still free first half of februrary
[17:47] <Uraeus> thomasvs: 425 Euros for 6 days is the price I get quoted
[17:51] <thomasvs> is that acceptable ?
[17:52] <Uraeus> well there is no cheaper option unless I and Wim is to share a bed
[17:52] <Uraeus> I was thinking to ask fluendo for some subsidizing of the appartment rent
[17:55] iain ([email protected]) left irc: Read error: 60 (Operation timed out)
[17:58] hallibaby ([email protected]) joined #gstreamer.
[18:00] <thomasvs> Uraeus: why not just get a room at a pension with two single beds ?
[18:00] <Uraeus> thomasvs: well I checked one service and that was not cheaper, but if you have an URL I am all for it
[18:02] <Uraeus> hmm, found a place for 300 Euros, that might be ok
[18:02] <thomasvs> Uraeus: how about the phone number from the pension around the corner ?
[18:03] <Uraeus> habla espanol?
[18:03] <thomasvs> the owner speaks english as well :)
[18:03] <thomasvs> I'll go and ask tonight
[18:03] <Uraeus> thomasvs: ok thanks, remember to ask for prices too then
[18:04] <Uraeus> hmm, guess I should mail Wim and ask, for all we know he might have booked a hotelroom for himself already
[18:05] <Uraeus> ugh, this early feb thing made the whole ordeal more stressfull than planned :)
[18:05] <alley_cat> hmm, got the nassink working with audioscale, but it's ugly
[18:06] melay ([email protected]) left #gstreamer ("Leaving").
[18:06] <Uraeus> alley_cat: due to how the plugin works or how nas works or how gstreamer works?
[18:06] <alley_cat> how nas works, or how i understand nas works, not sure yet
[18:07] <Uraeus> thomasvs: that 300 Euro place I found on the net is next to the paral-lel subway station
[18:07] <alley_cat> i have to close and reopen the connection to the server when i want to change samplerate after initally setting it, not sure if that's the way it works or if i don't understand nas correctly yet
[18:08] <thomasvs> Uraeus: that is close to work
[18:09] <Uraeus> the people who has stayed at that hostel seems to say it is very nice
[18:10] <Uraeus> thomasvs: http://www.barcelona-on-line.es/eng/reserves/plantilla_hotel.asp?local=1&Moneda=6&accio=p31-hotelbooking.asp
[18:12] <thomasvs> yeah, looks ok
[18:12] <Uraeus> thomasvs: well I hold of booking anything until you have checked that place close to you
[18:13] <Uraeus> and I also mail Wim to see what he thinks
[18:22] Action: thomasvs fixes BBB's nonvalidating docbook
[18:23] <teuf> thomasvs: you'll soon be an xml/xslt guru and give up hacking on gstreamer to work on libxml instead ;)
[18:23] <thomasvs> teuf: no, xslt is just for fun :)
[18:24] <thomasvs> teuf: i'll never leave gst anymore
[18:24] <thomasvs> at least, not while I have this job :)
[18:24] <teuf> thomasvs: dolphy isn't looking, you can tell us you are secretly hacking on libxml at work ;)
[18:24] Action: dolphy hits his head on walls
[18:24] <thomasvs> well, he is looking
[18:24] <thomasvs> :)
[18:25] <thomasvs> and he's not hitting his head on any wall around here
[18:25] <teuf> nah, he is busy with walls
[18:25] <dolphy> avidemux is completely even unaware
[18:25] <dolphy> it simply read like a stupid thing from riff lib
[18:25] <dolphy> so as soon as upstream element sends an event it makes an error
[18:25] <dolphy> cannot read from ressource
[18:25] <dolphy> how cute
[18:26] Action: dolphy cross fingers so that ronald comes back fast
[18:26] yippi ([email protected]) joined #gstreamer.
[18:26] ct_ ([email protected]) joined #gstreamer.
[18:26] <Uraeus> hello yippi and ct_
[18:26] <ct_> Uraeus: hey
[18:26] <Uraeus> ct_: so lymric is getting back on track?
[18:26] <ct_> yep
[18:26] <ct_> it's coming along pretty well
[18:27] <ct_> i've a couple bugs to fix, after i rewrote our player
[18:27] <ct_> it broke somethings in the app, so, i'll get build fixes finished pretty soon and i'll work on checking them into sf cvs
[18:27] <Uraeus> dolphy: looking at the changelog files it seems some seeking issues has been fixed, does that mean seeking works again now?
[18:28] Nick change: harshy -> harshyWork
[18:30] <dolphy> Uraeus: i m working on that
[18:31] <dolphy> Uraeus: it s not finished though
[18:31] <Uraeus> ah ok
[18:32] <taaz> thomasvs: pong
[18:32] <dolphy> Uraeus: but when it will be finished you will have a seeking so performant you won't believe it :)
[18:32] mathrick ([email protected]) joined #gstreamer.
[18:32] <mathrick> yo
[18:32] <Uraeus> dolphy: :)
[18:32] <Uraeus> hey mathrick
[18:32] <Uraeus> mathrick: started to work on your mms lib yet?
[18:34] <mathrick> Uraeus: kinda, there's libmms name overtake request pending on sf ;)
[18:34] Nick change: bluejay|sleep -> bluejay
[18:34] <mathrick> yo bluejay
[18:34] <Uraeus> bluejay: I am assuming you are using the blueJ editor to hack? ;)
[18:35] <bluejay> hello. No. I hate that editor
[18:35] <bluejay> I had the name first. ;)
[18:35] Action: mathrick is amazed that no MyDOOM hit him yet ;)
[18:35] <teuf> mathrick: do you want some ?
[18:36] <mathrick> teuf: no, but it's just unusual :). When swen broke out, I had instantly 300+ mails in my inbox
[18:36] <teuf> mathrick: maybe your mail server is filtering them ?
[18:37] <mathrick> teuf: mebbe
[18:37] Action: teuf goes home
[18:37] teuf ([email protected]) left irc: "Client exiting"
[18:39] <Uraeus> thomasvs: so how much is something we can use in that Intel programming handbook for GStreamer?
[18:42] <thomasvs> Uraeus: not much, I'm afraid
[18:42] <thomasvs> some ideas we could reuse though for our docs
[18:42] <thomasvs> but I'm going to ask him if he wants to help out on our manuals instead
[18:43] <Uraeus> thomasvs: is this the indian fellow who has been mailing the list? and did he say why he wrote it?
