IRC Logs

*******************************************************************
[03:45] _gst_newt_ joined #gstreamer.
[03:45] <BBB> dunno, I'm sure you'll now explain me why it's a good thing ;)
[03:45] <ds-work> it's better than dory, whose goal is to merely wrap an element
[03:46] <ds-work> so I can do filesrc ! dory child=mad ! osssink
[03:46] ChrisHJW_log ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[03:46] <ds-work> baby steps, here
[03:47] <BBB> I have a dory :)
[03:48] <BBB> it wraps gst-rec source elements
[03:48] <BBB> I'm supposed to document on how to do that and when it's useful in the manager thing of the PWG
[03:48] <BBB> it's fairly useful
[03:49] <BBB> I did the same thing for interface emulation (that sucked... I wrapped the 'wrapped element' inside another wrapper)
[03:49] <BBB> but that was too slow :/
[03:51] ChrisHJW_log ([email protected]) joined #gstreamer.
[03:54] <taaz> blah
[03:54] <BBB> ola
[04:00] <taaz> aren't people supposed to maybe ask debian maintainers what's up before announcing an "adoption" which is really a hijacking?
[04:02] <BBB> ?
[04:02] <Rounin> Ech, I'll do it some other time
[04:02] <BBB> taaz: ?
[04:02] <Rounin> Most of my files are MP3s anyway, so the mad sink works OK
[04:02] <BBB> mad is not a sink?
[04:03] <Rounin> Err... The mad whatever
[04:03] <Rounin> The mad thing
[04:03] <BBB> element
[04:03] <Rounin> #define everything G_THING
[04:03] <Rounin> How about mad hatter
[04:03] <BBB> dead hatter
[04:04] <Rounin> Tes,
[04:04] <Rounin> Yes,
[04:04] <BBB> taaz: what's wrong?
[04:05] <ds-work> taaz: you don't acknowledge bugs and otherwise look like an MIA maintainer.  What is he supposed to do?
[04:18] <taaz> ds-work: send email?  send patches?
[04:19] <ds-work> taaz: you look like an MIA maintainer
[04:19] <BBB> what is MIA and wtf are you guys talking about? :p
[04:20] <ds-work> missing-in-action
[04:20] <Rounin> This discussion is meaningful
[04:20] <taaz> BBB: some guy decided to "adopt" gst debian work without even pinging me first
[04:20] Action: BBB just looked through the debian mailinglist archives but couldn't find much
[04:20] <BBB> debian is too big ;)
[04:20] <Rounin> It's open source isn't it
[04:20] <BBB> got an uri? :)
[04:20] <ds-work> "some guy" == the main Debian GNOME developer
[04:21] <BBB> I guess this mess is why I never wanted to be maintainer or packager or have any important function... I'd rather just code ;)
[04:22] Action: BBB pokes taaz to re-start development instead of doing that package management stuff
[04:26] <taaz> BBB: there's probably a delay before msgs get to the archives.  it's probably on gmane though
[04:27] <BBB> which list?
[04:28] <taaz> 239114 debian-devel debian-gtk-gnome
[04:28] <taaz> man, i guess i just suck
[04:29] <BBB> found it
[04:29] <BBB> well, he's nice, no?
[04:30] <BBB> you should let thomasvs do the deb. packaging ;)
[04:30] <BBB> and he's not taking over, he's just helping, right?
[04:30] <BBB> he doesn't say he's taking over the package
[04:31] <taaz> um... "adopt" means take over vs just a NMU
[04:31] <BBB> it does?
[04:31] Action: BBB wonders why debian uses all those terms in a different fashion than what he's used to
[04:31] <BBB> is that part of your social contract?
[04:32] <taaz> there are some docs related to this stuff
[04:33] <taaz> usually people orphan packages and others adopt them.  people like me apparently look like they are MIA and people just adopt those packages too.  i think it's better to NMU packages till the maintainer specifically gives them up or is really mia
[04:34] <BBB> can he just "adopt" it?
[04:34] <BBB> I mean
[04:34] <BBB> doesn't he need permission?
[04:35] <taaz> not really.  it's sort of a general understanding vs strict rules
[04:36] <BBB> well, sorry for you... are you still maintainer or does this make you completely debian-maintainer-off?
[04:37] <BBB> Feel free to join the team (see the
[04:37] <BBB> gnome-pkg-tools package) if you're going to resume actively maintaining
[04:37] <BBB> GStreamer in future.
[04:37] <BBB> that's not so bad...
[04:44] sub_pop ([email protected]) left irc: "Client exiting"
[04:48] <taaz> yeah, except i didn't want to use svn ;)
[04:49] <taaz> just as i finally get this arch thing under control...
[04:49] <taaz> eh, i'll just start some dialog about it and at a minimum offer up my current work
[04:49] <BBB> look at it from the bright side, you didn't lose your deb account :)
[04:50] <taaz> which i figured i'd upload sometime this week
[04:50] <taaz> just need to merge 1 or 2 patches from CVS and rebuild at this point
[04:51] <BBB> go for the dialog
[04:51] <BBB> :)
[04:51] Action: BBB goes for bed/sleeptime
[04:51] <BBB> seeya later 'all
[04:51] BBB ([email protected]) left irc: "Client exiting"
[04:51] <taaz> question is if the gnome team wants to take over all the gst stuff... core, plugins, editor, player, python, ruby, c#, etc etc
[04:52] <taaz> and what do the kde folk think about that? ;)
[04:54] <Rounin> Well, doesn't Gstreamer rely heavily on Glib?
[04:54] <Rounin> I don't see how KDE could have anything to say about it
[05:07] <ds> KDE is essentially planning to use GStreamer in 3.0
[05:07] <ds> taaz is just being a troll :)
[05:09] <Rounin> KDE should use GTK in 3.0 too
[05:09] <Rounin> I keed I keed; after spending two days on putting a background image onto a widget, I've realized GTK sucks too
[05:09] <ds> uh, why?  they alread have Qt
[05:09] <Rounin> Exactly. They have Qt. That was sort of the essence of my point. =P
[05:09] Action: ds dismissed Rounin as a troll, too
[05:10] <Rounin> In that case I dismiss you as a troll
[05:10] Action: ds dismisses everyone as a troll
[05:10] <taaz> i think both gnome and kde should switch to OpenStep
[05:10] <ds> :)
[05:10] <Rounin> taaz, hang on, isn't that really ugly?
[05:10] Nick change: Jaramir -> Jara[zZ]
[05:11] <taaz> isn't what ugly? gnome? kde? yes! ;)  <-- troll
[05:11] <Rounin> OpenStem
[05:11] <Rounin> I mean OpenStem
[05:11] <Rounin> I mean "I mean OpenStep"
[05:11] <taaz> i used NeXTStep for 5+ years without thinking so
[05:12] <Rounin> Oh OK
[05:41] spyder482 ([email protected]) joined #gstreamer.
[05:42] brock ([email protected]) joined #gstreamer.
[05:42] <brock> Yo yo yo!
[05:46] <brock> Um, is this thing on?!
[05:47] <brock> c'mon Andy Wingo, anyone?
[05:51] <brock> This internet thingy is like broken... ...and stuff.
[05:51] <brock> I'm outta here.
[05:51] brock ([email protected]) left irc: "Leaving"
[05:55] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[06:07] Rounin ([email protected]) left irc: "Client exiting"
[06:13] <taaz> anyone want to check my response about the debian packages before i send it off?  ds?
[06:18] <ds> um, ok
[06:25] <ds> it's a bit wordy, perhaps
[06:25] <taaz> yeah ;)
[06:30] <ds> yes, you are being a troll
[06:30] <ds> nobody fucking cares about KDE vs. GNOME
[06:30] <ds> especially not people in Debian
[06:32] <Phython> taaz: are you telling netscipe to fuck off and that you'll get to uploading gstreamer 0.8 to experimental at some point?
[06:32] <taaz> i'm trying not to
[06:33] <ds> he pretty much deserves it, though
[06:35] <taaz> i guess i'm not sure why gst should be done by gnome team other than just more man power to work on things
[06:36] <taaz> maybe it's just the "team" concept encourages something that the "send patches" method doesn't... i dunno
[06:36] <Phython> taaz: just cut the kde stuff and say you'll keep the package for the above reasons and upload something to experimental
[06:37] <Phython> taaz: start a gstreamer-team on alioth ;)
[06:37] sub_pop ([email protected]) joined #gstreamer.
[06:43] <taaz> Phython: yeah, i thought about doing that and putting all my gst related packages there as well as any that other people want there
[06:54] ChrisHJW ([email protected]) joined #gstreamer.
[07:12] mxpxpod ([email protected]) joined #gstreamer.
[07:50] tvon ([email protected]) joined #gstreamer.
[07:51] <tvon> Howdy folks
[07:51] <tvon> I'm having an issue where gstreamer opens up a seperate window for video
[07:51] <tvon> eg, in totem or gst-player, if I open a video it leaves the totem/gst window blank and opens a seperate window to show the video in
[07:51] <tvon> took me a while to realize what was going on since the seperate window is real tiny at first, I have to resize it to see whats going on
[07:59] <ds> what gstreamer version?
[08:05] Shoragan ([email protected]) joined #gstreamer.
[08:09] spyder482 ([email protected]) left irc: "Leaving"
[08:16] <tvon> ds 8
[08:18] markey ([email protected]) joined #gstreamer.
[08:19] <ds> given that there's only been one version with an 8 in it, I'll assume you mean 0.8.0
[08:19] <ds> tvon: file a bug, please.  it shouldn't do that
[08:21] Action: dilinger blinks
[08:21] <dilinger> taaz: was netsnipe's "adoption" of gst worked out w/ you?
[08:22] <ds> er, no
[08:22] <ds> it's a hijack to fix a wishlist bug
[08:23] <dilinger> oh, i see there's been discussion in here about this already
[08:23] Action: dilinger reads scrollback
[08:23] <ds> not a lot
[08:23] <tvon> ds: k, will do
[08:27] <dilinger> taaz: tell netsnipe to fuck off, start a gstreamer alioth packaging project (i'd join, once i have my packages in better shape), and tell him he's welcome to join ;)
[08:27] <dilinger> taaz: my $0.02
[08:28] sub_pop ([email protected]) left irc: "Client exiting"
[08:29] ChrisHJW ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[08:29] sub_pop ([email protected]) joined #gstreamer.
[08:34] <taaz> dilinger: i need to nap on it.  i'm inclined to do that vs moving this stuff to gnome management
[08:35] <taaz> i see seb128 threw my unfinished packages online ;)
[08:37] <dilinger> taaz: it really has no place being maintained by the gnome team
[08:38] <dilinger> and on that note..
[08:39] Action: dilinger packages a long overdue devmapper
[09:27] bitshifter ([email protected]) joined #gstreamer.
[09:50] abo ([email protected]) joined #gstreamer.
[09:59] tromey ([email protected]) left irc: Remote closed the connection
[10:03] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[10:04] alley_cat ([email protected]) left irc: "May the Source be with you!"
[10:10] alley_cat ([email protected]) joined #gstreamer.
[10:15] ChrisHJW_log ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:18] alley_cat ([email protected]) left irc: "May the Source be with you!"
[10:19] alley_cat ([email protected]) joined #gstreamer.
[10:25] alley_cat ([email protected]) left irc: "May the Source be with you!"
[10:29] jd_Zzz ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:29] alley_cat ([email protected]) joined #gstreamer.
[10:31] Nick change: markey -> markey|afk
[10:36] bitshifter ([email protected]) left irc: "Client exiting"
[10:43] jdahlin ([email protected]) joined #gstreamer.
[10:49] bitshifter ([email protected]) joined #gstreamer.