[18:45] <taaz> i wonder if a Word doc writer will be resistant to docbook ;)
[18:47] <thomasvs> Uraeus: nope, no idea why he wrote it :)
[18:48] <Uraeus> thomasvs: but basically he hasn't documented anything not documented already?
[18:49] <thomasvs> Uraeus: yeah
[18:49] <thomasvs> sad no ?
[18:50] <Uraeus> yeah, well it seems we are having a really good doc drive over the last two days
[18:53] mchal ([email protected]) joined #gstreamer.
[18:53] <mchal> hello
[18:53] <taaz> thomasvs: seen buildbot? http://buildbot.sourceforge.net/
[18:54] <mchal> does gst support mpc files?
[18:55] <Uraeus> mchal: I don't think we do, but if they are supported in ffmpeg we could relativly easily add support for it
[18:57] <mchal> hm
[18:57] <Uraeus> mchal: but just so I remember correctly, mpc is that 'mp3 variation' which is supposed to be higher quality right?
[18:57] <mchal> Uraeus, yes
[18:58] <Uraeus> not sure if ffmpeg support it, not sure there is a free software decoder at all for it actually? do you know any linux software which currently support it?
[18:58] <mchal> Uraeus, xmms support it
[18:59] <Uraeus> mchal: hmm, can you play those files using the madplay command line tool?
[19:00] somex1 ([email protected]) joined #gstreamer.
[19:00] <mchal> Uraeus check http://sourceforge.net/projects/mpegplus/
[19:02] <Uraeus> ok, that project use the LGPL which is a good start for getting us to use it :)
[19:02] <yippi> hello uraeus.  :)
[19:03] <Uraeus> mchal: any chance you would be interested in making a plugin out of that yourself? the current developers have their hands full stabilizing what we already have in preparation for GNOME2.6
[19:03] <mchal> Uraeus, sorry i'm not a coder
[19:04] <Uraeus> mchal: if you add a bugzilla for it with this link I try to see if someone is interested in adding support for it
[19:04] <mchal> Uraeus sure, thanks for advice
[19:07] <Uraeus> mchal: we got monkeys audio support based on a user request so we do try adding stuff people ask for :)
[19:08] <mchal> :)
[19:08] <thomasvs> taaz: rock, looks nicer than tinderbox !
[19:08] <thomasvs> taaz: have you tried it ?
[19:10] <taaz> thomasvs: nope ;)  just remembered you were looking for an auto build system
[19:10] <mchal> MPC musepack is on wishlist :)
[19:11] <taaz> twisted would allow evil things like spamming #gstreamer when builds fail
[19:11] <thomasvs> taaz: looks really sweet
[19:11] <thomasvs> taaz: will definately help
[19:11] <thomasvs> taaz: yeah, was just envisioning the same thing
[19:11] <thomasvs> it would say "Company - fifth time you broke it this week.  no cvs commit rights for two days"
[19:11] <thomasvs> yummy :)
[19:11] <taaz> heh
[19:12] Action: thomasvs is going to test it asap
[19:12] <thomasvs> well, after fixing the modules set of pages to be built from only one xml file
[19:14] Nick change: Jaramir -> Jara[zZ]
[19:15] markey ([email protected]) joined #gstreamer.
[19:18] Rotty ([email protected]) joined #gstreamer.
[19:19] <mchal> bye
[19:19] mchal ([email protected]) left irc: "Leaving"
[19:22] Nick change: bluejay -> bluejay|gone
[19:24] <thomasvs>   Running cvs diff to find changes.
[19:24] <thomasvs> cvs: rcs.c:3028: RCS_getrevtime: Assertion `rcs != ((void *)0)' failed.
[19:24] <thomasvs> cvs [diff aborted]: received abort signal
[19:25] ct_ ([email protected]) left irc: "Client exiting"
[19:31] ChrisHJW ([email protected]) joined #gstreamer.
[19:31] Manny ([email protected]) joined #gstreamer.
[19:31] <Manny> hi
[19:32] <mathrick> hi Manny, ChrisHJW
[19:32] <Manny> I've got a low-quality MP3 (mono, 96kbps, 41000 khz). I'd like to convert this file into wav that is burnable on CD
[19:32] <Manny> it has got a length of 50 minutes
[19:33] <Manny> when converting it to wav using mpg123, it results in a file having ~300 MB which isn't enough obviously for a 50 minute wav
[19:33] <Manny> when burning this on CD using cdrdao, it sounds horrible
[19:33] <Manny> it doesn't even seem to be a valid audio signal
[19:34] <Manny> now my gstreamer-related question: Is there a pipe that has proven to convert low-quality mp3s to valid wav files
[19:34] <Manny> ?
[19:34] <thomasvs> record.c: line 459 (gst_rec_link_pre): assertion `(rec->audio_encoder != NULL || rec->audio_filter != NULL) && !(rec->audio_encoder != NULL && rec->audio_filter != NULL)' failed
[19:34] Action: thomasvs kicks BBB for indecipherable asserts
[19:36] <Misirlou> Manny: It sounds horrible because it's a monophonic MP3 at 96 kb/s.
[19:36] Action: thomasvs gives up trying to screenshot the recorder
[19:36] <mathrick> Manny: gst-launch-0.7 filesrc location="filename.mp3" ! typefind ! spider ! audio/x-wav ! filesink location="filename.wav" <-- you can try this, although I don't know if it would help much with lo-fi sound
[19:37] <mathrick> Manny: does it sound horrible when being played from mp3?
[19:37] <Manny> no, the wav itself is bogus
[19:37] <Manny> and cdrdao obviously can't handle it
[19:37] <Manny> no music at all
[19:37] <Manny> just noise
[19:38] <mathrick> Manny: hmm, so mp3 is ok, but produces faulty wav?
[19:39] <Manny> I don't know whether the MP3 is "ok"
[19:39] <mathrick> Manny: can you listen to it?
[19:39] <Manny> yes
[19:39] <Manny> using mpg123/321
[19:39] <mathrick> Manny: w/o noise you hear when playing wav?
[19:39] <Manny> and mp3blaster processes it, too
[19:39] <Manny> gst-launch filesrc location="in.mp3" ! typefind ! spider ! audio/x-wav ! filesink location="out.wav" <-- that's my pipeline (0.6)
[19:40] <Manny> error: syntax error, unexpected $undefined
[19:40] <Manny> ERROR: pipeline could not be constructed: Invalid syntax
[19:40] <mathrick> Manny: wouldn't work
[19:40] <mathrick> Manny: there's no filtered caps syntax in 0.6
[19:40] <Manny> :/
[19:40] <Manny> so I'll go installing CVS HEAD first :)
[19:40] <Manny> mathrick: what modules do I need?