[10:59] bitshifter ([email protected]) left irc: "Client exiting"
[11:07] Nick change: mathrick|sleep -> mathrick
[11:07] <mathrick> g'mornin
[11:07] <mathrick> (for some values of morning ;)
[11:08] Action: mathrick notes there were quite some amusing ppl hanging around in here today ;)
[11:08] thomasvs ([email protected]) joined #gstreamer.
[11:09] <mathrick> tvon: still there? first check if you don't have xvideosink set as default in gconf key
[11:09] <mathrick> tvon: you need to have x[v]imagesink in there, cuz xvideosink doesn't implement proper embedding interface an will result in separate window popping up
[11:10] <mathrick> yo thomasvs
[11:11] <thomasvs> morning
[11:11] <mathrick> thomasvs: what tz are you? GMT +1?
[11:12] Action: mathrick tries to place barcelona somewhere on map
[11:14] <wheels> mathrick: Spain is on the same as the rest of western europe...
[11:14] <mathrick> wheels: ok, so what would it be? +1?
[11:14] <wheels> mathrick: yeah
[11:15] <mathrick> and with DST applied, it reduces to +0 i presume?
[11:15] <wheels> mathrick: Uhm, other way.  :-)
[11:16] <mathrick> ah, right
[11:16] Action: wheels wonders if mathrick spent time in the US -- Europeans don't usually say "daylight savings time"
[11:17] <ds> hmmm...
[11:17] <ds> euopeans awake.  must be time to go to sleep
[11:17] <mathrick> nope, that's the easiest thing to type ;)
[11:17] <mathrick> and besides, what do Europeans say for DST? ;)
[11:17] <wheels> mathrick: just "summer time" or "winter time"
[11:18] <wheels> mathrick: The English do anyway...
[11:19] dolphy ([email protected]) joined #gstreamer.
[11:19] sublett ([email protected]) joined #gstreamer.
[11:19] <jdahlin> ds, what time is it over there?
[11:19] <mathrick> wheels: ah, that's exact translation of what we say in polish
[11:19] <jdahlin> in sweden we mostly say summer time and winter time
[11:19] <mathrick> jdahlin: 2:18 :)
[11:20] <jdahlin> yeah, bed time :)
[11:21] <mathrick> jdahlin: nope, bed time is 4:00 :)
[11:21] iain ([email protected]) joined #gstreamer.
[11:21] <ds> 2:20
[11:21] <mathrick> especially fun if your classes start @ 8:30 :)
[11:22] <jdahlin> I'm a little bit wasted, since I didn't go to bed until 03
[11:22] <iain> I went to bed at two and woke at 3:30 thinking I had a full nights sleep
[11:23] <mathrick> iain: that's what I call "convention syndrome" :)
[11:31] alley_cat ([email protected]) left irc: "May the Source be with you!"
[11:32] alley_cat ([email protected]) joined #gstreamer.
[11:32] bitshifter ([email protected]) joined #gstreamer.
[11:40] wheels ([email protected]) left irc: "using sirc version 2.211+KSIRC/1.3.9"
[11:58] ChrisHJW_log ([email protected]) joined #gstreamer.
[11:59] mxb ([email protected]) joined #gstreamer.
[11:59] mxb ([email protected]) left #gstreamer ("Leaving").
[13:05] bilboed_school ([email protected]) joined #gstreamer.
[13:16] teuf ([email protected]) joined #gstreamer.
[13:16] alley_cat ([email protected]) left irc: Remote closed the connection
[13:18] alley_cat ([email protected]) joined #gstreamer.
[13:28] ensonic ([email protected]) joined #gstreamer.
[13:29] <ensonic> hi hi
[13:42] teuf ([email protected]) left irc: "Client exiting"
[13:42] kmaraas ([email protected]) joined #gstreamer.
[14:05] ploum ([email protected]) joined #gstreamer.
[14:12] markey|afk ([email protected]) left irc: "leaving"
[14:29] bitshifter ([email protected]) left irc: Remote closed the connection
[14:33] bitshifter ([email protected]) joined #gstreamer.
[15:13] jdahlin ([email protected]) left irc: "Leaving"
[15:13] jdahlin ([email protected]) joined #gstreamer.
[15:23] Jara[zZ] ([email protected]) left irc: Read error: 110 (Connection timed out)
[15:26] Jara[zZ] ([email protected]) joined #gstreamer.
[15:32] Rounin ([email protected]) joined #gstreamer.
[15:32] <Rounin> Hm... a'rpi has left mplayer
[15:42] markey ([email protected]) joined #gstreamer.
[15:44] kmaraas ([email protected]) left irc: Read error: 110 (Connection timed out)
[16:26] wingo ([email protected]) joined #gstreamer.
[16:27] <wingo> yop
[16:28] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[16:29] <jdahlin> hello wingo
[16:29] walters ([email protected]) joined #gstreamer.
[16:34] Action: ensonic noticed that gst not yet has a memory-source and sink - how would you guys construct a pipeline, where you want to get the data at the end (read in a sample) and not just push it through
[16:38] <bitshifter> ensonic: fdsink with fd = one end of a pipe;  alternatively, you could probably use fakesink and set 'signal-handoffs' to TRUE, and then just ref and use the buffer you get passed via the 'handoff' signal
[16:39] <wingo> hi jdahlin
[16:40] <wingo> ensonic: the idea, normally, is to keep things in the pipeline. but for this kind of thing where it might take time to render the data to a buffer, but playback needs to happen quickly,
[16:40] <wingo> that's more like what marlin implemented as a backend.
[16:40] <wingo> i'm planning to use it as well in an app -- but it really can't be a single array, it has to be chunked to allow nonlinear editing
[16:41] <wingo> something like an emacs buffer.
[16:42] <wingo> the reason it hasn't been done is because most existing apps are just of the playback / recording / transcoding variety.
[16:46] ploum ([email protected]) left irc: "Leaving"
[16:51] teuf ([email protected]) joined #gstreamer.
[16:52] sublett ([email protected]) left irc: "I quit, therefore I was"
[16:53] <ensonic> wingo, you know we are writing a sound composer and there we need to load samples, for that we would like to write pipelines to load the samples to memory and the offer a source that offers these waves
[16:54] <ensonic> wingo: btw. as you do sound stuff as well, do you use internal elements or so? current gst has not many sound related plugins yet
[16:56] <wingo> i have a fuckload of internal elements :)
[16:57] <wingo> i would look to marlin as a model. perhaps it doesn't suit your purposes, but i think it's pretty neat. i don't know very much about working with samples, though (unfortunately)
[16:57] <ensonic> wingo, what the heck is marlin?
[16:58] <teuf> a cute girl
[16:59] <teuf> ensonic: http://marlin.sourceforge.net/
[16:59] <teuf> or http://www.lenemarlin.com/ as you prefer ;)
[17:04] Action: ensonic is getting marlin from gnome cvs
[17:05] <ensonic> wingo: would be useful to have some of you internal gstelems as gst-plugins (I guess)
[17:05] Action: jdahlin prefers lenemarlin.com
[17:07] <ensonic> irks, it needs glib 2.3.0 (have only 2.2.3)
[17:12] <thomasvs> gtk-doc is so frigging fragile
[17:13] <jdahlin> wingo: can you create new elements with the guile bindings?
[17:15] <wingo> jdahlin: yes, but processing with guile is slow
[17:15] <wingo> see the sinesrc in examples/gstreamer/sinesrc.scm
[17:15] spyder482 ([email protected]) joined #gstreamer.
[17:15] <wingo> i need to investigate integration with guile-numeric
[17:16] <wingo> bins are quick of course
[17:16] <bitshifter> thomasvs: typo in configure.ac, line 326 should be USE_PROFILING, not UES_PROFILING
[17:16] <thomasvs> bitshifter: what where how
[17:16] <wingo> ensonic: most are special-purpose synthesis elements
[17:16] <jdahlin> wingo: same with python
[17:17] <jdahlin> wingo: but I wrote a small filesrc like element which wasn't that much slower
[17:17] <wingo> jdahlin: can you do caps now? and represent the buffer as something other than characters?
[17:17] <bitshifter> thomasvs: sorry, gstreamer module, configure.ac, line 326
[17:17] <jdahlin> wingo: some caps yes. I don't think I can create new ones yet
[17:18] <thomasvs> bitshifter: thanks, fixed
[17:18] <thomasvs> wingo: have you ever thought about writing an assembler plugin that rearranges buffer sizes going out ?
[17:18] <thomasvs> wingo: or did you just never need something like that for audio ?
[17:18] <wingo> thomasvs: rearranges buffer sizes?
[17:19] <wingo> like it chunks buffers differently?
[17:19] <thomasvs> wingo: for video for example - getting input from the network has smaller buffer chunks than the actual video frame
[17:19] <thomasvs> wingo: so you need to reassemble complete frame buffers before display
[17:19] <thomasvs> wingo: yeah, basically
[17:19] <thomasvs> wingo: I'm wondering if that applies to audio in some cases as well.
[17:19] <wingo> thomasvs: my buffer-frames-convert still isn't in cvs (ak, haven't had time), but that's what it does
[17:20] <wingo> only it uses caps to decide how to chunk buffers
[17:20] <thomasvs> wingo: yeah, it would have to do the same for video
[17:20] <thomasvs> look at caps, figure out how big a data buffer those caps would need, reassemble
[17:20] <thomasvs> wingo: so, it's not in cvs why ?
[17:20] <wingo> but in general (i think) you should use bytestream for that
[17:20] <wingo> thomasvs: lemme fix that ;)
[17:20] <thomasvs> wingo: maybe it should have a different name tough
[17:20] <thomasvs> like, rechunk, or rebuffer or something
[17:21] <wingo> well... lemme fix that tomorrow. it will take time to update cvs, rebuild, etc.
[17:21] <thomasvs> sure
[17:21] <thomasvs> well, you have two testers for your new element on the video side :)
[17:21] <wingo> why don't you want to use bytestream?
[17:22] <wingo> it can be very efficient, merging buffers when possible
[17:23] <thomasvs> wingo: why rewrite xvimagesink to use bytestream if it already works well ?
[17:23] <thomasvs> and, why do you write the element in the first place if you can use bytestream ? :)
[17:23] sublett ([email protected]) joined #gstreamer.
[17:23] bitshifter ([email protected]) left irc: "Client exiting"
[17:25] <wingo> thomasvs: because for me it's a boundary between different caps-spaces
[17:25] <wingo> buffer-frames is a caps property in float systems
[17:25] <wingo> it's somewhat analogous to the size of a video frame, in that sense
[17:25] <wingo> so you need an element that sits between the caps of a network source, and the caps that can link to a video sink
[17:26] <wingo> that element should do the defragmentation
[17:26] <wingo> at first glance, anyway ;)
[17:26] alley_cat ([email protected]) left irc: "May the Source be with you!"
[17:28] <thomasvs> right, which is why I'm asking you would use bytestream for our case but not yours
[17:28] <thomasvs> and, if we should have a separate one
[17:28] <thomasvs> ie, how far to go in abstracting it
[17:29] <wingo> you need an element to convert caps. that element uses bytestream (or something like it). for audio, it's buffer-frames-convert between different values of buffer-frames. for video, ... what element actually gives you video caps from the network source?
[17:30] <wingo> what is your pipeline?
[17:30] <thomasvs> wingo: tcpsrc ! xvimagesink for example
[17:30] <thomasvs> wingo: the caps are set filtered on that link
[17:30] <wingo> hm.