[19:40] <mathrick> you can try replacing audio/x-wav with wavenc, but I suggest CVS HEAD :)
[19:41] <mathrick> Manny: well, gstreamer and gst-plugins obviously :)
[19:41] <Manny> cvs server: cannot find module `gst-plugins' - ignored
[19:41] <Manny> :pserver:[email protected]:2401/cvs
[19:41] <mathrick> Manny: huh?
[19:42] <mathrick> :pserver:[email protected]:/cvs/gstreamer
[19:42] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[19:42] <Manny> ok
[19:42] <mathrick> Manny: then you'd need to make sure you have enough builddeps for gst-plugins
[19:42] <Manny> I hope so
[19:42] <Manny> we'll see
[19:43] <mathrick> Manny: I suggest running ./configure | tee plugins.log, to spare you some of the joy of waiting for it multpile times to finish ;)
[19:43] <Manny> mathrick: 0.7/0.6 can be installed in parallel, right?
[19:43] <Manny> parallell
[19:43] <mathrick> Manny: sure
[19:43] <Manny> uhm no the first one was correct
[19:44] <mathrick> Manny: then, at the end of plugins.log, you'll have list of plugins to be built, and those which aren't going to
[19:45] <Manny> tee is a nice tool
[19:45] <Manny> never used it
[19:45] <Manny> thanks for your help
[19:46] <mathrick> Manny: np
[19:55] <Manny> what is poisioning?
[20:02] <Misirlou> ?
[20:03] <mathrick> Manny: it's for maintainer builds, invalidates memory before freeing it
[20:03] <Manny> oh
[20:03] <mathrick> Manny: after ./autogen.sh, run ./configure with no options manually
[20:03] <Manny> no, I passed --prefix and --sysconfdir
[20:04] <Manny> I know, it complained when not running configure after autogen
[20:04] <mathrick> Manny: these are ok
[20:04] <Manny> when trying to install it
[20:04] <Manny> I know
[20:05] pb_ ([email protected]) left irc: "Client exiting"
[20:08] <thomasvs> http://www.freedesktop.org/~gstreamer/modules/
[20:08] <thomasvs> whee !
[20:08] <thomasvs> can somebody click some links there to verify ?
[20:09] Manny ([email protected]) left irc: Remote closed the connection
[20:11] <thomasvs> somebody better cheer me on in the next three minutes or I give up working on gst altogether
[20:12] <Uraeus> thomasvs: yay go thomasvs!!
[20:12] <mathrick> thomasvs: seems to work fine :)
[20:12] <mathrick> thomasvs: you rock, doing all that auto* voodoo I can't even understand :)
[20:13] somex1 ([email protected]) left #gstreamer.
[20:14] <mathrick> thomasvs: confirmed, every link I happened to click works, really w00t :)
[20:15] <thomasvs> cool
[20:15] <thomasvs> I think I'm getting the hang of it now
[20:15] <thomasvs> Uraeus: so, about those tasks, when are you going to fix them ?
[20:16] <Uraeus> thomasvs: I am on it I am on it, stopp nagging mom
[20:16] Action: sxpert is back (gone 10:54:25)
[20:19] <thomasvs> thanks 
[20:20] <thomasvs> mathrick: thanks for checking
[20:20] <thomasvs> now I need to make new screenshots
[20:25] <mathrick> thomasvs: on gst-editor page, second screenshot is repeated in large version at the bottom
[20:25] <thomasvs> mathrick: yeah, I need to make a new screenshot
[20:25] <thomasvs> mathrick: I just recycled the second one for now
[20:25] <mathrick> thomasvs: k
[20:28] <thomasvs> ok, going home
[20:29] markey ([email protected]) left irc: "brb"
[20:34] dolphy ([email protected]) joined #gstreamer.
[20:35] <dolphy> re
[20:35] <mathrick> wb dolphy
[20:37] markey ([email protected]) joined #gstreamer.
[20:37] Action: dolphy gets back to that seeking thing
[20:37] thomasvs ([email protected]) left irc: Read error: 60 (Operation timed out)
[20:38] yippi ([email protected]) left irc: "Client exiting"
[20:45] sublett ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[20:45] C ([email protected]) left irc: Remote closed the connection
[20:54] sublett ([email protected]) joined #gstreamer.
[21:03] Nick change: KA_afk -> KA
[21:08] yippi ([email protected]) joined #gstreamer.
[21:10] thomasvs ([email protected]) joined #gstreamer.
[21:17] <dolphy> thomasvs: hmm i found why the discont event manage to pass
[21:17] <dolphy> thomasvs: that's really bad :)
[21:23] pb_ ([email protected]) joined #gstreamer.
[21:23] Company ([email protected]) joined #gstreamer.
[21:24] <Company> good evening angels
[21:24] <dolphy> hi
[21:24] <dolphy> do you have some bytestream knowledge ?
[21:24] <Company> yeah
[21:25] <dolphy> can you tell me why bytestream function are so poor at event handling ?
[21:25] <Company> because bytestreams are older then events :)
[21:25] <dolphy> ah
[21:25] <Company> someone should have redesigned the API long ago...
[21:26] <dolphy> i don't see why event is interrupting bytestream reading
[21:26] <dolphy> i m trying to fix avidemux
[21:26] <dolphy> because if you send an push an event to avidemux it stops and say error reading from ressource
[21:34] Zeenix ([email protected]) joined #gstreamer.
[21:34] <Zeenix> hi
[21:36] <Uraeus> hi Zeenix
[21:36] <Uraeus> thomasvs: seems the page build doesn't work as it seems the build system doesn't see that I have updated the files
[21:38] <Uraeus> with current CVS of plugins and core do I get this when running gst-register-0.7:
[21:38] <Uraeus> added plugin flac with 3 feature(s)
[21:38] <Uraeus>  
[21:38] <Uraeus> (process:23461): GStreamer-CRITICAL **: Could not convert static caps "video/x-raw-yuv, form at = (fourcc) YUY2, width = (int) 720, height = (int) { 480, 576 }, framerate = (float) { 25 .0, 29.997 }; video/x-raw-rgb, bpp = (int) 32, depth = (int) 32, endianness = (int) 4321, re d_mask =   (int) 0x000000ff, green_mask = (int) 0x0000ff00, blue_mask =  (int) 0x00ff0000, w idth = (int) 720, height = (int) { 480, 576 }, framerate = (float) { 25.0, 29
[21:38] <Uraeus> .997 }; video/x -raw-rgb, bpp = (int) 24, depth = (int) 24, endianness = (int) 4321, red_mask =   (int) 0x00 0000ff, green_mask = (int) 0x0000ff00, blue_mask =  (int) 0x00ff0000, width = (int) 720, hei ght = (int) { 480, 576 }, framerate = (float) { 25.0, 29.997 }"
[21:38] <Uraeus> added plugin dvdec with 1 feature(s)
[21:39] <thomasvs> Uraeus: which files did you update ?
[21:39] <thomasvs> Uraeus: and what did it say after commiting ? can you paste to me privately ?
[21:43] <thomasvs> Uraeus: you didn't run make locally to test ?
[21:43] <thomasvs> Uraeus: because your changes do not validate at all
[21:44] <Uraeus> thomasvs: locally it complains about no rule to make index.html
[21:45] <thomasvs> Uraeus: well, NEVER commit unless it builds locally
[21:45] <thomasvs> sigh, and you pasted some content without spaces between lines and one big line of 1000 chars
[21:47] <Uraeus> thomasvs: well I figured I proof read it online
[21:53] <Uraeus> thomasvs: commit the index.html fix and I clean up my stuff locally and recommit
[21:54] <thomasvs> Uraeus: nope, NEVER do that !
[21:54] <thomasvs> Uraeus: as the README says - you should check locally
[21:54] <thomasvs> Uraeus: never EVER "check online"
[21:56] <mathrick> hmm, are elements with possibly multiple {sink,source} pads considered filters?
[21:56] <mathrick> or are filters strictly of one in - one out type?
[21:57] <mathrick> ie. is demuxer a filter?
[21:59] <thomasvs> Uraeus: ok, fixed, commited and uploaded now
[22:04] <dolphy> thomasvs: it's quite hard to get that even in rifflib fixed
[22:04] <dolphy> s/even/event
[22:10] <dolphy> that bytestream thing is a big issue
[22:10] Nick change: harshyWork -> harshy
[22:11] thaytan ([email protected]) left irc: Read error: 113 (No route to host)
[22:18] ChrisHJW_log ([email protected]) joined #gstreamer.
[22:19] hallibaby ([email protected]) left irc: "Bye bye"
[22:19] <mathrick> ds-work: so filter is strictly one input, one output, the same caps elem? Seems rather restricted
[22:22] <mathrick> ds, ds-work, ds-gromit: which one of you is real one?
[22:26] yippi ([email protected]) left irc: "Client exiting"
[22:27] <Company> well, audiofilter and videofilter are supposed to work on one format only
[22:28] <Company> so iain could basically write all his marlin effects with audiofilter only...
[22:30] <Company> ha, not to self:read more backlog
[22:30] <Company> mathrick: currently everything that has at least one input and one output pad is considered a filter
[22:31] <Company> mathrick: so yes, demuxers are filters
[22:31] <mathrick> Company: http://bugzilla.gnome.org/show_bug.cgi?id=125890 <-- so it's correct?
[22:31] <mathrick> Company: ds commited entirely different opinion just at the same time, we even managed to get conflict ;)
[22:32] <Company> mathrick: _currently_ it's done as you say, but I share ds' view
[22:32] <Company> mathrick: i'd prefer if we made decoders, demuxers and friends not filters
[22:33] <mathrick> Company: does being a filter involve some specific subclassing?
[22:33] <Company> mathrick: nope
[22:33] <Company> mathrick: the classes are important for two things mainly:
[22:33] <Company> mathrick: 1) finding an element in gst-editor
[22:33] <Company> mathrick: 2) restricting the elements an autoplugger uses
[22:33] <mathrick> Company: I think that any one in - one out elem should be a filter
[22:34] <Company> mathrick: why would that be good?
[22:34] <mathrick> Company: so 'we made' == 'we started calling them' ?
[22:34] <mathrick> Company: well, it seems natural :)
[22:35] <Company> mathrick: yeah, i'd like to get a bit more "root" categories - not just src, sink and filter
[22:35] <mathrick> Company: and filter is such a beautiful, lockpick term ;)
[22:35] <mathrick> Company: yep, it would be nice to have some generic terms at hand
[22:36] <Company> dictionary.com says a filter is "Any of various electric, electronic, acoustic, or optical devices used to reject signals, vibrations, or radiations of certain frequencies while allowing others to pass"
[22:37] <Company> which doesn't match close to nothing we call a filter ;)
[22:37] <mathrick> Company: kind of ;)
[22:38] <Company> anyway, the only important part for me is classifying what should be autoplugged
[22:38] <Company> so you'd say autoplugger elements=/converter/transcoder:/converter/decoder for a decoding autoplugger
[22:39] <Company> autoplugger elements=/filter/converter/transcoder:/filter/converter/decoder buys you nothing there...
[22:39] <ds-work> filter, imo, should have one source and sink pad, and have the same caps on both pads
[22:40] <mathrick> Company: yeah, I agree, but it's rather about terminology used in docs which matters here
[22:40] <ds-work> and input/output buffers at the same rate
[22:40] <Company> ... with the same timestamps
[22:40] <mathrick> ds-work: very restrictive definition
[22:41] <Company> but it fits a load of plugins
[22:41] <Company> elements
[22:41] <ds-work> mathrick: it's a definition that makes sense
[22:41] <ds-work> because that's what people think of for filters
[22:41] <mathrick> ok, so contest: how do we call element that has both input and output?
[22:41] <ds-work> an element
[22:42] <mathrick> ds-work: no, then you will match sources and sinks too
[22:42] <ds-work> so
[22:42] <Company> ds-work: what would you name goom, festival and friends?
[22:42] <mathrick> ds-work: it's kinda bad when describing various elem types in PWG ;)
[22:42] <ds-work> Company: converters
[22:43] <ds-work> Company: specifically, synthesizers
[22:43] <mathrick> ds-work: and are filters converters too?
[22:43] <ds-work> no, because they don't convert
[22:43] <mathrick> ds-work: then, we need some generic term
[22:43] <Company> mathrick: why?