[17:30] <thomasvs> so ideally it would be tcpsrc ! assembler ! video_caps ! xvimagesink
[17:31] <thomasvs> and assembler can figure out buffer size from the caps easily, but with video-specific code
[17:31] <thomasvs> although wim had a nice idea to use buffer size querying
[17:31] <wingo> i think ideally it would be tcpsrc ! video_caps ! xvimagesink maybe
[17:31] <thomasvs> wingo: I don't know - why put the functionality in tcpsrc at all ?
[17:31] <wingo> that the video_caps does the assembly
[17:31] <thomasvs> hm
[17:31] <thomasvs> that would involve new API then
[17:31] <thomasvs> ie, buffer_size_from_caps or something
[17:31] <wingo> because it's a property of video caps that there's one frame per buffer
[17:32] <thomasvs> actually, not such a bad idea
[17:32] <thomasvs> well, yeah, exactly
[17:32] <thomasvs> in fact
[17:32] <thomasvs> by now it just worked by chance, really :)
[17:32] <wingo> anything that claims to convert to video caps with not one frame per buffer is lying ;)
[17:32] <thomasvs> since all video-providing sinks just happen to output one frame per buffer
[17:32] <wingo> quick, make it a rule :)
[17:32] <thomasvs> but since audio doesn't have a concept of "frame" above the sample level ....
[17:32] <thomasvs> ... it wasn't really an issue there
[17:32] <thomasvs> interesting
[17:33] <wingo> gotta go, bro. good luck :)
[17:33] Nick change: wingo -> wingo-beer
[17:33] <thomasvs> hm.  well, htat just leaves me with four different ways to implement the same idea.  which I hate :)
[17:33] <wingo-beer> hehe
[17:51] tvon ([email protected]) left #gstreamer.
[18:04] <ensonic> anyone knows how to make gnome devhelp aware of gst docu?
[18:05] <jdahlin> ensonic: it is already
[18:05] <jdahlin> ask thomasvs
[18:05] <thomasvs> ensonic: it should be, gtk-doc does that automatically
[18:05] <thomasvs> ensonic: is it generating the html correctly ?
[18:06] <ensonic> thomasvs: jup html is okay
[18:06] <Rounin> Hey, what would gstreamer need to replace mplayer anyway? Some codecs and splitters, right?
[18:08] <ensonic> thomasvs: can I manually add the docs?
[18:09] Nick change: markey -> markey|afk
[18:09] <thomasvs> Rounin: yeah
[18:09] <thomasvs> Rounin: and output sinks
[18:09] <thomasvs> ensonic: well, is there a .devhelp file that gets generated too ?
[18:09] <thomasvs> during docs build that is ?
[18:11] <Rounin> thomasvs: Hmm, it seems like that could work
[18:15] sub_pop ([email protected]) left irc: "Client exiting"
[18:15] <ensonic> thomasvs: no, under docs is just an empty devhelp folder
[18:21] ChrisHJW ([email protected]) joined #gstreamer.
[18:33] teuf ([email protected]) left irc: "Client exiting"
[18:36] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[18:41] sub_pop ([email protected]) joined #gstreamer.
[18:47] mxpxpod ([email protected]) joined #gstreamer.
[18:49] <thomasvs> ensonic: huh ?
[18:50] <thomasvs> ensonic: I meant, does it generate gstreamer.devhelp along with the html ?
[18:50] <thomasvs> ensonic: check buildbot for an example of build output and compare with yours for the docs
[18:50] <mxpxpod> does gstreamer do quicktime?
[18:53] <ensonic> thomasvs,: ok, there is ./docs/gst/html/gstreamer.devhelp
[18:55] <ensonic> thomasvs: where is the files installed on you machine?
[18:58] <thomasvs> /usr/share/gtk-doc/html/gstreamer-0.8/gstreamer-0.8.devhelp
[18:58] <thomasvs> ensonic: ie, in the same install location as the gtk-doc stuff
[18:58] <thomasvs> ensonic: if you're compiling from source however, you are probably installing not in /usr so your /usr-installed devhelp doesn't pick it up
[18:59] <ensonic> thomasvs: it is /opt/gnome/share/gtk-doc/html/ here, but there is no gstreamer-0.8 there
[19:00] <ensonic> thomasvs: there is no gtk-doc under /usr it's all under /opt/gnome
[19:02] <thomasvs> ensonic: ok, then check your build log why it's not installing that file there
[19:04] wingo_ ([email protected]) joined #gstreamer.
[19:08] wingo-beer ([email protected]) left irc: Read error: 110 (Connection timed out)
[19:17] Nick change: markey|afk -> markey
[19:23] Action: ensonic has no buildlog :-( but will try to findout what is going wrong anyway
[19:23] sub_pop ([email protected]) left irc: "Client exiting"
[19:30] jdahlin ([email protected]) left irc: "Leaving"
[19:30] <ensonic> have a nice easter break, cu
[19:30] sub_pop ([email protected]) joined #gstreamer.
[19:30] ensonic ([email protected]) left #gstreamer ("Leaving").
[19:43] Rotund ([email protected]) left irc: Remote closed the connection
[19:45] thomasvs ([email protected]) left irc: Read error: 60 (Operation timed out)
[19:56] jdahlin ([email protected]) joined #gstreamer.
[19:58] Phython_ ([email protected]) joined #gstreamer.
[19:58] Phython ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[20:07] pb_ (~pb@2002:5160:4550:1:240:95ff:fe30:aa84) joined #gstreamer.
[20:16] mat1t ([email protected]) joined #gstreamer.
[20:18] <mat1t> has anbody installed GStreamre 0.8.0 under FreeBSD 5.2.1 ?
[20:21] arana ([email protected]) joined #gstreamer.
[20:24] LeRoutier ([email protected]) joined #gstreamer.
[20:25] <LeRoutier> hello
[20:27] wheels ([email protected]) joined #gstreamer.
[20:30] <mat1t> Hi
[20:38] spyder482 ([email protected]) left irc: Read error: 113 (No route to host)
[20:39] mathrick ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[20:41] Nick change: arana -> spyder482
[20:46] arana ([email protected]) joined #gstreamer.