[22:43] <ds-work> no, we don't
[22:43] <Company> mathrick: btw, bin, pipeline and thread are "container" elements
[22:44] <mathrick> Company: for easier explanation in PWG, for easier referring
[22:44] <Company> mathrick: "data passing elements"
[22:44] <mathrick> Company: ah, ok, I hoped someone will revise that :)
[22:45] <mathrick> Company: but what do you think of it as general explanation?
[22:45] <ds-work> "inline elements"
[22:45] <mathrick> ds-work: nice, i like it
[22:45] <ds-work> but don't use it
[22:45] <mathrick> ds-work: although, it will be a hell when translating that to Polish ;)
[22:45] <ds-work> because an "inline element" isn't a special case of an element
[22:45] <mathrick> ds-work: huh?
[22:46] <ds-work> sources and sinks are special cases of element
[22:46] <ds-work> because they only have one direction of pad
[22:46] <sxpert> what's happening there ?? : /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.2/../../../libgstplay-0.7.so: undefined reference to `gst_x_overlay_get_type'
[22:46] <Company> well there's 4 categories with and witrhout source or sink
[22:46] <Uraeus> sxpert: old core version?
[22:47] <ds-work> cdplayer is uberspecial
[22:47] <Company> as is v4lelement
[22:47] <mathrick> Company: I think you can exclude one, that without sink and source altogether ;)
[22:47] <sxpert> Uraeus: just compiled and installed it
[22:47] <Company> sxpert: does libgstplay link against libgstinterfaces?
[22:47] <mathrick> sxpert: ldconfig ?
[22:48] <Company> mathrick: "containers" + v4lelement
[22:48] <sxpert> Company: nope
[22:48] <mathrick> Company: hmm, ok, but why v4l ?
[22:48] <Company> mathrick: 4vlelement is an element that reads from v4l and uses overlay capabilities to display directly in X
[22:48] <Company> v4lelement, too
[22:49] <Company> sxpert: then bug dolphy about it :)
[22:49] <sxpert> one thing is that my X doesn't have Xv support
[22:49] <mathrick> Company: ahh, funky :)
[22:49] <mathrick> ds-work: why is cdplayer special?
[22:49] <Company> sxpert: that doesn't matter
[22:49] <ds-work> mathrick: same reason
[22:49] <sxpert> Company: ok
[22:49] <sxpert> will do
[22:49] <mathrick> ds-work: it's not source?
[22:50] <ds-work> no
[22:50] <Company> mathrick: it plays audio cds directly to the sound card
[22:50] <ds-work> I don't even know if it exists anymore
[22:50] <Company> mathrick: via this funky direct onnection cable that's always in the way
[22:50] <Company> when you switch parts inside your pc
[22:51] <mathrick> ahh, so it's not really an element, it's pipeline itself :)
[22:51] <Company> something like that
[22:51] <Company> it's one of those "entirely in hardware" elements
[22:51] <ds-work> more like a bastardization
[22:51] <mathrick> Company: yeah, last time i changed CD drive I didn't even connect it :)
[22:52] Action: mathrick doesn't like whole CD playing funkyness
[22:53] Action: Company would prefer to run v4lsrc ! application/direct-graphics-rendering ! ximagesink instead of having v4lsrc, too
[22:53] <Company> but that's not how things work unfortunately
[22:55] <mathrick> Company: but there is v4lsrc, but no cdsrc apparently
[22:56] <Company> cdsrc is called cdparanoia
[22:56] <Company> and there was a cddasrc or cdda2wavsrc once, too
[22:57] <mathrick> I think cdparanoia should be renamed to cd{da}src 
[22:58] <Company> there should be a native cdsrc
[22:58] <Company> cdparanoia is ugly...
[22:58] <ds-work> I'm planning on rewriting libcdparanoia sometime soon
[22:58] <Company> before or after the autoplugger? ;)
[22:58] <ds-work> after
[22:58] <ds-work> much after
[22:59] Action: mathrick wonders what've happend to his cdparanoia elem
[22:59] <ds-work> 'sometime soon' == 6 month timeframe
[22:59] <Company> i had imagined the sound-juicer guys write a good cd src
[22:59] <mathrick> I remember having it once
[22:59] <Company> apparently cdparanoia wasn't bad enough ;)
[23:00] Action: Company wonders what sound-juicer will do once vorbisenc doesn't do 16bit int anymore
[23:01] <Company> is cd output little endian or big endian?
[23:03] <ds-work> depends on the drive
[23:03] <ds-work> you can set it on most drives, iirc
[23:04] <ds-work> er, wait for someone to fix int2float
[23:05] <Uraeus> Company: isn't the cdplayer plugin a native cdsrc?
[23:06] <mathrick> Uraeus: read above ;)
[23:06] <Uraeus> ok :)
[23:06] Nick change: KA -> KA_afk
[23:16] <dolphy> weeeeeeeee
[23:16] Action: dolphy fixed seeking :)
[23:16] <dolphy> immediate seeking even :)
[23:17] <thomasvs> dolphy: commit !!!
[23:17] <dolphy> thomasvs: well :)
[23:17] <Uraeus> yay :)
[23:17] <dolphy> thomasvs: better check that i did not break everything
[23:17] <dolphy> i had to make some changes in bytestream
[23:25] <Uraeus> dolphy: does this fix mean that gst-player bascially works again?
[23:26] <dolphy> Uraeus: hmm not completely i m afraid
[23:26] <dolphy> Uraeus: there is still a time issue that Company is investigating
[23:27] <Uraeus> ok, well being done to one hurdle is also a good thing :)
[23:28] yippi ([email protected]) joined #gstreamer.
[23:31] <Uraeus> wb yippi
[23:31] <dolphy> waaaoow
[23:31] <dolphy> totem seeks magnificently on avis :)
[23:31] <dolphy> instant seeking :)
[23:31] <dolphy> that's cute :)
[23:32] <dolphy> mpeg seems buggy though...
[23:32] <Company> dolphy: is it in ync?
[23:33] <Company> sync?
[23:33] <dolphy> it seems
[23:33] <Company> that's cool
[23:33] <Company> but the "next" issue is still there?
[23:34] <dolphy> Company: haven't checked yet
[23:34] <dolphy> Company: maybe my fix will help because of DISCONT 
[23:34] <dolphy> Company: which was eaten by avidemux till now
[23:35] <thomasvs> dolphy: I WANT TO TEST !
[23:36] <dolphy> thomasvs: can you please test mpeg seeking with your current version and stop shouting ? :)
[23:36] Action: thomasvs is just trying to commit his gst_element_error -> GST_ELEMENT_ERROR
[23:36] <thomasvs> after that, time to check the libtool versioning
[23:37] <thomasvs> ds-work: have you put in that patch you wanted in ?
[23:37] Action: thomasvs is liking buildbot
[23:39] <taaz> dolphy: mpeg2dec plugin still doesn't use the new mpeg2_reset(...) call.  that might be needed to make seeking work better
[23:39] <ds-work> thomasvs: yes
[23:39] <taaz> thomasvs: did you get a chance to look at that pkgconfig patch?
[23:39] <yippi> heya!
[23:40] <thomasvs> taaz: yeah, which was why I poked you, but after spending another day on the site I completely forgot my remakers
[23:40] <thomasvs> taaz: so send me the patch by mail so I can reply in mail and not forget everything
[23:42] <dolphy> committed
[23:43] <dolphy> thomasvs: TEST !
[23:44] Action: thomasvs curses his dog slow telefonica link
[23:45] <dolphy> Company: the next issue is still there
[23:47] <taaz> thomasvs: email no good today.  i've only gotten a few emails since monday.  virus problems took out entire univ mail system: http://www.vt.edu/news/showitem.php?id=1075410215
[23:47] Action: ds-work wants to see pkg-config --cflags gstreamer-0.8
[23:48] <taaz> dolphy: does seeking work now with mpeg2dec?
[23:48] <dolphy> taaz: well yeah it works
[23:48] <dolphy> taaz: dunno what you meant with mpeg2_reset though
[23:48] <taaz> hmm... back when i was working on vdv i had bad seeking problems with the mpeg2dec/a52dec combo
[23:48] <thomasvs> ds-work: I'm going to do that after the next release
[23:49] <taaz> the lib never really handled seeking properly.  i think the current plugin still has some hacks for that.  using _reset() is a more proper way to ensure decoder state is reset when you seek.  at least that's how i think it's supposed to work ;)
[23:50] <ds-work> thomasvs: after?
[23:51] <sxpert> dolphy : /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.2/../../../libgstplay-0.7.so: undefined reference to `gst_x_overlay_get_type'
[23:51] <thomasvs> ds-work: not really much point right now, is there ?
[23:51] <dolphy> sxpert: your built is fucked up :)
[23:51] <mathrick> sxpert: did you run ldconfig?
[23:52] <sxpert> mathrick: yeah
[23:52] <sxpert> dolphy: erh...
[23:52] <dolphy> sxpert: which app ? examples/gstplay ? gst-player ? totem ?
[23:52] <ds-work> thomasvs: 0.7.4 will be the first release with a frozen ABI
[23:52] <sxpert> nah, my lil player
[23:53] <thomasvs> ds-work: sure, but no point to break all apps' configure yet before we tell them we're about to
[23:53] <ds-work> thomasvs: ah, right, of course
[23:55] markey ([email protected]) left irc: "nite"
[23:56] <dolphy> sxpert: so fix your lil player then :)
[23:56] <dolphy> sxpert: because all other apps compile and run just fine for me :)
[23:56] <dolphy> thomasvs: are you testing seeking ?
[23:56] <ds-work> does anyone know of a graph-based framework for image or data manipulation, somewhat similar to gstreamer, but more geared for still images?
[23:57] <mathrick> ds-work: someone doing something similar was here not so long ago
[23:57] <mathrick> ds-work: also GEGL is something like that
[23:57] <thomasvs> ds-work: there's some gimp-related lib for that, yeah
[23:57] <thomasvs> ds-work: btw, I'm not sure esd should be "fixed" to only support those formats
[23:58] <thomasvs> ds-work: a) esd can be fixed and b) apps can get the same behaviour as your fix by doing a filtered connect
[23:58] <thomasvs> ds-work: would you mind if I reverted those bits ?
[23:58] taaz ([email protected]) got netsplit.
[23:58] Misirlou ([email protected]) got netsplit.
[23:58] sxpert ([email protected]) got netsplit.
[23:59] <ds-work> thomasvs: gstreamer does better audio resampling than esd
[23:59] taaz ([email protected]) returned to #gstreamer.
[23:59] Misirlou ([email protected]) returned to #gstreamer.
[23:59] sxpert ([email protected]) returned to #gstreamer.
[23:59] <ds-work> thomasvs: esdsink theoretically _should_ work with different formats, but it doesn't work in practice
[23:59] Action: sxpert gets a fresh cvs
[23:59] <thomasvs> ds-work: sure - now.  but in general, it seems like the wrong solution.  esp. since I am thinking of rewriting esd using gst
[00:00] --- Fri Jan 30 2004
[00:00] <ds-work> thomasvs: don't rewrite esd
[00:00] <ds-work> thomasvs: the esd protocol is absolute crap
[00:01] <thomasvs> ds-work: yes - but we're stuck with it in gnome.  which is why I a) want to rewrite it using gst then b) extend it
[00:01] <ds-work> thomasvs: and a sound server doesn't really do anything complicated enough to need gstreamer
[00:01] <thomasvs> ds-work: nope, but the very first thing it would bring that i'd like is runtime audio backend
[00:01] <dolphy> k i m so tired
[00:01] Action: dolphy is off to bed
[00:01] <dolphy> night all
[00:01] <thomasvs> ds-work: I spent an hour trying to figure out why my jamboree didn't work ...
[00:01] <mathrick> nite dolphy
[00:01] <thomasvs> ds-work: ... when it was because it chose to compile against alsa just because I had the libs installed
[00:02] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[00:02] Nick change: Jara[zZ] -> Jaramir
[00:02] <thomasvs> ds-work: as a bonus, it would force the sun people to actually make sure the solaris plugin works :)
[00:02] <ds-work> heh
[00:02] <taaz> ah... found out why gst-python not likeing these enum values.  somewhere need to run glib-mkenums to autogen some code.  yuck.
[00:03] <taaz> like how core does gst/gstenumtypes.{ch}
[00:04] <thomasvs> I had this cool idea for speeding up autogen.sh, but threw it away after spending an hour on it
[00:04] <thomasvs> someone want to script it for me ?