[20:59] apoc ([email protected]) joined #gstreamer.
[21:00] jdahlin ([email protected]) left irc: "Leaving"
[21:02] spyder482 ([email protected]) left irc: Read error: 110 (Connection timed out)
[21:03] KoRnouille ([email protected]) left irc: "Leaving"
[21:04] teuf ([email protected]) joined #gstreamer.
[21:05] apoc ([email protected]) left irc: "Leaving"
[21:09] mat1t ([email protected]) left irc: Remote closed the connection
[21:16] bilboed_school ([email protected]) left irc: "Leaving"
[21:21] spyder482 ([email protected]) joined #gstreamer.
[21:29] Action: ds-work wonders where Company is
[21:32] Nick change: spyder482 -> spyder_away
[21:33] arana ([email protected]) left irc: Read error: 113 (No route to host)
[21:42] arana ([email protected]) joined #gstreamer.
[21:42] Nick change: arana -> spyder482
[21:44] sub_pop ([email protected]) left irc: "Client exiting"
[21:46] yippi ([email protected]) joined #gstreamer.
[21:51] taaz ([email protected]) left irc: Remote closed the connection
[21:52] taaz ([email protected]) joined #gstreamer.
[21:52] taaz ([email protected]) left irc: Remote closed the connection
[21:53] taaz ([email protected]) joined #gstreamer.
[21:54] spyder_away ([email protected]) left irc: Read error: 113 (No route to host)
[22:05] LeRoutier ([email protected]) left irc: "Leaving"
[22:05] <ds-work> rb still leaks about 50 MB loading 5000 songs
[22:13] <wheels> ds-work: Have you kissed valgrind this week?
[22:13] <ds-work> no
[22:14] <ds-work> I suppose I could look at the problems, but I'm not feeling motivated
[22:15] <wheels> ds-work: A valgrind a day keeps the, uhm...
[22:16] <ds-work> leak moster away?
[22:23] ChrisHJW ([email protected]) left irc: Connection timed out
[22:28] wingo_ ([email protected]) left irc: Remote closed the connection
[22:30] thomasvs ([email protected]) joined #gstreamer.
[22:33] Nick change: spyder482 -> spyder_away
[22:34] sxpert ([email protected]) left irc: Remote closed the connection
[22:45] Zeenix ([email protected]) joined #gstreamer.
[22:47] <Zeenix> hi
[22:58] sxpert ([email protected]) joined #gstreamer.
[23:03] <Zeenix> why we use xml and not sgml for the docs?
[23:04] markey ([email protected]) left irc: "leaving"
[23:05] <Rounin> Isn't it the other way round?
[23:05] Ed_Mack ([email protected]) joined #gstreamer.
[23:06] <Zeenix> Rounin: don't know, i dont think that I understand the www build system at all
[23:09] <Rounin> Oh... I haven't even heard about it
[23:10] <Ed_Mack> Would you ever consider having the volume element have a dbus connection by default (is the dependancy a big deal)? It would be great to have Gnome-wide application volume control, so they could all be turned down by say a gnome-meeting call, or just nicely managed. Is that a stupid idea?
[23:10] <ds-work> Ed_Mack: the volume element is specifically for changing the stream, not changing the system volume
[23:11] Nick change: harshy -> harshyGone
[23:11] <Ed_Mack> I know, that's what's nice about it
[23:12] <Ed_Mack> Then you can have a central consistant control, as well as individual apps' approaches
[23:12] <Ed_Mack> Longhorn does this I think too
[23:12] mxpxpod ([email protected]) left irc: Read error: 110 (Connection timed out)
[23:13] <ds-work> Ed_Mack: imo, that's all a sound server issue
[23:13] <Ed_Mack> ok
[23:13] <ds-work> Ed_Mack: of course, no sound server bothers to solve that :)
[23:13] bilboed ([email protected]) joined #gstreamer.
[23:14] <Ed_Mack> heh, thanks for your input. It's just an idea I had to bat off my head
[23:14] mxpxpod ([email protected]) joined #gstreamer.
[23:15] <ds-work> Ed_Mack: I think of a sound server like a window manager -- if you get a gnome-meeting call, it's just like if you have a window pop up
[23:15] <ds-work> so the sound server/window manager should put other things into the background
[23:15] <ds-work> of course, there may be some interaction with the application to handle that, depending on the protocol
[23:16] <Ed_Mack> What is Gnome's sound server? I'm guessing me not knowing is a positive gui thing :)
[23:16] <ds-work> esd
[23:16] <Ed_Mack> Ok
[23:16] <ds-work> it's particularly crappy
[23:17] <Ed_Mack> Is the new x.org related one (MAS?) much better?
[23:17] <ds-work> heh, don't get me started about MAS
[23:18] <Zeenix> Ed_Mack: I've heard that their latest version works :)
[23:19] <Ed_Mack> I hope you don't mind me asking, but what's bad about esd?
[23:19] <ds-work> critical bugs, or just major design flaws?
[23:20] <Ed_Mack> The biggies :)
[23:21] <ds-work> for the former, it's impossible for a gstreamer application to use gnome sound events, since libesd can't handle multiple connections
[23:21] <ds-work> also, esd gets very confused if the application does things outside of exactly what it expects
[23:22] <Ed_Mack> Are there better alternatives? esd does sound crappy!
[23:22] <ds-work> such as bug #136572
[23:22] <ds-work> jack is allegedly nice, but it doesn't work with OSS
[23:24] <ds-work> as for major design flaws, it doesn't really understand streaming well
[23:24] <ds-work> it doesn't handle low-latency apps (games)
[23:24] <ds-work> the protocol is crappy
[23:24] <ds-work> it only handles stereo (not 5.1)
[23:25] <ds-work> it's network transparency is flawed
[23:25] <Ed_Mack> I think you're off on a tangent now
[23:25] <ds-work> <ds-work> heh, don't get me started about MAS
[23:25] <ds-work> s/MAS/ESD/
[23:31] sxpert ([email protected]) left irc: Remote closed the connection
[23:36] sxpert ([email protected]) joined #gstreamer.
[23:38] sublett ([email protected]) left irc: "I quit, therefore I was"
[23:43] sublett ([email protected]) joined #gstreamer.
[23:44] yippi ([email protected]) left irc: "Client exiting"
[23:49] yippi ([email protected]) joined #gstreamer.
[00:00] --- Fri Apr  9 2004
[00:10] teuf ([email protected]) left irc: Remote closed the connection
[00:48] sub_pop ([email protected]) joined #gstreamer.
[00:58] Ed_Mack ([email protected]) left irc: "Leaving"
[00:59] LeRoutier ([email protected]) joined #gstreamer.
[00:59] <LeRoutier> re
[01:01] <LeRoutier> anyone know this error ? ** ERROR **: file gstcolorspace.c: line 330 (gst_colorspace_get_format): should not be reached
[01:09] <ds-work> yeah, that's bad
[01:10] <ds-work> that should not ever happen :)
[01:10] <ds-work> what pipeline?
[01:10] <LeRoutier> well, it happened
[01:10] <iain> morning
[01:10] <LeRoutier> gst-launch v4lsrc device=/dev/video0 ! spider ! ximagesink
[01:10] <LeRoutier> hi iain
[01:10] <ds-work> LeRoutier: ah, ok
[01:10] <LeRoutier> same for : gst-launch v4lsrc device=/dev/video0 ! colorspace ! ximagesink
[01:11] <LeRoutier> but : gst-launch v4lsrc device=/dev/video0 ! ffmpegcolorspace ! ximagesink   works ok, got the video
[01:11] <ds-work> LeRoutier: could you file a bug with a --gst-debug=GST_PADS:5 log?
[01:11] <LeRoutier> ok, no prob, doing the log now
[01:12] <LeRoutier> doing the same log with ffmpegcolorspace to be able to compare both ?
[01:13] <ds-work> no
[01:13] <LeRoutier> ok, log done, opening the bug now
[01:18] <LeRoutier> ds-work, http://bugzilla.gnome.org/show_bug.cgi?id=139532
[01:18] <LeRoutier> i'll use ffmpegcolorspace for now, no hurry
[01:19] <LeRoutier> i still have to figure out how to setup the tuner frequency & standard but i'll look to the code for that
[01:19] <ds-work> you should, anyway
[01:20] <ds-work> colorspace is complete crap
[01:21] <LeRoutier> well, i did not choose it first, spider choosed it
[01:21] <LeRoutier> perhaps some plugin priority to change between the 3 colorspace converters
[01:22] <ds-work> i thought I fixed that
[01:22] <ds-work> oh well
[01:22] <Rounin> I have one of those odd errors again
[01:22] <Rounin>                 while(gst_element_get_state(myThread)==GST_STATE_PLAYING)
[01:22] <Rounin>                 {
[01:22] <Rounin>                 ;//gtk_main_iteration_do(NULL);
[01:22] <Rounin>                 }
[01:22] <Rounin> That function , gst_element_get_state(myThread)==GST_STATE_PLAYING, takes up 80-100% of my CPU
[01:22] <Rounin> And it didn't do that yesterday
[01:23] <ds-work> Rounin: er, why are you spinning in a loop?
[01:23] <Rounin> It's supposed to print the stream position to a stream
[01:23] <Rounin> Err, to a label
[01:23] <Rounin> The ; // is just to assure that it doesn't do anything
[01:23] <Rounin> ANd yet it takes up almost 100% CPU
[01:24] <ds-work> so it sits there and takes 100% cpu
[01:24] <Rounin> If I comment out the while it doesn't do that
[01:24] <Rounin> Isn't that at all odd?
[01:24] <ds-work> no
[01:24] <LeRoutier> well, you should not loop
[01:25] <ds-work> why is it surprising that while(1){} takes 100% cpu?
[01:25] <LeRoutier> imho, you should see from time to time what the position is, with a timer
[01:25] <Rounin> I see... So I'll go back to calling it every .10 seconds from the event loop
[01:25] <Rounin> Thanks
[01:25] yippi ([email protected]) left irc: "Client exiting"
[01:25] <LeRoutier> yep, it would avoid this awfull infinite loop
[01:26] Shoragan ([email protected]) left irc: "Leaving"
[01:31] <LeRoutier> Rounin, on what kind of app are you working ?
[01:37] <Rounin> LeRoutier: A simple MP3 player
[01:37] <Rounin> Just thought I'd do it for fun
[01:38] <LeRoutier> if it's a way to learn the API, go for it
[01:38] Action: wheels giggles a bit at the loop above...
[01:43] <Rounin> Ah, now, I remember why I called this so often
[01:43] <Rounin> When I put it in its own function, I get this:
[01:43] <Rounin>                 printf(" "); //This printf just has to be here for the function to work
[01:43] <Rounin>                 if(myThread!=NULL)
[01:43] <Rounin> I need to pritnf something each time it gets called, or myThread evaluates to 0
[01:44] <Rounin> It's true
[01:44] <Rounin> I had this bug yesterday and I never figured it out
[01:45] <LeRoutier> should i say it looks really odd ?
[01:45] <Rounin> No shit
[01:45] <Rounin> Well, at least it works
[01:45] <Rounin> I wonder if an sprintf would work too
[01:47] Zeenix ([email protected]) left irc: "The whole problem with the world is that fools and fanatics are always so certain of themselves, but wiser people so full of 
[01:47] walters ([email protected]) left irc: Client Quit
[01:47] <Rounin> Hm, if I remove the if statement altogether it doesn't run either
[01:48] <Rounin> So I need to have a printf and if there in order to make it work
[01:49] <Rounin> Hm, and if (1) doesn't work either
[01:49] <Rounin> I hate C
[01:52] pb_ (~pb@2002:5160:4550:1:240:95ff:fe30:aa84) left irc: "Client exiting"
[01:56] BBB ([email protected]) joined #gstreamer.