[00:09] <thomasvs> (process:20105): GStreamer-CRITICAL **: Could not convert static caps "video/x-raw-yuv, format = (fourcc) YUY2, width = (int) 720, height = (int) { 480, 576 }, framerate = (float) { 25.0, 29.997 }; video/x-raw-rgb, bpp = (int) 32, depth = (int) 32, endianness = (int) 4321, red_mask =   (int) 0x000000ff, green_mask = (int) 0x0000ff00, blue_mask =  (int) 0x00ff0000, width = (int) 720, height = (int) { 480, 576 }, framerate = (float
[00:09] <thomasvs> ) { 25.0, 29.997 }; video/x-raw-rgb, bpp = (int) 24, depth = (int) 24, endianness = (int) 4321, red_mask =   (int) 0x000000ff, green_mask = (int) 0x0000ff00, blue_mask =  (int) 0x00ff0000, width = (int) 720, height = (int) { 480, 576 }, framerate = (float) { 25.0, 29.997 }"
[00:09] <thomasvs> any idea what this is on running gst-register ?
[00:10] <ds-work> (float)
[00:10] Action: ds-work apparently forgot to commit that fix
[00:10] <ds-work> ext/dv
[00:10] <Uraeus> ok night everyone
[00:10] Uraeus ([email protected]) left irc: "Client exiting"
[00:12] <ds-work> thomasvs: update
[00:15] <thomasvs> hm
[00:15] <thomasvs> the seeking is indeed very fast when it's not causing segfaults now
[00:15] <thomasvs> but there still seem to be some problems
[00:15] <thomasvs> esp. with resyncing video output after a seek
[00:19] ChrisHJW ([email protected]) left irc: Connection timed out
[00:21] Action: thomasvs thinks bytestream leaks heavily
[00:21] <thomasvs> when I move the slider on matrix.avi, gst-player's cpu usage goes up 6% :)
[00:22] <thomasvs> so I can seek about 15 times before it gets killed :)
[00:22] <ds-work> cpu usage?
[00:22] thaytoo ([email protected]) joined #gstreamer.
[00:22] <ds-work> or mem usage?
[00:22] <thomasvs> mem usage, sorry
[00:22] <ds-work> gstreamer developers should start using ulimit on their local machines
[00:23] <ds-work> although that would cause problems when running galeon :)
[00:24] <thomasvs> for some reason it cannot negotiate my mpeg files with oss anymore
[00:24] <thaytoo> morning, guys
[00:24] <ds-work> buffer-frames, perhaps?
[00:24] <ds-work> someone removed that from audio.h
[00:25] <thaytoo> Company, I think?
[00:26] <thomasvs> that was only for float no ?
[00:26] <thomasvs> ok, so
[00:26] <thomasvs> app. my oss doesn't accept mono
[00:26] <ds-work> it was removed from int
[00:26] <thomasvs> gst-launch sinesrc ! audioconvert ! osssink doesn't scale it properly though
[00:27] <thomasvs> you think it would affect all nego ?
[00:27] <ds-work> it shouldn't
[00:27] <Zeenix> thomasvs: you are in spain right?
[00:27] <thomasvs> Zeenix: yep
[00:27] <thomasvs> ds-work: well, Company would've at least tested some audio pipelines before committing no ?
[00:27] <Zeenix> thomasvs: still being paid to work for gst?
[00:27] <ds-work> thomasvs: um, sure
[00:28] <thomasvs> this works fine
[00:28] <thomasvs>  gst-launch sinesrc ! audioconvert ! audio/x-raw-int,channels=2 ! osssink
[00:28] <thomasvs> without hte filtered caps, it doesn't
[00:28] <thomasvs> so audioconvert is broken again :)
[00:28] <thomasvs> Zeenix: yep
[00:28] <ds-work> osssink probably doesn't know the available formats
[00:28] <thomasvs> ds-work: any idea what could have gone wrong there ?
[00:28] <ds-work> so it refuses in the link function
[00:29] <thomasvs> hm, but it used to work fine
[00:29] <thomasvs> maybe someone changed that in osssink though
[00:29] <ds-work> perhaps
[00:29] <thomasvs> argh, for once I'd like to see consistent behaviour when I'm trying to test stuff :)
[00:29] <ds-work> it's one of the bad features with the current negotiation system
[00:30] yippi ([email protected]) left irc: "Client exiting"
[00:30] <ds-work> osssink should subtract out known-bad caps from getcaps
[00:30] <thomasvs> how should it do that ?
[00:31] <thomasvs> argh, well, that's for tomorrow, I'm wasted
[00:31] <thomasvs> feel free to send me a mail about that, I want our audio sinks to work
[00:31] Nick change: thomasvs -> thomasvz
[00:33] <mathrick> nite
[00:33] Nick change: mathrick -> mathrick|sleep
[00:39] walters ([email protected]) joined #gstreamer.
[00:40] <Company> hey, buffer-frames was never an int property
[00:40] <Company> no element tested for it, it just got added because someone was confused
[00:40] Action: Company thinks people should blame him for every bug they have
[00:41] <thaytoo> "should"?
[00:41] <Company> should stop blaming
[00:41] <ds-work> Company: we do :)
[00:41] <Company> freud probably
[00:42] <Company> is ogg working fine for you btw?
[00:42] <thaytoo> rhythmbox is
[00:43] <Company> and do you use a recent gst-plugins (from today)?
[00:43] <thaytoo> almost always (yes)
[00:43] <Company> then it uses oggdemux/vorbisdec :)
[00:43] <Company> i was expecting at least something to be broken today
[00:44] <thaytoo> not sure - I haven't deleted the vorbisfile yet, so it might still be choosing that?
[00:44] Rotty ([email protected]) left irc: Remote closed the connection
[00:44] <taaz> the core has support for disabling enum types code gen.  i doubt anyone has tested that though. ;)  should plugins have same flags?  i added glib-mkenums to the gst-libs/gst/ dirs that need it.  not sure if it's ok to throw that in without flags to shut it off though
[00:45] <ds-work> loaded 149 plugins with 484 features
[00:45] <ds-work> only 516 more features for 1000
[00:46] <taaz> only 999516 till 1000000!
[00:47] <walters> heh
[00:49] <Company> taaz: the only disabling support that is supposed to work is --sidable-gst_debug
[00:49] <Company> taaz: everything else is most definitely broken because nobody tried it since 0.3 or so
[00:51] <Company> and we only have 484 features because each typefind function is a feature
[00:51] <Company> :p
[00:51] <ds-work> and I have 100+ ladspa plugins
[00:51] <Company> i think typefindfunctions is more than ffmpeg already ;)
[00:52] <ds-work> added plugin ladspa with 188 feature(s)
[00:52] <ds-work> added plugin typefindfunctions with 41 feature(s)
[00:52] <Company> yeah, i'm around 260 or so
[00:52] <Company> ffmpeg is 54 iirc
[00:55] Action: Zeenix goes into a chat-room where everyone is given the operator status :)
[00:56] <ds-work> kickban everyone :)
[00:59] <Zeenix> ds-work: actually i am on #anarchism on undernet so they are just practicing their theory of equality for all
[01:02] Zeenix ([email protected]) left irc: "Leaving"
[01:06] Januz ([email protected]) joined #gstreamer.