[01:57] <LeRoutier> hello ronald
[02:00] <BBB> evenin'
[02:00] Action: BBB ill
[02:00] <BBB> :/
[02:00] Action: ds-work sympathetic
[02:03] Action: ds-work wonders why ffmpegcolorspace has a rank of NONE
[02:05] <BBB> dunno
[02:05] <BBB> probably to prevent spider from messing up?
[02:05] <ds-work> LeRoutier: try that
[02:05] <ds-work> well, it's PRIMARY now
[02:05] <Rounin> Hey BBB, you wanna try solving my bug?
[02:05] <Rounin>                 printf(" ");
[02:05] <Rounin>                 if(myThread!=NULL)
[02:05] <Rounin> Evaluates as 1
[02:06] <Rounin> if(myThread!=NULL)
[02:06] <Rounin> Evaluates as 0
[02:06] <Rounin> The printf is actually what makes the difference
[02:06] <BBB> I told you to use valgrind for debugging :)
[02:06] <LeRoutier> Rounin, add a 'return 1;' at the end
[02:06] <Rounin> LeRoutier: Of the function?
[02:06] <LeRoutier> yep
[02:07] <Rounin> .............Ooooh
[02:07] <Rounin> I'll give it a shot
[02:07] <Rounin> It is supposed to return 1 when I think about it
[02:07] <LeRoutier> ds-work, trying what ? changing ffmpeg rank to have spider choose it ?
[02:07] <ds-work> LeRoutier: I checked some stuff in
[02:08] <Rounin> LeRoutier: My hero. :D
[02:08] <ds-work> (heh, wasn't that obvious?)
[02:08] <LeRoutier> sorry, did not see it, updating it
[02:08] <LeRoutier> Rounin, it's a question of logic
[02:08] <Rounin> printf returns 1?
[02:08] ds-work ([email protected]) left irc: Remote closed the connection
[02:08] <LeRoutier> even if called in void contect, printf evaluates to something
[02:08] <Rounin> LeRoutier: I think my brain still runs on CMOS
[02:10] ds-work ([email protected]) joined #gstreamer.
[02:10] <ds-work> dude, mozilla can go kiss my ass
[02:10] <ds-work> it just crashed X _again_
[02:11] <ds-work> arguably, X shouldn't crash, but
[02:11] <LeRoutier> ds-work, still xfree or did you update to xorg ?
[02:12] <ds-work> xfree from sid
[02:12] <Rounin> The only acceptable incarnations of mozilla are Thunderbird and Firefox
[02:12] <ds-work> or epiphany
[02:12] <BBB> why are our man pages versioned as 0.8.1 and our tools as 0.8?
[02:13] <LeRoutier> BBB: seems you're really hill
[02:13] <LeRoutier> -h
[02:14] <BBB> thank you
[02:14] <BBB> why?
[02:14] <LeRoutier> man 1 gst-inspect-0.8
[02:14] <BBB> er
[02:14] <BBB> wait
[02:14] <BBB> hm
[02:14] Action: BBB kicks himself
[02:14] Action: BBB goes back to bed
[02:16] abo ([email protected]) left irc: Read error: 113 (No route to host)
[02:18] <LeRoutier> BBB: got a small question : is there a way to set tuner frequency/standard in the pipeline (via a v4lsrc element property) or should i go to C directly ?
[02:18] <BBB> no, I removed those
[02:18] <BBB> company re-added them to v4l2src
[02:18] <BBB> but never did that for v4lsrc
[02:18] <BBB> I guess I should re-add them at some point
[02:19] <LeRoutier> ok, unfortunatly, i've got v4lsrc working but not v4l2src
[02:19] <LeRoutier> i'll wait, no prob
[02:19] <LeRoutier> BBB: does gst-recorder now require another gstreamer than HEAD ? (needs an unknow .pc file to pass configure)
[02:22] <BBB> er
[02:22] <BBB> oh
[02:22] <BBB> I probably committed some brokenness as for the quality interface
[02:22] <BBB> I'll remove that
[02:22] <BBB> I added that to the configure file, but it shouldn't be there yet
[02:22] <BBB> feel free to remove it
[02:23] <LeRoutier> ok, i'll do that
[02:25] <LeRoutier> ds-work, spider still doesn't like me : it uses ffmpegcolorspace now but it negotiates really bad video size
[02:26] <BBB> v4lsrc probably cannot handle size re-negotiations very well or so
[02:26] <BBB> I'll have a look
[02:26] <LeRoutier> k, thx
[02:27] <LeRoutier> BBB: should i fill a bug as a reminder for that or the frequency/standard properties readdition to v4lsrc ?
[02:27] <BBB> yes please
[02:27] Action: BBB feels sick and wants to go to bed but he promised his company to do some server updates this night
[02:27] <BBB> grmlb
[02:28] <LeRoutier> ok, doing so now
[02:31] <LeRoutier> ds-work, look at this one :) : http://bugzilla.gnome.org/show_bug.cgi?id=139119
[02:39] Action: BBB just installed 0.8.0 at work...
[02:44] sublett ([email protected]) left irc: "I quit, therefore I was"
[02:47] <LeRoutier> BBB: 2 reminder "bugs" added : http://bugzilla.gnome.org/show_bug.cgi?id=139535 & 139537
[02:47] iain ([email protected]) left irc: "!"
[02:53] <BBB> ty
[02:53] <BBB> I'll check them later
[02:53] <BBB> like the media playback bugs
[02:53] Action: BBB hates being sick
[02:53] <BBB> I can't work then
[02:53] <BBB> :(
[02:54] <LeRoutier> don't worry, you'll have plenty of time to fix them while i'll be skiing
[02:54] Action: LeRoutier feels cruel
[02:55] <BBB> er...
[02:55] <BBB> right
[02:55] <BBB> :)


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