[01:06] <Januz> hi all, does anyone know where i can find the alsasink plugin?
[01:07] <ds-work> gst-plugins, like everything else
[01:07] <Januz> ok, thx
[01:10] <Januz> and where can i get the source code of it for 0.6.4?
[01:10] wheels ([email protected]) joined #gstreamer.
[01:13] Januz ([email protected]) left #gstreamer.
[01:16] harshy ([email protected]) left irc: "I quit for now"
[01:22] LeRoutier ([email protected]) joined #gstreamer.
[01:23] <LeRoutier> hello
[01:41] <LeRoutier> there is a problem while registration for a plug-in :
[01:41] <LeRoutier> added plugin mad with 2 feature(s)
[01:41] <LeRoutier>  
[01:41] <LeRoutier> (process:32581): GStreamer-CRITICAL **: Could not convert static caps "video/x-raw-yuv, format = (fourcc) YUY2, width = (int) 720, height = (int) { 480, 576 }, framerate = (float) { 25.0, 29.997 }; video/x-raw-rgb, bpp = (int) 32, depth = (int) 32, endianness = (int) 4321, red_mask =   (int) 0x000000ff, green_mask = (int) 0x0000ff00, blue_mask =  (int) 0x00ff0000, width = (int) 720, height = (int) { 480, 576 }, framerate = (float)
[01:41] <LeRoutier>  { 25.0, 29.997 }; video/x-raw-rgb, bpp = (int) 24, depth = (int) 24, endianness = (int) 4321, red_mask =   (int) 0x000000ff, green_mask = (int) 0x0000ff00, blue_mask =  (int) 0x00ff0000, width = (int) 720, height = (int) { 480, 576 }, framerate = (float) { 25.0, 29.997 }"
[01:42] <LeRoutier> added plugin dvdec with 1 feature(s)
[01:42] <LeRoutier> i don't know if it's with mad or dvdec
[01:42] <thaytoo> dvdec
[01:43] <ds-work> LeRoutier: update ext/dv
[01:43] <LeRoutier> k, compiling gst-plugins actually
[01:43] <LeRoutier> thx
[01:44] <LeRoutier> hum, quite a bunch of updated files. you're really active guys
[01:45] pb_ ([email protected]) left irc: "Client exiting"
[01:45] <ds-work> naw, we just change things randomly
[01:46] <LeRoutier> could have been a simple sed line that changed the copyright header in those files too :)
[01:54] harshy ([email protected]) joined #gstreamer.
[01:55] ct_ ([email protected]) joined #gstreamer.
[01:55] desrt ([email protected]) joined #gstreamer.
[01:58] <ds-work> is gst-player working for anyone?
[01:59] <LeRoutier> there's a regression in gst-player : i was able to resize the video window while playing an avi file. now, it segfaults when resizing
[01:59] <ds-work> specifically, mpeg seems to stop after about 2 seconds of playing
[01:59] <LeRoutier> trying a mpg now
[02:00] <LeRoutier> for me, it plays the mpg, still at 1fps / s but it plays it
[02:00] <ds-work> yeah, it seems to play it very slowly
[02:01] <LeRoutier> clock/time or fps nego issue
[02:01] <ds-work> obviously not a negotiation issue
[02:01] <LeRoutier> it does it for every mpg i have. since at least 1 week
[02:02] <ds-work> yeah
[02:02] <ds-work> it's not a gst-player issue
[02:03] <LeRoutier> resizing a playing mpg doesn't segfault like it does with avi
[02:04] <ds-work> ximagesink or xvimagesink?
[02:04] <LeRoutier> xv
[02:05] <LeRoutier> another new issue with gst-player+avi that i didn't have yesterday is the duration+slider : it says the film i'm looking at is 19s long. unfortunatly, current time is actually 1m40 and going
[02:06] Action: desrt cvs up
[02:10] <LeRoutier> going off-line. good night
[02:10] LeRoutier ([email protected]) left irc: "Leaving"
[02:10] <thaytoo> ds-work: did you just update dv and I just haven't seen the commit?
[02:11] <ds-work> thaytoo: hours ago
[02:11] <thaytoo> hrmn. I haven't seen the log email
[02:13] <ds-work>  odd
[02:17] <Company> the sf email is flakey
[02:17] <Company> i'm 99% sure i wrote an email sying oggdemux/vorbisdec is in
[02:24] <sxpert> ah, it somewhat works better, now it *compiles*, but still doesn't work 
[02:25] <sxpert> filename: /media/music/Carl Cox/U60311 Compilation techno division Vol. 3 (Disc 2)/10 - Slam - Stepback (smith + selway mix).ogg
[02:25] <sxpert> position 3 1075425908040577000
[02:25] <sxpert> position 4 1075425910040577000
[02:25] <sxpert> what did I do wrong ? do I need to seek to the beginning or something ?
[02:34] sublett ([email protected]) left irc: Remote closed the connection
[02:34] walters ([email protected]) left irc: Read error: 60 (Operation timed out)
[02:36] Company ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[02:38] Company ([email protected]) joined #gstreamer.
[02:46] <ds> Company: do we still need a float2int that works?
[02:46] <Company> ds: dunno
[02:47] <Company> ds: i didn'T just delete the old one as i didn'T want to upset anyone
[02:47] <Company> and was to bored to write ntoone
[02:47] <ds> -EPARSE
[02:48] <ds> er, no
[02:48] <ds> nm
[02:49] <ds> so how does vorbisdec work?
[02:49] <Misirlou> I think it decodes Vorbis.
[02:55] <Company> vorbisdec takes a raw vorbis stream and uses libvorbis to make a raw float audio stream out of it - just like libvorbis does
[02:55] <ds> and how does one play the raw float audio?
[02:55] <Company> and it's different from vorbisfile in that vorbisfile uses vorbis/vorbisfile.h and vorbisdec uses vorbis/codec.h
[02:55] <Company> ds: float2intnew
[02:56] <ds> ah, ok


-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn