IRC Logs

*******************************************************************
[03:08] <Company> man, marlin is dead sexy
[03:08] <Company> it doesn't work most of the time, but it looks gorgeous
[03:08] <Company> kinda like gst-recorder
[03:09] <BBB> :X
[03:09] <BBB> :D
[03:09] <BBB> gst-rec works, sort of ;)
[03:10] <BBB> I just don't release it since some people tell me v4lsrc is broken and I don't know how to test that
[03:10] <BBB> apart from that, it's currently in a workable state
[03:10] <BBB> I can record from videotestsrc/sinesrc ;)
[03:11] Action: Company pokes ds again
[03:11] <LeRoutier> and would it work from a dv1394 source ?
[03:12] <BBB> dunno
[03:12] <BBB> I don't have such a source
[03:16] <LeRoutier> it lists a "Firewire (1394) DV Source) for videosrc but no such source for sound
[03:17] <BBB> I guess that to integrate nicely, it'll require some small modifications
[03:17] <BBB> but basic recording should work
[03:17] <Company> yeah, but basic is broken ;)
[03:17] <Company> wah, i want Martin on IRC _now_
[03:17] <LeRoutier> ok, so i'll try to import a video from my cam in the next days
[03:18] <BBB> Company: not *that* basic ;)
[03:18] <BBB> the other basic
[03:18] <BBB> LeRoutier: ok...
[03:18] Action: BBB fears crashes ;)
[03:19] <LeRoutier> well, if something show up in "video encoder" section. i know i have several of them but there are none in the listbox
[03:20] <BBB> what does dv1394 output?
[03:20] <LeRoutier> might be some crappy mjpeg like i think
[03:20] <Company> video/dv, no?
[03:20] <BBB> a systemstream?
[03:21] <BBB> because gst-rec does 'source' -> ['encoder' ->] 'muxer'
[03:21] <BBB> so if the muxer doesn't know dv, it won't work
[03:21] <BBB> and I don't think avimux knows dv
[03:21] <LeRoutier> oh, no spider element in the middle
[03:21] <BBB> ?
[03:22] <BBB> spider is for playback
[03:22] <BBB> not for encoding ;)
[03:22] <BBB> actually
[03:22] <BBB> that's not entirely true
[03:22] <BBB> but it's a good approximation
[03:22] <LeRoutier> something to convert dv to a more friendly format in the middle (between src & encoder)
[03:23] <BBB> nah
[03:23] <BBB> it should just record dv then
[03:23] <BBB> if you have a hardware MPEG encoder, you don't want it to decode MPEG in software and encode to divx, do you?
[03:23] <BBB> that'd be crappy
[03:24] <LeRoutier> native dv format takes loads of space, barelly compressed
[03:25] <BBB> yup
[03:25] <BBB> but dv decoding is expensive...
[03:26] <LeRoutier> doesn't dvgrab do it in real time with reencoding and all ?
[03:26] <BBB> no :)
[03:27] <LeRoutier> ok, so he just builds an avi file without changing video format
[03:28] <BBB> yes
[03:28] <BBB> it's a dv type 1 (interleaved) or type 2 (deinterleaved) AVI
[03:28] <BBB> I guess that for special DV support, I'd need to add some DV specific code to gst-rec
[03:29] <LeRoutier> but you won't be able to test it as you lack hardware
[03:30] Nick change: harshy -> harshyOut
[03:30] <BBB> yup
[03:31] <LeRoutier> would there be a way for gst-recorder in the future to choose the channel from which to record when using v4l ? (setting tunner and the like)
[03:31] <LeRoutier> or is it more for a tv app ?
[03:31] <BBB> it has such settings
[03:31] <BBB> it just lacks the tuner frequency right now
[03:31] <LeRoutier> oh, did not see that
[03:31] <BBB> I didn't add that yet
[03:31] <LeRoutier> k
[03:31] <BBB> but I will
[03:31] <BBB> it's supposed to be a TV/capture app
[03:31] <BBB> like xawtv and the like
[03:32] <BBB> but more capture-oriented
[03:32] <Company> i want a GSet
[03:33] <BBB> hm?
[03:33] <LeRoutier> BBB: i see a 'frequency' property actually but i don't know if it's just UI or if it is wired behind
[03:33] <BBB> the UI doesn't use frequency yet
[03:33] <BBB> apart from that, it's all there
[03:34] <LeRoutier> and anyone working on gst-tv those days ?
[03:34] <BBB> I think gst-tv and gst-overlay are largely dups
[03:34] <LeRoutier> would it be more like tvtime or more like zapping ?
[03:34] <BBB> gst-overlay is mine, gst-tv is ds' and company's
[03:34] <BBB> :)
[03:34] <BBB> iirc
[03:34] <Company> nope
[03:34] <BBB> what's the difference?
[03:34] <Company> my gst-tv is in libgstui
[03:35] <BBB> oh
[03:35] <BBB> ok
[03:35] <BBB> so we have 3 tv apps
[03:35] <BBB> gst-overlay works ;)
[03:35] <Company> my app is just a testbed for my widgets so far
[03:36] <BBB> should I release gst-overlay some day? :)
[03:36] <BBB> prolly not
[03:36] <BBB> too simple :p
[03:38] <Company> yeah, channel selection kinda sucks ;)
[03:38] <BBB> *nod*
[03:38] <Company> s/channel/frequency/
[03:38] <BBB> like in gst-rec ;)
[03:38] <BBB> I need to work on that... I had some ideas for a channel editor like in src or xawtv
[03:38] <BBB> but then HIG compliant
[03:38] <BBB> but given the amount of work, I never finished that so far :/
[03:39] <BBB> mail me if you have cool ideas on that
[03:39] <BBB> I need to get a drink
[03:39] <BBB> ;)
[03:40] <LeRoutier> and i need some sleep. alcohol doesn't make any effect any more. and it's 4AM here
[03:41] <LeRoutier> good night ppl
[03:43] Action: Company doesn't get why that pipeline ever worked qith basic
[03:45] LeRoutier ([email protected]) left irc: "Leaving"
[03:45] <BBB> it worked some time ago :)
[03:45] <BBB> say, two weeks
[03:45] <BBB> should I do it without threads?
[03:47] <BBB> ERROR         default(13310) ./grammar.y(383):gst_parse_element_lock:<esdsink0> (null)
[03:47] Last message repeated 1 time(s).
[03:47] <BBB> ERROR: from element /pipeline0/esdsink0: Internal GStreamer error: state change failed.  File a bug.
[03:47] <BBB> what's that?
[03:47] <BBB> (process:13310): GStreamer-WARNING **: internal error: push on pad audioconvert0:sink but it has no chainhandler
[03:49] <Company> esd doesn't run
[03:49] <Company> most likely ;)
[03:49] <Company> hm no, that would result in its own error
[03:49] <Company> might be caps nego
[03:49] <BBB> gst-launch-0.8 filesrc location=/media/clips/lego_troopers_divx.avi ! avidemux name=d .audio_00 ! audioconvert ! audioscale ! esdsink d.video_00 ! divxdec ! ffcolorspace ! ximagesink --gst-scheduler=basicgthread
[03:49] <BBB> RUNNING pipeline ...
[03:49] <BBB> ERROR    audioconvert(13333) gstaudioconvert.c(245):gst_audio_convert_chain:<audioconvert0> (null)
[03:49] <BBB> ERROR    audioconvert(13333) gstaudioconvert.c(245):gst_audio_convert_chain:<audioconvert0> Sink pad not negotiated before chain function
[03:49] <BBB> ERROR: from element /pipeline0/audioconvert0: Internal GStreamer error: negotiation problem.  File a bug.
[03:49] <BBB> and same error with alsasink
[03:50] <BBB> or osssink
[03:51] Action: BBB wonders how a GNOME channel editor would look like
[03:52] <Company> list of channels, preselected according to LANG
[03:52] <Company> wqith the properies option to change the channels to something else
[03:52] <BBB> hm, actually, I guess you wouldn't do that, since it differs per region
[03:52] <BBB> so you'd probably include a list of known-good lists or so
[03:53] <BBB> and let the user select
[03:53] <BBB> e.g. 'NL - Groningen/Casema'
[03:53] <BBB> 'NL - Utrecht/Casema'
[03:53] <BBB> etc.
[03:56] <Company> well, you can just set "CHannel 1", "Channel 2", ...
[03:56] <Company> there are options to read out channel names though, so that might work
[03:59] <BBB> options? how?
[03:59] walters ([email protected]) left irc: Client Quit
[03:59] <BBB> isn't that through VBI?
[03:59] <BBB> I don't do any of that yet..
[04:01] <BBB> http://ronald.bitfreak.net/images/chanedit.png
[04:01] <BBB> rough sketch ;)
[04:02] <BBB> and then a non-editable combobox to select a channel in the preferences
[04:02] <BBB> bbl
[04:02] BBB ([email protected]) left irc: "Client exiting"
[04:25] sxpert ([email protected]) left irc: Read error: 110 (Connection timed out)
[04:30] <dtm> who else here is serious on mjpegtools, particularly mplex?  :)
[04:32] <dtm> i'm outta synch!
[04:33] sxpert ([email protected]) joined #gstreamer.
[05:14] danb ([email protected]) left irc: Read error: 110 (Connection timed out)
[05:22] mxpxpod ([email protected]) joined #gstreamer.
[05:32] Celtic-Fox ([email protected]) joined #gstreamer.
[05:34] Celtic-Fox ([email protected]) left #gstreamer.
[05:40] Nick change: spyder_away -> spyder482
[05:40] spyder482 ([email protected]) left irc: "Leaving"
[06:03] steve_b ([email protected]) joined #gstreamer.
[06:09] Nick change: steve_b -> steveb_away
[06:09] danb ([email protected]) joined #gstreamer.
[06:22] <Company> KDE :/
[06:43] jae ([email protected]) joined #gstreamer.
[06:46] jae ([email protected]) left irc: Client Quit
[06:46] jaebird ([email protected]) joined #gstreamer.
[06:49] <jaebird> What is up!
[06:57] jaebird ([email protected]) left irc: "Leaving"
[08:04] ChriHJW_log ([email protected]) left irc: Read error: 60 (Operation timed out)
[08:10] sub_pop ([email protected]) left irc: "Client exiting"
[08:12] sub_pop ([email protected]) joined #gstreamer.
[08:29] sub_pop ([email protected]) left irc: "Client exiting"
[09:12] defendguin ([email protected]) joined #gstreamer.
[09:12] <defendguin> how do you play a sone from the commandline with gst?
[09:13] apoc ([email protected]) joined #gstreamer.
[09:16] sxpert ([email protected]) left irc: "Client exiting"
[09:19] apoc_ ([email protected]) left irc: Read error: 60 (Operation timed out)
[09:28] danb ([email protected]) got netsplit.
[09:28] harshyOut ([email protected]) got netsplit.
[09:28] Company ([email protected]) got netsplit.
[09:28] mathrick|sleep ([email protected]) got netsplit.
[09:28] sjoerd_ ([email protected]) got netsplit.
[09:28] tromey ([email protected]) got netsplit.
[09:28] Jara[zZ] ([email protected]) got netsplit.
[09:28] noah ([email protected]) got netsplit.
[09:28] sjoerd ([email protected]) got netsplit.
[09:28] bitshifter ([email protected]) got netsplit.
[09:28] alley_cat ([email protected]) got netsplit.
[09:28] lilo ([email protected]) got netsplit.
[09:28] bilboed_ ([email protected]) got netsplit.
[09:28] Misirlou ([email protected]) got netsplit.
[09:28] wtay_ ([email protected]) got netsplit.
[09:28] dtm ([email protected]) got netsplit.
[09:28] wheels ([email protected]) got netsplit.
[09:28] taaz ([email protected]) got netsplit.
[09:28] defendguin ([email protected]) got netsplit.
[09:28] Rotund ([email protected]) got netsplit.
[09:28] ds ([email protected]) got netsplit.
[09:28] lilo_booter ([email protected]) got netsplit.
[09:28] md` ([email protected]) got netsplit.
[09:28] apoc ([email protected]) got netsplit.
[09:28] dholmes ([email protected]) got netsplit.
[09:28] mxpxpod ([email protected]) got netsplit.
[09:28] thomasvs ([email protected]) got netsplit.
[09:29] danb ([email protected]) returned to #gstreamer.
[09:29] harshyOut ([email protected]) returned to #gstreamer.
[09:29] Company ([email protected]) returned to #gstreamer.
[09:29] mathrick|sleep ([email protected]) returned to #gstreamer.
[09:29] mxpxpod ([email protected]) returned to #gstreamer.
[09:29] thomasvs ([email protected]) returned to #gstreamer.
[09:29] noah ([email protected]) returned to #gstreamer.
[09:29] sjoerd_ ([email protected]) returned to #gstreamer.
[09:29] sjoerd ([email protected]) returned to #gstreamer.
[09:29] tromey ([email protected]) returned to #gstreamer.
[09:29] Jara[zZ] ([email protected]) returned to #gstreamer.
[09:29] bitshifter ([email protected]) returned to #gstreamer.
[09:29] alley_cat ([email protected]) returned to #gstreamer.
[09:30] apoc ([email protected]) returned to #gstreamer.
[09:30] defendguin ([email protected]) returned to #gstreamer.
[09:30] bilboed_ ([email protected]) returned to #gstreamer.
[09:30] dtm ([email protected]) returned to #gstreamer.
[09:30] wheels ([email protected]) returned to #gstreamer.
[09:30] dholmes ([email protected]) returned to #gstreamer.
[09:30] ds ([email protected]) returned to #gstreamer.
[09:30] Misirlou ([email protected]) returned to #gstreamer.
[09:30] lilo_booter ([email protected]) returned to #gstreamer.
[09:30] md` ([email protected]) returned to #gstreamer.
[09:30] taaz ([email protected]) returned to #gstreamer.
[09:30] lilo ([email protected]) returned to #gstreamer.
[09:30] wtay_ ([email protected]) returned to #gstreamer.
[09:30] Rotund ([email protected]) returned to #gstreamer.
[09:30] defendguin ([email protected]) left #gstreamer ("Leaving").
[09:32] Nick change: mathrick|sleep -> mathrick
[09:32] <mathrick> mornint
[09:34] <mathrick> s/t//
[09:44] KoRnouille ([email protected]) joined #gstreamer.
[09:46] <KoRnouille> hello all
[09:46] <mathrick> hi KoRnouille
[09:46] <KoRnouille> :)
[09:54] ChrisHJW_log ([email protected]) joined #gstreamer.
[09:56] markey ([email protected]) joined #gstreamer.
[10:21] ChrisHJW ([email protected]) joined #gstreamer.
[10:29] thomasvs ([email protected]) left irc: Read error: 60 (Operation timed out)
[10:33] jdahlin ([email protected]) joined #gstreamer.
[10:41] wheels ([email protected]) left irc: "using sirc version 2.211+KSIRC/1.3.9"
[10:53] dolphy ([email protected]) joined #gstreamer.
[11:08] thomasvs ([email protected]) joined #gstreamer.
[11:14] markey ([email protected]) left irc: "bbl"
[11:16] <jdahlin> I think ronald really want that gnomevfssink patch to go in
[11:20] <thomasvs> or he just really likes mailing he has patches
[11:20] Action: thomasvs fears looking at buildbot
[11:23] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) joined #gstreamer.
[11:26] bilboed_ ([email protected]) left irc: "Client exiting"
[11:26] teuf ([email protected]) joined #gstreamer.
[11:34] sublett ([email protected]) joined #gstreamer.
[11:34] mathrick ([email protected]) left irc: ""Warp speed: 2.6""
[11:37] mathrick ([email protected]) joined #gstreamer.
[11:38] <teuf> thomasvs: just received spiritualized/ladies and gentlemen... and it seems pretty neat at first listening :)
[11:39] <thomasvs> teuf: ah, great :) do give it some time, it took me a while to get used to
[11:40] iain ([email protected]) joined #gstreamer.
[11:40] <teuf> I already like it when I listen to it while working, so it should be ok :) I'll take more time to listen to it this evening I guess
[12:02] <thomasvs> another day, another segfault fixed
[12:09] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) left irc: "Client exiting"
[12:09] Shoragan ([email protected]) joined #gstreamer.
[12:18] ensonic ([email protected]) joined #gstreamer.
[12:21] mathrick ([email protected]) left irc: Read error: 110 (Connection timed out)
[12:38] jdahlin ([email protected]) got netsplit.
[12:38] wtay_ ([email protected]) got netsplit.
[12:38] Misirlou ([email protected]) got netsplit.
[12:38] dtm ([email protected]) got netsplit.
[12:39] wheels ([email protected]) joined #gstreamer.
[12:41] jdahlin ([email protected]) returned to #gstreamer.
[12:41] dtm ([email protected]) returned to #gstreamer.
[12:41] Misirlou ([email protected]) returned to #gstreamer.
[12:41] wtay_ ([email protected]) returned to #gstreamer.
[12:42] <iain> where do you get id3tag from?
[12:42] <iain> is that wheels' thing?
[12:42] markey ([email protected]) joined #gstreamer.
[12:42] <thomasvs> no, it's a gneeral lib afaik
[12:42] <wheels> iain : No, my thing is TagLib
[12:43] <iain> oh, ok
[12:43] <wheels> iain : I think the one you're referring to is the one that comes with libmad.
[12:44] Action: markey recommends TagLib
[12:48] <iain> wheels: yeah, that looks like it, but Muine searches for a .pc file and one doesn't come with the tarball I have :/
[12:49] <wheels> iain : It doesn't come with one (I ranted about this recently), but some packages add one.
[12:50] <iain> oh, thats nice
[12:55] Action: markey looking @ muine info
[12:57] <markey> is it related to rhythmbox?
[12:58] <jdahlin> testsuite/elements/Makefile.am:17: invalid unused variable name: `gst_compprep_check_SOURCES'
[12:58] Last message repeated 1 time(s).
[12:58] LeRoutier ([email protected]) joined #gstreamer.
[12:58] <jdahlin> thomasvs?
[12:58] <LeRoutier> hello
[13:00] <wheels> markey : Well, especially since I just started looking at the libid3tag API again.  Users have to deal with the id3 frames and fields directly...
[13:00] <markey> wheels: frankly, they should consider switching to TagLib
[13:00] <markey> wheels: it's a nice lib
[13:01] <markey> wheels: C++ allergy is keeping them from adoption
[13:03] <wheels> markey : Well, even for a C API the libid3tag API leaves a bit to be desired.  Like it's not hard for me because I know how ID3v2 tags are structured, but if you didn't know that ID3v2 tags are build of frames with certain IDs and that those frames have certain fields it wouldn't be comprehensible...
[13:03] <wheels> (Though I do prefer it to id3lib by quite a bit.)
[13:03] <markey> TagLib's API is quite convenient
[13:03] ploum ([email protected]) joined #gstreamer.
[13:03] ploum ([email protected]) left irc: Client Quit
[13:04] ensonic ([email protected]) left irc: Read error: 60 (Operation timed out)
[13:07] <LeRoutier> hum, no one has problem with anon cvs ?
[13:11] <bitshifter> LeRoutier: gnome.org you mean? 
[13:13] <LeRoutier> nope, fdo
[13:14] <wheels> LeRoutier : works here
[13:14] <LeRoutier> ok, might be specific to my 'cvs' tool version which could be buggy
[13:15] <LeRoutier> last changed it 20 days ago, that's odd
[13:15] <bitshifter> works fine here, only anoncvs.gnome.org is stuck at the state from around March 25 for me
[13:16] <LeRoutier> it insults me with :
[13:16] <LeRoutier> cvs update: warning: server is not creating directories one at a time
[13:16] <LeRoutier> cvs [update aborted]: there is a version in gstreamer already
[13:16] <teuf> the gnome anoncvs servers doesn't seem to update anymore after the upgrade of the main cvs server
[13:16] <LeRoutier> i'm googling right now to find a similar problem (and hopefully a solution)
[13:16] <teuf> LeRoutier: rm -rf gstreamer doesn"'t help ?
[13:17] <LeRoutier> i already deleted the local cvs tree to do a clean checkout, it  didn't help
[13:17] <bitshifter> LeRoutier: try doing a checkout in another directory
[13:18] <LeRoutier> might be a gentoo specific problem (my CVS ebuild), i have to check
[13:28] Shoragan ([email protected]) left irc: "Leaving"
[13:32] Rotty ([email protected]) joined #gstreamer.
[13:41] jmmv ([email protected]) joined #gstreamer.
[13:41] <jmmv> hi
[13:50] bitshifter ([email protected]) left irc: "bye"
[13:53] <LeRoutier> hi jmmv
[13:53] <LeRoutier> argh, i confirm it is not fdo CVS server fault, i have it with every CVS ebuild i try, with packages from fdo, gnome & sf.net CVS. so it is a local problem
[13:59] ensonic ([email protected]) joined #gstreamer.
[14:10] teuf ([email protected]) left irc: "Client exiting"
[14:10] <LeRoutier> bbl
[14:10] LeRoutier ([email protected]) left irc: "Leaving"
[14:17] foser ([email protected]) joined #gstreamer.
[14:24] smoke ([email protected]) joined #gstreamer.
[14:27] Shoragan ([email protected]) joined #gstreamer.
[14:29] Nick change: markey -> markey|afk
[14:34] <ensonic> i've posted a bug as #138846, can anyone please try this and either confirm or reject it?
[14:46] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) joined #gstreamer.
[14:56] Nick change: markey|afk -> markey
[15:19] <Company> good morning angels
[15:19] <Company> ensonic: fixed
[15:25] <wheels> Company : Oh, you greet the voices when you wake up too?
[15:25] <wheels> Company : I thought it was only me that saw them...
[15:25] ensonic ([email protected]) left irc: Read error: 60 (Operation timed out)
[15:27] <lilo_booter> wheels, don't have an immediate answer to your LGPL query :-) - will reply after discussing the issue with our customer
[15:29] <wheels> lilo_booter : Yeah, as you can imagine -- while an annoyance -- that's somewhat important for us...
[15:30] <lilo_booter> totally understand - my preference was, and is, for LGPL
[15:30] <lilo_booter> unfortunately, doing this stuff for money restricts my freedoms a bit too :-/
[15:31] <markey> hey Company angel
[15:31] <markey> hi lilo_booter 
[15:31] <wheels> lilo_booter : That's why I usually ask for payment in food and drugs.
[15:31] <markey> rofl
[15:31] <lilo_booter> heh
[15:34] ensonic ([email protected]) joined #gstreamer.
[15:34] <ensonic> Company: many thanks, I will try in a minute
[15:34] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[15:38] taaz ([email protected]) left irc: Read error: 110 (Connection timed out)
[15:40] <ensonic> Company: what was it, I see now changes in the plugins/ext/ladspa
[15:41] dolphy ([email protected]) joined #gstreamer.
[15:41] ploum ([email protected]) joined #gstreamer.
[15:41] ploum ([email protected]) left irc: Client Quit
[15:42] <Company> ensonic: audioconvert'S getcaps function returned a wrong buffer-frames value
[15:45] bitshifter ([email protected]) joined #gstreamer.
[15:46] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[15:49] <ensonic> Company: ahh, thats definitely a long way from the actual error
[15:51] walters ([email protected]) joined #gstreamer.
[15:52] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[15:59] dolphy ([email protected]) joined #gstreamer.
[16:11] alley_cat ([email protected]) left irc: "May the Source be with you!"
[16:11] <wheels> heh -- everyone at work is stopping by to see my new bass.  I feel like I just had a child or something.  :-)
[16:12] <wheels> (This was new toy day -- hadn't bought a new instrument in a few years...)
[16:16] alley_cat ([email protected]) joined #gstreamer.
[16:17] ensonic ([email protected]) left irc: Read error: 60 (Operation timed out)
[16:20] ensonic ([email protected]) joined #gstreamer.
[16:21] <ensonic> now I get : gstmad.c:688: error: parse error before "GST_TIME_FORMAT"
[16:24] <thomasvs> ensonic: update core
[16:25] <thomasvs> ensonic: always update underlying modules first if you have errors like that
[16:32] <Company> wheels: is there a way to get rid of KDEs deprecation warnings?
[16:33] <Company> wheels: -Werror and the recernt KCmdLineArgs change don't like each other
[16:35] <ensonic> thomasvs,: I *have* updated the core, it had no new files, therefore I just used make + make install
[16:43] <thomasvs> ensonic: if it had no new files, you didn't need make and make install
[16:43] <thomasvs> ensonic: in any case, gstinfo.h got changed, and those two symbols added
[16:43] <thomasvs> anyone happen to know about a libmad-using player example that supports seeking and resyncing ?
[16:43] taaz ([email protected]) joined #gstreamer.
[16:53] wheels_ ([email protected]) joined #gstreamer.
[16:53] dholmes ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[16:54] <wheels_> Company : Hmm, looking.  It might just be possible to do a "#define KDE_DEPRECATED"
[16:55] Action: wheels_ uses GCC 2.95, which doesn't support the __deprecated__ attribute, so I don't see them...
[16:56] wheels ([email protected]) left irc: Nick collision from services.
[16:56] Nick change: wheels_ -> wheels
[16:56] ensonic ([email protected]) left irc: "Leaving"
[16:57] ensonic ([email protected]) joined #gstreamer.
[16:57] <ensonic> join #multisync
[16:57] <ensonic> sorry, xchat drives me crazy today
[16:58] <Company> wheels: you should start writing changelog entries btw
[16:59] <wheels> Company : Oh, you guys do that even for stuff that hasn't previously been released?
[17:00] <Company> wheels: yeah, we do that for nearly anything
[17:01] foser ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[17:02] foser ([email protected]) joined #gstreamer.
[17:02] <wheels> Crap.  There's no check in the KDE header to see if KDE_DEPRECATED is already defined, so redefining it will just cause yet another warning.  ;-)
[17:03] foser ([email protected]) left irc: Client Quit
[17:03] Action: wheels knows a dirty hack that would probably work, but hopes to find something better...
[17:08] <ensonic> I am still getting the error :-(  gstmad.c:688: error: parse error before "GST_TIME_FORMAT"
[17:10] <Company> ensonic: gstinfo.h:525
[17:16] <wheels> Company : Ok, I just checked something into kdelibs that will make it in the future such that KDE_NO_DEPRECATED can be defined to turn them off, but that doesn't help us now.  :-)
[17:17] <Company> wheels: can we just #ifdef the KCmdLineArgs:init to use the old one on 3.1 and the new one on 3.2?
[17:18] Action: thomasvs is getting a suspicion avidemux is handing corrupt mp3 data to mad
[17:19] <wheels> Company : yeah, that would work
[17:19] Toastie ([email protected]) joined #gstreamer.
[17:20] <Toastie> anyone packages 0.8 for debian?
[17:20] <ensonic> Company, ahh, it's new (I updated two hours ago)
[17:20] <Company> Toastie: taaz does
[17:20] <Toastie> taaz?
[17:21] <Toastie> Company: you know where?
[17:24] <Company> Toastie: i think he does it in his arch repo, but i dunno wherer that is, you gotta poke him
[17:24] <Company> Toastie: though it's probably still a bit early in the US to poke him right now
[17:26] <Toastie> oh
[17:26] Toastie ([email protected]) left irc: "swoosh"
[17:28] Action: ensonic is gritting his teeth, now doing cvs up, autoregen,... the 3rd time today
[17:28] <Company> poor ensonic
[17:28] <Company> but you don't need autoregen
[17:28] <Company> just make && make install is fine
[17:31] iain ([email protected]) left irc: Read error: 110 (Connection timed out)
[17:31] <ensonic> Company: unfortunately not (see previous post of thomasvs)
[17:33] <jdahlin> ensonic: builds fine on FC1 (three boxes tested), RH9, YD3 (ppc) ...
[17:34] fbenites ([email protected]) joined #gstreamer.
[17:34] <fbenites> hi, do somebody knows libdc1394?
[17:41] iain ([email protected]) joined #gstreamer.
[17:46] <ensonic> jdahlin: I guess I managed to somehow got an inconsistent state here and therefore lost a day doing development with gst (I am still rebuilding)
[18:05] sub_pop ([email protected]) joined #gstreamer.
[18:23] foser ([email protected]) joined #gstreamer.
[18:35] <ensonic> I've found out that a specific ladspa plugin advertises its parameter-range in the wrong order (swapped lower and upper bound)
[18:35] <ensonic> this results in a critical gobject failure
[18:36] <ensonic> should I prepare a patch that checks the ranges before using it (thus actually cure the bugs in the ladspa plugin)
[18:36] <ensonic> might be better that showing this to the user
[18:38] <ensonic> is there a define analog to DEBUG() like  WARN () ?
[19:00] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[19:01] <ensonic> Company: can you please have a look at the private msg
[19:03] sxpert ([email protected]) joined #gstreamer.
[19:10] sublett ([email protected]) left irc: "Why are all the really good songs short?"
[19:18] <Company> ensonic: there's LOG, DEBUG, INFO, WARN and ERROR
[19:18] <Company> ensonic: though i'd say the ladspa plugin needs fixage
[19:20] <ensonic> Company: I attached a patch to #138635 that cures it and mailed the original author, maybe you can review and accept/reject the patch for me (I am not yet member)
[19:22] <ensonic> Company: about the priv. post, I was trying to use the visualisation, is the pipline up-to-date or should I use a different set of plugins
[19:22] <Company> ensonic: you should make thomasvs get you cvs access, applying your patches is boring
[19:22] <Company> even though it boosts my buzilla closing stats
[19:22] <ensonic> Company: thomas has all details from already (since about week or so)
[19:23] <Company> ensonic: for historic reasons spider doesn't handle colorspace conversions
[19:25] <ensonic> gst-launch-0.8 esdmon ! goom ! { queue ! colorspace ! ximagesink } does not work either (could not link goom0 to queue0)
[19:26] <ensonic> I am just asking, because I ve got one who could post a thousand visualisation over to gst
[19:26] <ensonic> post=>port
[19:29] <Company> try ffcolorspace
[19:29] <Company> or is it ffmpegcolorspace?
[19:29] <Company> one of them
[19:30] <ensonic> uhm, now it is the time to check out gst-ffmpeg ;-)
[19:30] <Company> nope
[19:30] <Company> the colorspace is in gst-plugins
[19:30] <thomasvs> ensonic: I'll poke some admins again
[19:32] <ensonic> thomasvs, thx ;-)
[19:32] foser ([email protected]) left irc: "[ I want to believe ]"
[19:32] <ensonic> thomasvs, it is not so urgent, now I still have an excuse to do actual work
[19:32] mxpxpod ([email protected]) joined #gstreamer.
[19:32] <ensonic> Company, but the ff-variant are not
[19:33] <Company> it _is_ urgent
[19:33] <Company> at least for me ;)
[19:33] <Company> ensonic: lemme check
[19:33] <ensonic> Company, it is found it
[19:34] foser ([email protected]) joined #gstreamer.
[19:34] teuf ([email protected]) joined #gstreamer.
[19:35] jdahlin ([email protected]) left irc: "Leaving"
[19:36] <ensonic> Company: it works (still showing the vararg field stuff, so it is not the spider causing it)
[19:37] <Company> that vararg thing should be dead easy to find in gdb with --gst-fatal-warnings
[19:38] <ensonic> Company, I'll look that up in a minute
[19:38] <ensonic> synaestesia draws a lot of buggy pixels at the bottom
[19:39] <Company> that could be a colorspace issue, too
[19:39] <Company> (can be checked by using ximagesink with no colorspace)
[19:42] LeRoutier ([email protected]) joined #gstreamer.
[19:42] <LeRoutier> bouh
[19:43] <LeRoutier> any gentooist on the chan that did not update its portage tree in the last 24 hours ?
[19:44] <ensonic> Company: that with synaestesia seems to be related to the plugin, I tried sdlvideo and xvideo, without colorspace it wont start
[19:46] <foser> LeRoutier: how come ?
[19:46] <Company> ok, then it uses one of those very weird colorspaces
[19:47] <LeRoutier> foser, it's because none of my cvs ebuilds works with portage > 2.0.50-r1 and working versions are no more in portage
[19:47] <ensonic> Company: the vararg stuff is in  esdmon.c:251, can you please fix it (saved us using bugzilla)
[19:47] <Company> ha, it's fixed already ;)
[19:48] <Company> (i hope, i haven't tested it)
[19:48] <sri> hi there
[19:48] <ensonic> Company, : about the synaestesia, to me it looks like it advertises a too big picture, as the graphics look good except about 50 lines at the bottom
[19:48] <foser> LeRoutier: what exactly doesn't work ?
[19:48] Nick change: Company -> there
[19:48] <there> hi sri
[19:48] Nick change: there -> Company
[19:48] <ensonic> Company, esdmon looks obvious, i'll test with tomorows build ;-)
[19:51] <ensonic> Company: i'll look into synae* tomorrow, cu
[19:53] <ensonic> cu
[19:53] ensonic ([email protected]) left #gstreamer ("Leaving").
[20:01] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) left irc: "Client exiting"
[20:03] KoRnouille ([email protected]) left irc: "Leaving"
[20:05] sublett ([email protected]) joined #gstreamer.
[20:08] thomasvs ([email protected]) left irc: Read error: 60 (Operation timed out)
[20:18] wheels ([email protected]) left irc: "BitchX Official FTP Site -- ftp://ftp.bitchx.com"
[20:19] wheels ([email protected]) joined #gstreamer.
[20:19] pb_ (~pb@2002:5160:4550:1:240:95ff:fe30:aa84) joined #gstreamer.
[20:20] dolphy ([email protected]) joined #gstreamer.
[20:21] <ds-work> !
[20:25] <Company> ds-work: any objections to me adding valgrind support in gst?
[20:25] <Company> ds-work: (optional, only used when valgrind is present and debugging is enabled of course)
[20:25] <ds-work> Company: like checks to see if it's running, and if so, disable bugs?
[20:25] <ds-work> seems reasonable
[20:26] <Company> ds-work: most important thing is disabling mempools
[20:26] <Company> ds-work: because you can't detect leaked GstBuffers if not
[20:26] <ds-work> I'd like a big fat reminder every time I start gstreamer in valgrind that it's executing different code
[20:26] <ds-work> Company: yeah, good idea
[20:27] <Company> ds-work: no problem
[20:27] Action: Company was just shown valgrind.h
[20:27] Action: ds-work wonders if he ever checked in gstbuffer.c changes
[20:27] kwm ([email protected]) joined #gstreamer.
[20:30] spyder482 ([email protected]) joined #gstreamer.
[20:32] The_Company ([email protected]) joined #gstreamer.
[20:34] alley_cat ([email protected]) left irc: Read error: 110 (Connection timed out)
[20:36] jdahlin ([email protected]) joined #gstreamer.
[20:36] thomasvs ([email protected]) joined #gstreamer.
[20:37] <ds-work> and the answer is no
[20:37] alley_cat ([email protected]) joined #gstreamer.
[20:41] <thomasvs> and the question is '?'
[20:42] <LeRoutier> thomasvs : * ds-work wonders if he ever checked in gstbuffer.c chang
[20:42] <LeRoutier> +es
[20:42] <ds-work> the answer is now yes
[20:42] <ds-work> "do not go to the elves for council, for they will tell you both no and yes"
[20:43] <walters> ds-work is an elf!
[20:44] <thomasvs> smelves
[20:44] Action: thomasvs just realized how very different mad seeking is for audio only as compared to inside an avi
[20:45] <ds-work> actually, that's "Go not to the elves for counsel, for they will say both yes, and no"
[20:46] <ds-work> of course, spelling counsel correctly
[20:49] Company ([email protected]) left irc: Read error: 110 (Connection timed out)
[20:51] <taaz> hmm... is disabling gst-compprep-check a good idea?  it exposes problems so seems like fixing distcheck issues would be a good idea if someone has the time
[20:52] <ds-work> thomasvs: btw, we have gstreamer.net now.  Were we going to give it to gnome.org?
[20:53] <The_Company> wee, i've written a working configure check in first try
[20:53] <The_Company> and it was with pkg-config
[20:53] <The_Company> wheels: beat that! :p
[20:53] Nick change: The_Company -> Company
[20:54] Action: wheels will be distracted from anything semi-productive for the next few hours while playing with his new bass.  :-)
[20:54] <Company> go wheels :)
[20:54] <taaz> wheels: i see you added a fix for arts debian stuff.  nice to see that
[20:55] <wheels> taaz: You mean the KDE includes?
[20:55] <wheels> taaz: (There was a debian fix for those...)
[20:56] Action: Company lures KDE devels to fix all KDE-related integration problems
[20:56] <taaz> wheels: yeah.  i just gave up on that with the debian packages. ;)  it confused me too much. i hope your patch fixes it ;)
[20:57] seb128 ([email protected]) joined #gstreamer.
[20:57] <Company> taaz: that was for the new kiosrc plugin though - so you will have more KDE stuff from 0.8.1 onwards ;)
[20:57] <seb128> hey
[20:57] <jdahlin> Company: we're not too far from tricking them into using it!
[20:57] <wheels> taaz: Actually it probably didn't.  But if you let me know the issue it probably could.  :-)
[20:57] <taaz> seb128: sure they can be public.  other than not having the source diff up yet they should work ok
[20:57] <seb128> hey jdahlin :)
[20:57] <jdahlin> hello seb128
[20:57] <seb128> taaz: ok, thanks
[20:58] <seb128> taaz: a lot of people are waiting on new rhythmbox/gstreamer apparently
[20:58] <jdahlin> taaz: I managed to embed xvimagesink in a gtk+ widget in python today
[20:58] BBB ([email protected]) joined #gstreamer.
[20:58] <BBB> thomasvs: could you do something for me?
[20:58] <BBB> thomasvs: add a note about gst-register somewhere in the FAQ?
[20:59] <taaz> seb128: yeah, i know. walters had said he wasn't going to uplaod 0.7.x rb to debian though
[20:59] <seb128> taaz: why ?
[20:59] <seb128> taaz: I should ask him so, I was going to upload it ...
[21:00] <taaz> jdahlin: cool.  when i did that in vdv it was hard to get right.
[21:00] <taaz> walters: ping
[21:00] <Company> valgrind rocks
[21:01] <walters> taaz: pong?
[21:01] <jdahlin> taaz: it's still quite hard
[21:01] <Company> just wanted to mention that
[21:01] <jdahlin> I was forced to use --sync, which is sort of a hack
[21:01] <seb128> walters: hey
[21:01] <walters> taaz: oh
[21:01] <walters> taaz: yeah.
[21:01] <seb128> walters: 0.7 should not be uploaded in experimental ?
[21:01] <walters> seb128: experimental is fine, sure
[21:01] <wheels> Company: See -- you've practically turned into a KDE developer already.  You're already praising the holy software.
[21:01] <seb128> walters: gnome2.6 is in experimental :)
[21:01] <walters> heh
[21:02] <walters> i don't think you have to be a KDE developer to appreciate that valgrind is cool :)
[21:02] <seb128> so it was not for unstable :p
[21:02] <Company> wheels: valgrind is not KDE
[21:02] <wheels> walters: Right, but I think you do to be a card-carying member of the cult of valgrind.
[21:02] <Company> wheels: you can see that easily because it has a .pc file :p
[21:02] Action: walters flashes his card :)
[21:04] <taaz> seb128: i really need to just get an arch mirror setup on arch.d.o.  then people can, in theory at least, help get the packages working better
[21:04] <Company> wheels: it doesn't require a KApplication to work either ;)
[21:05] <seb128> taaz: ok
[21:05] <seb128> taaz: do you want to use the Gnome's Team SVN ?
[21:05] <seb128> (probably not if you're using arch)
[21:05] <taaz> um... well, since i just started using arch...
[21:05] <wheels> Company: Well, I guess you'll have to talk to walters about obtaining your card then.  He's apparently nabbed a stash of them...
[21:06] <taaz> i just needed to use something other than cvs ;)
[21:06] <seb128> taaz: just let me know if you want to use the SVN, adding you to the alioth project is a 1min job :)
[21:06] <taaz> does svn support distributed development yet?  last time i tried it it was still single server based, which is just no good
[21:06] <seb128> single server ...
[21:07] <taaz> sigh... i do like svn ideas but the distributed repo concept is very useful
[21:08] <seb128> I'm not convinced
[21:08] <seb128> BTW arch is usefull to have a history of local changes
[21:08] <taaz> does svn let you unplug from network and make commits?
[21:08] <tromey> no
[21:08] Zeenix ([email protected]) joined #gstreamer.
[21:09] <tromey> svn is a central-server model like cvs
[21:09] <Zeenix> hi
[21:09] <seb128> taaz: no, but is that usefull ?
[21:09] <tromey> my personal fave for next-gen version control is monotone. http://www.venge.net/monotone/
[21:09] <taaz> tromey: long time ago they were saying it was designed so it could be extended at some point to do such things
[21:09] <tromey> I think there may be some layer over it which allows it
[21:10] <tromey> with monotone distributed development is built in
[21:10] Action: wheels thinks that the next generation is just FTP.  Version history encourages carelessness.
[21:10] <ds-work> anyone around a working v4lsrc right now?
[21:11] <LeRoutier> are you all using valgrind directly on the command line ? or are you using some front-end like alleyoop ?
[21:11] <walters> tromey: monotone is cool.
[21:11] <walters> tromey: i just don't like the idea of sha1sums as a UI
[21:11] <walters> tromey: patch-1, patch-2 etc is much nicer.
[21:12] <Company> LeRoutier: i'm using it from the commandline
[21:13] <Company> LeRoutier: but i'm a command line junkie anyway
[21:13] <taaz> so what's the plan for apps?  are they always goingn to be based on leading edge api?
[21:13] <walters> taaz: btw, have you seen my arch presentation?
[21:13] <taaz> walters: uri?
[21:13] <walters> taaz: http://web.verbum.org/tla/grokking-arch/grokking-arch.html
[21:13] <LeRoutier> if you want to try alleyoop, it's there : http://alleyoop.sourceforge.net/
[21:14] <taaz> specifically, do i need to start providing versioned apps?
[21:14] <taaz> debian packaged apps that is
[21:14] Action: wheels dreads the day when he has to have 30 differen version control software packages on his system...
[21:15] <tromey> walters: yeah.  I find I don't mind them too much
[21:15] <taaz> like the 0.8.x gst-editor isn't going to work with 0.6.x.  some people might want to use 0.6.x.  this is, imho, a mess
[21:16] <walters> wheels: i doubt it'll ever approach anywhere near 30
[21:17] <wheels> I guess in a couple of years people will stop writing mp3 taggers and IRC clients and start writing version control software.  :-)
[21:18] <walters> heh
[21:18] <walters> revctl is a sufficiently hard problem that i doubt we'll see that many
[21:19] <wheels> walters: Since when has not understanding the problem stopped OSS developers?  :-)
[21:19] <tromey> well, many more than we already have :-)
[21:19] <walters> heh
[21:21] <ds-work> BBB: why does v4lsrc run v4l-conf?
[21:22] trow ([email protected]) joined #gstreamer.
[21:25] <taaz> walters: nice slides.
[21:26] <tromey> yeah, I'm looking at them too
[21:30] <walters> taaz: thanks.  did you find them helpful?
[21:30] <walters> taaz: anything that could use improvement?
[21:31] <walters> taaz: i am hoping to give this presentation here soon.
[21:32] <wheels> walters: "here" == "work"?
[21:32] <walters> wheels: yeah
[21:33] Action: wheels uses perforce at work -- I've never loved CVS so much...  :-)
[21:33] <walters> haha
[21:35] <thomasvs> ds-work: I wouldn't give it to gnome.org, no - but don't know who
[21:35] <thomasvs> maybe fdo
[21:35] <thomasvs> BBB: what do you mean ?
[21:36] <ds-work> freedesktop.org isn't a legal entity
[21:36] <BBB> thomasvs: we're getting loads of questions from gentoo users that didn't run gst-register
[21:37] <BBB> I'd like to refer them to the FAQ instead of just saying that they should go to gentoo bugzilla
[21:37] <BBB> ds-work: for overlay... I guess for webcams it shouldn't do that
[21:37] Action: walters now has a check in Rhythmbox to see whether a thread element can be created, and if not flames the user
[21:37] <BBB> ds-work: but it's allowed to fail, IIRC
[21:37] <LeRoutier> that's odd, in post-installation of every gstreamer gentoo ebuild, it calls gst-register
[21:37] <BBB> ds-work: make a bug if it annoys you
[21:37] <BBB> ds-work: problem is that v4l-conf needs to be run with SUID root
[21:37] <BBB> ds-work: so we cannot do that in the plugin itselof
[21:38] <ds-work> BBB: ah...
[21:38] <LeRoutier> BBB: but what provides this v4l-conf ?
[21:40] jdahlin ([email protected]) left irc: Read error: 110 (Connection timed out)
[21:41] <thomasvs> ds-work: and the fsf ?
[21:46] <ds-work> perhaps
[21:46] <ds-work> or SPI or KDE
[21:47] <thomasvs> ds-work: what's the minimum legal entity that can own domains ?
[21:47] <thomasvs> is there some sort of nonprofit we can do ourselves or something ?
[21:47] Action: Zeenix has finally convinced his boss to use linux/ARM in our embedded systems
[21:47] <ds-work> we could start a non-profit, but that seems like a lot of work
[21:48] <ds-work> I'd much prefer to latch on to whatever other non-profit
[21:48] <thomasvs> xiph ? :)
[21:49] <ds-work> heh, maybe
[21:53] coqRe ([email protected]) joined #gstreamer.
[21:57] <ds-work> thomasvs: what's wrong with the GNOME Foundation?
[22:01] <markey> couldn't gst-register get run automagically from make install?
[22:01] <markey> might make things a bit easier
[22:02] Action: Zeenix begs thomasvs to fix the broken-links to video-wall..
[22:02] coqRe ([email protected]) left #gstreamer ("Leaving").
[22:05] <Zeenix> thomasvs: i was to apply for linuxfund. if these links are fixed then my chances of getting funds are higher
[22:05] <Zeenix> thomasvs: if i get these funds, i'll be able to work for gst/gst-plugins again
[22:06] <thomasvs> markey: it's an admin thing to run - ie, packagers, root, or the person reading README
[22:07] <Company> markey: question is what things would get easier - certainly not hacking oin the core ;)
[22:07] seb128 ([email protected]) left #gstreamer ("I like core dumps").
[22:07] <markey> for the user it would be easier
[22:07] <markey> I'd consider it
[22:07] <markey> especially the need for running it as user _and_ root is a bit weird
[22:08] <thomasvs> markey: you don't, just root is fine.
[22:08] <thomasvs> Zeenix: argh, ok, will do it tomorrow
[22:08] <Zeenix> thomasvs: thats my man.
[22:08] <markey> well, opinions seem to differ ;)
[22:08] <markey> some say so, others so
[22:12] <Company> markey: does make install run ldconfig?
[22:12] <markey> no clue
[22:13] <wheels> (It's actually not needed in KDE since we put hints in our binaries about the lib paths.)
[22:19] jdahlin ([email protected]) joined #gstreamer.
[22:19] Action: Company kicks Debian
[22:19] <Company> why are glibc manpages not installed by glibc-dev or glibc-doc but manpages-dev?
[22:23] <BBB> LeRoutier: overlay
[22:28] <ds-work> Company: because they're not part of glibc
[22:33] Action: Company kick glibc
[22:33] <Company> ds-work: uh, i just remember another question about Debian
[22:33] <Company> ds-work: if I install a -dbg package, where do I need to put the source so gdb finds it?
[22:36] <taaz> i LD_PRELOAD the libs i need directly
[22:38] <Company> taaz: no, i meant inside gdb
[22:38] <taaz> oh source... no clue ;)
[22:38] <taaz> i think it's just worked for me... haven't done it in a while though
[22:39] <LeRoutier> anyone know if fd.o CVS server was updated lately ?
[22:43] <ds-work> Company: don't know
[22:47] patrick76 ([email protected]) joined #gstreamer.
[22:48] <patrick76> hi there, anyone got some experience with gstreamer & ruby?
[22:51] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[22:51] <Company> hooray
[22:51] <Company> ds-work: what's the difference between ANONYMOUS mmap and malloc?
[22:52] <BBB> mmap is kernel memory and can thus be particular hardware (like PCI bus memory)
[22:52] <BBB> malloc is always virtual memory
[22:53] <BBB> I guess mmap in your case is RAM
[22:53] <BBB> malloc is VT (and can thus be swap)
[22:53] <Company> mmap can not be swapped?
[22:54] <walters> glibc malloc uses anonymous mmaps IIRC
[22:55] <patrick76> i want to create a filesrc >> decoder >> audiosink pipeline, i want the decoder to be anyone that can decode audio/x-mp3 to audio/raw, not a special one like 'mad', how?
[22:57] <Company> patrick76: short solution: use an autoplugger like spider
[22:58] <Company> patrick76: long solution get all element factories and search for an element factory that has the sinkcaps that you want and the srccaps that you want
[22:59] <ds-work> Company: anonymous mmap is lower level than malloc
[23:02] kwm ([email protected]) left irc: "Leaving"
[23:03] <patrick76> Company: ahhh, thanks a lot !
[23:04] markey ([email protected]) left irc: "g'nite"
[23:06] <ds-work> this sounds like it's straight out of the onion: "Girl, 15, charged with child porn after posting her sex pics online"
[23:10] bilboed ([email protected]) joined #gstreamer.
[23:11] <Company> ds-work: wouldn't be surprised if that was illegal...
[23:14] <Company> ds-work: you know that problem: get sex pics of your girlfriend when both of you are 15, get busted for possessing child porn 10 years later
[23:16] <LeRoutier> unless there is a problem, you prefer to have photos of her in her twenties than when she was 15
[23:17] <Company> LeRoutier: you might have forgotten you had the pictures
[23:17] Rounin ([email protected]) joined #gstreamer.
[23:18] <Rounin> Hey... I'm getting a bizarre error
[23:18] <Rounin> (smmx:3976): GStreamer-CRITICAL **: file gstelement.c: line 1533 (gst_element_link_many): assertion `element_1 != NULL && element_2 != NULL' failed
[23:18] <Rounin> Nothing is NULL... I've checked everything
[23:19] <spyder482> hmm
[23:19] <spyder482> Nothing IS NULL ;)
[23:20] <Rounin> You're right... It all makes sense!
[23:20] <Rounin> Too bad my program won't work.. But you're right, nothing is null
[23:20] <spyder482> hmm
[23:21] <BBB> Company: it can be
[23:21] <spyder482> can you paste the code snip again?
[23:21] sack ([email protected]) joined #gstreamer.
[23:21] <Rounin> Shoulda used more example code
[23:21] <BBB> Company: depends on what fd you mmap from
[23:21] <Rounin> Sure, hang on
[23:21] <sack> Hi everyone .
[23:21] <Company> BBB: ANONYMOUS
[23:21] <BBB> Company: kernelspace can always share vt pages
[23:21] <Rounin> myPipeline = gst_pipeline_new ("pipeline");
[23:21] <jmmv> hmm could anybody be so kind to look at the last mail I posted to the devel list, please?
[23:21] <Rounin> myFilesrc = gst_element_factory_make ("filesrc", "disk_source");
[23:21] <Rounin> [GTK stuff here]
[23:21] <BBB> er
[23:22] <BBB> vm pages
[23:22] <sack> anyone knows if gstreamer works on ARM arch or there's any future plans to make it work ?
[23:22] <Rounin> g_object_set (G_OBJECT (myFilesrc), "location", gtk_file_chooser_get_filename(
[23:22] <Rounin> 		GTK_FILE_CHOOSER(myFileDialog)), NULL);
[23:22] <BBB> sack: some elements don't work because of alignment issues, we're awaiting a patch
[23:22] <Rounin> That whole last bit is equivalent to the file name... Believe me, I checked
[23:22] dolphy ([email protected]) joined #gstreamer.
[23:22] <Rounin> myDecoder = gst_element_factory_make ("mad", "decoder");
[23:22] <Rounin> 	myAudiosink = gst_element_factory_make ("alsasink", "play_audio");
[23:22] <Rounin> 	gst_bin_add_many (GST_BIN (myPipeline), myFilesrc, myDecoder, myAudiosink, NULL);
[23:22] <Rounin> gst_element_link_many (myFilesrc, myDecoder, myAudiosink, NULL); <-- Fails here
[23:22] <sack> BBB : ok , thanks Ronnald :-P
[23:22] <BBB> Company: anonymous is a *flag*, right? I mean, don't yuo still need an fd to mmap from?
[23:22] Action: ds-work sees dolphy and remembers to file a bug on GstPlay
[23:23] <Company> Rounin: are you sure myDecoder isn't NULL?
[23:23] <ds-work> BBB: you can map fd -1
[23:23] <BBB> ah
[23:23] <Company> Rounin: which might happen when mad isn't found
[23:23] <BBB> I didn't know
[23:23] <Rounin> Company: I did a test
[23:23] <Rounin> if(myDecoder==NULL)gtk_statusbar_push(GTK_STATUSBAR(myStatusBar),
[23:23] <Rounin> 	gtk_statusbar_get_context_id(GTK_STATUSBAR(myStatusBar),"lol"),"Decoder");
[23:23] <BBB> that's probably just a vm kernel mmap then
[23:23] <BBB> that can be swap
[23:23] <BBB> just like malloc
[23:23] <Rounin> If myDecoder is NULL, it prints "Decoder" in the status bar
[23:23] <ds-work> BBB: it's a somewhat silly extension to mmap
[23:23] <Rounin> And it didn't
[23:23] <BBB> ds-work: it doesn't sound very useful...
[23:23] <Rounin> I guess I could just see if it finds mad though
[23:24] <Rounin> That would save us all a lot of work
[23:24] <ds-work> BBB: glibc's malloc implementation uses it extensively
[23:26] <Rounin> ERROR: pipeline could not be constructed: No such element mad
[23:26] <Rounin> My bad =P
[23:26] <Rounin> But it isn't NULL for some reason
[23:28] <Company> Rounin: maybe the statusbar isn't updated?
[23:29] <Company> Rounin: screen updates in gtk run in an idle handler
[23:29] <Rounin> Odd...
[23:29] <Rounin> Can I just install mad now though, or do I need to recompile gstreamer in its entirety?
[23:30] <Company> you need the plugin
[23:30] <Company> if you installed from source you need to rebuild gst-plugins
[23:30] <Company> or at least the part that installs the mad plugin
[23:30] <Rounin> Oh, ok... Thanks
[23:33] <Rounin> configure: These plugins will not be built: <name of pretty much all plugins here>
[23:33] <Rounin> I must have done something wrong some place
[23:34] <Rounin> Is there a --noimeanitbuildtheplugins command?
[23:34] <Rounin> Or is this just Gentoo mindfucking me again
[23:34] <Rounin> It likes to disale stuff that is totally elementary
[23:35] <ds-work> btw, you need libmad, libid3, and libid3tag to build the mad plugin
[23:35] <Rounin> Yeah, I see
[23:35] <Rounin> Thanks
[23:36] <LeRoutier> Rounin, are you using gstreamer/gst-plugins/gst-plugins-mad  official ebuilds ?
[23:37] <Rounin> Yes... And everything's disabled... 
[23:37] <LeRoutier> Company, on gentoo, mad plugin is a separate package (gst-plugins-mad). so he has to do : emerge gst-plugins-mda
[23:37] <LeRoutier> Rounin, it's perflectly normal
[23:37] <Rounin> Not any more apparently
[23:38] <Rounin> Anyway, I've specified "mad" in my USE flags
[23:38] bilboed_ ([email protected]) joined #gstreamer.
[23:38] <Rounin> But it's disabled flac, vorbis, xmms, well, it's disabled pretty much everything
[23:38] <Rounin> And there aren't enough flags to turn everything on
[23:38] Action: Company blames Gentoo
[23:38] <LeRoutier> Rounin, there is no such flag
[23:38] <BBB> Rounin: don't do gentoo bugs here
[23:38] Action: LeRoutier blames Rounin
[23:38] <BBB> Rounin: please mark them on gentoo bugzilla
[23:39] <BBB> foser: da?
[23:39] <LeRoutier> Rounin,  : each plugin is in a separate package/ebuild in gentoo
[23:39] <foser> BBB: you in luck
[23:39] <Company> BBB: why shouldn't they discuss gentoo issues here?
[23:39] <BBB> foser: can you add a big flashing annoying and noticeable banner text after the gstreamer ebuild to not report bugs upstream unless a gentoo gstreamer ebuild maintainer confirms it?
[23:40] <Rounin> LeRoutier: Oh I'm sorry... The dependency mechanism made it seem like gst-plugins was the only package
[23:40] <foser> BBB: well i think i did, but hmm some ppl won't listen
[23:40] <Rounin> Guess I'll just try emerging them all then
[23:40] <BBB> Company: because I'm slightly annoyed by the amount of gentoo bugs ("gentoo" bugs, so gentoo system bugs ionstead of our software bugs) in gnome-media bugzilla
[23:40] <BBB> foser: oh, hm...
[23:40] <foser> BBB: i'm really sorry.. but we got always say 'gentoo bugzilla first'
[23:40] <LeRoutier> Rounin : gst-plugins-ffmpeg, gst-plugins-oss, gst-plugins-*
[23:40] <foser> -got
[23:40] <Rounin> Where's gst-plugins-all when you need it
[23:40] <foser> Rounin: you don't need it
[23:40] <BBB> foser: well, if the banner is there then there's nothing more I can ask :/
[23:40] <jmmv> you should discuss netbsd bugs, not gentoo! ;)
[23:41] <foser> BBB: i'll update the gentoo gnome page one of these days..
[23:41] <LeRoutier> BBB: it is there, like the gst-register that is automatically launched at post-install
[23:41] <foser> BBB: it used to be on the gstreamer release page as well, but it got removed i noticed not so long ago
[23:41] <BBB> LeRoutier: that's really weird then, but I see loads of bugs that are caused by people not having run gst-register, it seems...
[23:41] <BBB> foser: oh... hmm, I'll re-add it tonight or so
[23:42] <BBB> foser: feel free to make bugzilla entries for each such things 
[23:42] <BBB> saves me a few other bugzilla entries ;)
[23:42] <Company> BBB: yeah, they shouldn't report gentoo bugs in our bugzilla, but they can discuss in this irc channel
[23:42] <foser> BBB: i wasn't sure i was looking at the right page ;) so changes there
[23:43] <foser> Company: its not a bug, it's uninformedness at best.. well .. ppl who can't find this out.. ghmm well .. should they be using gentoo 
[23:43] <Rounin> Why should people run gst-register
[23:43] <Company> yeah, i know
[23:43] <Rounin> You seem to think all informed users are telepaths
[23:43] <BBB> Company: don't know... maybe... ask me again tomorrow, I'm annoyed today ;)
[23:44] <BBB> Rounin: there's a bugzilla entry to add it to the FAQ :p
[23:44] <Company> BBB: :)
[23:44] <BBB> I'll do that tonight, too
[23:44] <Rounin> Hm
[23:44] <LeRoutier> Rounin, after each plugin install, to make gstreamer aware that there are new ones available
[23:44] <foser> Rounin: _any_ gentoo ebuild plugin runs gst-register...
[23:44] <Rounin> Yeah... It would be a good idea to do this automatically post-install on any OS
[23:44] <foser> Rounin: if it doesn't thats a bug, but i really don't believe you if you say that..
[23:45] <foser> adding an ebuild for a plugin not available yet is a pita, isn't it LeRoutier ?
[23:46] <foser> and everything should work smooth for you
[23:47] <LeRoutier> foser, well, i don't use the official ebuilds for all gstreamer related packages but cvs ebuilds (that i try to keep for myself)
[23:48] <Rounin> Hm... I think it would be easier if it just installed all the plugins though
[23:48] <LeRoutier> but as i remember the gst-plugins.eclass, and gst-plugins-*.ebuild content, that doesn't sound too hard
[23:48] <Rounin> Who needs hard drive space anyway
[23:48] <LeRoutier> Rounin, no, too much deps
[23:48] <Rounin> LeRoutier, but the user could disable stuff with USE flags anyway
[23:48] <Rounin> -mad -oggvorbis -xmms etc.
[23:48] <LeRoutier> that would be horrible, too much use flags after that
[23:49] <Rounin> It's even more horrible having to resolve dependencies in an application at runtime though
[23:49] <Rounin> "Sorry, the program's running, but you need to download an integral part. Goodbye."
[23:49] <LeRoutier> why looking at run-time ?
[23:50] <Rounin> Well, let's say I want to play an xvid file, and I don't have the xvid plugin
[23:50] Shoragan ([email protected]) left irc: "Leaving"
[23:50] <Rounin> ...It won't play
[23:50] <LeRoutier> yep, normal
[23:50] <Rounin> Well, it's normal having the entire computer crash too, but it's not good
[23:50] <Rounin> Guess I should switch to Debian
[23:51] <foser> Rounin: too much deps and it introduces a lot of instability in my opinion
[23:51] <LeRoutier> Rounin, as you wish, what use do yoiu have of gstreamer ? end user or are you developping something around ?
[23:52] <Rounin> Something in between end-user and developer? =P
[23:52] <foser> LeRoutier: actually i don't think cvs ebuilds are all that useful, especially when you are active in development
[23:52] <Rounin> More like I decided to try it out
[23:52] <foser> Rounin: we provide all necessary deps as needed by the packages, no user should in principle need to install anything extra to get the application working at a decent level
[23:53] <Rounin> True
[23:53] <Rounin> Well, thanks for the info
[23:53] <LeRoutier> foser, i won't say i'm an active dev. i only use it to test small patches before sending them to bugzilla.
[23:55] bilboed ([email protected]) left irc: Read error: 110 (Connection timed out)
[23:55] <LeRoutier> Rounin, i was unhappy with the gst-plugins-xxx ebuilds at first, then i understood how crappy it would be on a dependency level if it was an all-in-one ebuild
[23:56] <BBB> bbl
[23:56] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[23:59] Action: ds-work notices that SuSv3 is now online for free
[23:59] <Company> if we follow wheels' arguments, GStreamer core now uses KDE technologies
[00:00] --- Tue Apr  6 2004
[00:02] <LeRoutier> kiosrc now working ?
[00:03] <jdahlin> Company: yes, please tell all KDE developers
[00:04] <Company> LeRoutier: valgrind support added
[00:04] <iain>  /. /. /.
[00:04] <wheels> Company: Actually -- to be pedantic -- I never said that valgrind was a KDE thing, but just that the most hard-core valgrind fans are KDE people.  :-)
[00:05] <LeRoutier> Company, that would be good, several people reported importants mem leaks in gstreamer
[00:06] <wheels> Company: what was the issue with GstBuffer?  (Since I was seeing some problems with it in my valgrinding of kiosrc.)
[00:06] Nick change: apoc -> apoc-zZz
[00:08] <Rounin> LeRoutier: I think it's crappy either way... But gstreamer in itself is great, so I don't mind to be honest
[00:08] <Company> wheels: GstBuffers are mempooled, so valgrind doesn't find leaks there
[00:08] <Rounin> I mean, wow... Systems should come with this preinstalled =P
[00:09] <wheels> Company: Ok, I was getting some invalid reads from inside of memory allocated using GstBuffer, so I guess that wouldn't be related?
[00:09] <wheels> Company: Hmm, that came out wrong...
[00:09] Action: wheels looks at the code again before finishing this thought...
[00:10] <wheels> Ah, I was sort of right -- allocated with gst_buffer_new_and_alloc()...
[00:11] <Company> i should have attached my cool little test program
[00:12] <Company> so people can see for themselves
[00:13] <iain> Company: did you ever get marlin compiled?
[00:13] <Company> iain: i committed a fix yesterday
[00:13] <iain> cool
[00:14] <iain> sweet, marlin has a gaelic translation
[00:15] Action: ds-work grumbles
[00:15] <ds-work> apparently, loading plugins to resolve currently unresolved symbols is not portable
[00:16] <Company> not portable? or not portable to NetBSD?
[00:17] <ds-work> portable as in "POSIX doesn't require it"
[00:17] <ds-work> basically, our problem is this:
[00:18] <jmmv> no... I've found the real problem here
[00:18] <ds-work> app loads plugin1
[00:18] <ds-work> plugin1 loads plugin2
[00:18] <jmmv> which I'll confirm when the build finishes ;)
[00:19] <jmmv> it's all glib2's fault, which doesn't detect RTLD_GLOBAL properly at configuration time, so it will always enforce local bindings when g_module_open is called
[00:19] <ds-work> at this point, plugin2's symbols can be accessed via dlsym(), but undefined references in plugin1'd code are not required to be resolved by plugin2's symbols
[00:20] <ds-work> jmmv: that could be, too.  are you using glib-2.6?
[00:20] <jmmv> 2.4, you mean?
[00:20] <ds-work> er, yeah
[00:20] <jmmv> yes
[00:20] <ds-work> nevertheless, under a strict reading of POSIX, we shouldn't rely on such behavior
[00:20] <jmmv> (anyway, thre will be problems when a platform doesn't provide RTLD_GLOBAL)
[00:21] <Company> ds-work: if that were the case, i'd just kick the platform that doesn't support this. Oh and someone should file a bug against posix so this is fixed whenever they update the spec ;)
[00:21] Action: Company doesn't care about the TOS Posix layer very much ;)
[00:22] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[00:24] <jmmv> that was it :_)
[00:24] <Company> i don't care very much about POSIX at all, because they want money to support them...
[00:24] <Company> s/support them/get access to the spec/
[00:24] <ds-work> Company: it's online now
[00:24] <LeRoutier> Company, no more the case
[00:24] <ds-work> http://www.opengroup.org/onlinepubs/007904975/mindex.html
[00:25] <LeRoutier> they also now allow the specs text to be included in man pages
[00:27] <Company> bookmark added
[00:28] <Company> nice, kcachegrind is in debian-ppc but valkgrind isn't ;)
[00:28] Action: Company notices, that the correct name would be valkrind anyway<
[00:28] 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 
[00:31] <ds-work> valgrind doesn't work on powerpc yet, uthc
[00:31] <Company> yeah, but kcachegrind would be pretty useless
[00:31] <Company> with out it
[00:33] Action: Company should try the experimental valgrind support on ppc some day
[00:34] <Company> though gst is pretty much the last app that'd work with that anyway
[00:34] Action: ds-work wishes someone would solve the generic run-time assembly/disassembly/emulation/compilation problem some day
[00:35] <Company> what problem?
[00:36] <ds-work> having a set of libraries that understands binary code and read/write/execute it generically
[00:36] <ds-work> right now, portions of all of it are in various applications
[00:38] <Company> you mean a generic debugging framework for binaries?
[00:39] <Company> gdb and valgrind with a cool UI (or possibly 3: terminal, gtk and qt)
[00:40] <ds-work> but also assembly capabilities
[00:40] <ds-work> and JIT code rewriting for other processors
[00:41] <ds-work> since they're mostly very related problems
[00:41] pb_ (~pb@2002:5160:4550:1:240:95ff:fe30:aa84) left irc: "Client exiting"
[00:42] <Company> yeah
[00:42] <Company> that will finally allow patching IE without having the source for it :)
[00:48] <jdahlin> heh
[00:49] <jmmv> night all
[00:49] <wheels> Company: Emacs!
[00:50] jmmv ([email protected]) left irc: "Leaving"
[00:50] <Company> wheels: huh?
[00:50] <wheels> Company: It has gdb and valgrind modes...
[00:50] <Company> wheels: i want gdb and valgrind combined
[00:50] <wheels> (Not that I actually use them, but it's just a reflex to plug emacs.)
[00:51] <Company> wheels: like valgrind break lost memory
[00:51] <wheels> Actually my big wish is reasonable scritability for gdb...
[00:51] <Company> yeah, that would be cool, too
[00:52] <jdahlin> wheels: there are some scriptablity for gdb
[00:52] <Company> and adding breakpoints to functions in plugins ;)
[00:52] <jdahlin> s/are/is/
[00:52] <jdahlin> but it's not especially good
[00:52] Action: wheels wants a Perl module.  :-)
[00:52] <jdahlin> http://cvs.sourceforge.net/viewcvs.py/*checkout*/python/python/dist/src/Misc/gdbinit?content-type=text%2Fplain&rev=1.6
[00:52] <jdahlin> the python guys has some wicked stuff
[00:52] <jdahlin> in a custom .gdbinit
[00:53] <jdahlin> for printing local variables and stack frames...
[00:55] <Company> i'd also like gdb to save state
[00:55] <Company> so when i exit gdb and restart it, it should have the same program with the same breakpoints loaded
[00:56] <jdahlin> you can have a macro that sets up breakpoints
[00:56] teuf ([email protected]) left irc: "Client exiting"
[00:56] <jdahlin> starts gdb. sets variables, stops it, modifes blah blah etc
[00:56] Action: ds-work wants a combo of gdb and valgrind that keeps track of the last 100 or so functions executed
[00:56] <ds-work> hmmm...
[00:56] <Company> yeah, options to what should be tracked
[00:57] <ds-work> anyone mind if I wander off for a few months to write that?
[00:57] <Company> so you can stop for common problems
[00:57] trow ([email protected]) left irc: "Leaving"
[00:58] <Company> ds-work: after gst 1.0, ok?
[00:59] <LeRoutier> where to send patches for marlin ? gnome bugzilla or sf.net ?
[01:04] <ds-work> bugzilla
[01:04] <LeRoutier> ok, doing it now
[01:05] <LeRoutier> trivial bug as usual, nothing big
[01:05] <Company> fasted is to poke iain probably
[01:06] <iain> send to me
[01:07] <LeRoutier> sorry, already created
[01:08] <iain> oh, no worries
[01:08] <LeRoutier> http://bugzilla.gnome.org/show_bug.cgi?id=139198
[01:08] <iain> bugzilla.gnome.org will work fine too
[01:08] <LeRoutier> adding a .desktop.in file :)
[01:08] <LeRoutier> with require Makefile changes
[01:09] <iain> thanks
[01:09] <LeRoutier> tested locally, works well. pot file would have to be updated of course to add the 2 news strings for i18n
[01:16] <Company> good, markey is gone
[01:16] Action: Company can now start ranting about amarok screenshots ;)
[01:18] <ds-work> you mean how the interface is insane?
[01:19] Action: ds-work wonders if he's a usability snob yet
[01:23] <Rounin> Hm, anyone know of an easy way to place a widget (button etc.) on top of another widget (a bg image)?
[01:23] <Rounin> GTK+ 2.4
[01:23] <Company> nope, that's not easy
[01:23] Action: Company still wants to write an overlay image that allows putting an image on top of another
[01:24] <Company> ds-work: yeah, the interface is insane
[01:24] <Rounin> Oh great
[01:24] <Company> ds-work: the most important parts get the smallest font
[01:26] jdahlin ([email protected]) left irc: Read error: 60 (Operation timed out)
[01:27] <ds-work> Company: that pretty much follows the "design for power users" theory
[01:28] <Company> i'll keep using apps with a proper UI
[01:29] <Company> oh, swfdec is an evil "print to console" lib
[01:31] jdahlin ([email protected]) joined #gstreamer.
[01:31] Action: Company thinks GType should be a typedef to gchar *
[01:32] <ds-work> Company: only because you never split out gst's debugging :)
[01:32] <ds-work> Company: also, by default swfdec only prints if there is an error
[01:32] <Rounin> GtkFixed <3
[01:33] <ds-work> Company: technically, any time an error is currently printed, you should file a bug
[01:33] <Company> ds-work: there's a lot of DEBUG: [...]: bla
[01:33] <Company> ds-work: i'm not sure what file it is, i just imported /some/dir into rb ;)
[01:34] <ds-work> Company: it's either an error or something I should look at.  unfortunately, I have 100's (1000's?) of swfs I already know I need to look at
[01:35] <walters> so did anyone recently fix memleaks?
[01:36] <ds-work> walters: we can now detect buffer leaks, so "soon"
[01:36] <Company> walters: there are so many, we always fix some ;)
[01:36] <walters> heh
[01:36] <Company> and i fixed one in esdmon today
[01:38] <ds-work> kiosrc.cpp: In function `gboolean plugin_init(GstPlugin*)':
[01:38] <ds-work> kiosrc.cpp:558: warning: `init' is deprecated (declared at 
[01:38] <ds-work>    /usr/include/kde/kcmdlineargs.h:258)
[01:38] <Company> ds-work: known issue
[01:38] <Company> ds-work: quick fix: add another string in there (breaks kde 3.1 though)
[01:38] <Company> ds-work: not so quick fix: poke wheels some more (i already started)
[01:39] <Company> ds-work: longest fix: figure out how to disable that
[01:40] <wheels> Company: Well, I already commited something to kdelibs to disable that, but that doesn't do us any good at the moment.  I'll probably do the ifdef based on the version for now...
[01:41] <Company> wheels: go ahead - problem is gst-plugins doesn't compile right now with KDE 3.2
[01:41] <wheels> Company: I'm still messing with my bass.  :-)
[01:41] <wheels> Company: But I suppose I can take a short break.  :-)
[01:42] <Company> wheels: great
[01:43] <wheels> ds-work: for background -- the issue is that it originally used something that was added in 3.2 (and as such didn't compile with KDE 3.1 or lower), but in 3.2 the old one was marked as deprecated -- which generates a warning with GCC > 3.
[01:44] Action: ds-work notes one could use '#include "kdemacros.h" "#undef KDE_DEPRECATED" "#define KDE_DEPRECATED"
[01:44] <wheels> Company: And it actually does build for me since I use 2.95.  :-)
[01:44] Action: walters notes that loading a library in valgrind in rb over nfs is slwo
[01:44] <walters> slow even.
[01:45] <wheels> ds-work: wouldn't help -- that's a compile time thing...
[01:45] Misirlou ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[01:45] <wheels> ds-work: But today I added an #ifndef KDE_DEPRECATED wrapper in kdemacros.h (in HEAD and BRANCH) so that it can be defined externally...
[01:45] <ds-work> as it's a compile time error, I think compile time solutions are relevant
[01:47] <wheels> ds-work: Well, I don't think what you're suggesting will work -- I was saying that it is compile time.  Here goes.  :-)
[01:47] <wheels> Hmm, actually I can't test this here...
[01:47] <wheels> ds-work: can you try it?
[01:48] <Company> wheels: define KDE_DEPRECATED to "psidjghsoidghsodighsoig" in kdemacros.h and see if it works
[01:48] <wheels> Company: true
[01:50] <ds-work> yes, it works
[01:50] <wheels> Yep, just saw that here too.
[01:50] <wheels> ds-work: care to commit?
[01:50] <ds-work> ok
[01:51] <ds-work> it's a nasty hack, one should check a kde version define or something
[01:51] Action: wheels notes that the C++ guys are always a little slow on macro magic.  :-)
[01:51] <wheels> ds-work: Well, the "new" version will just be "#define KDE_DEPRECATED"
[01:52] <iain> http://mattcamp.paunix.org/usound/
[01:52] <iain> uh oh
[01:53] sub_pop ([email protected]) left irc: "Client exiting"
[01:53] <wheels> iain: I'm sure it's set to become the official mediao framework of FreedomBox.  :-)
[01:54] <ds-work> wheels: checked in
[01:54] <wheels> ds-work: thanks
[01:54] sub_pop ([email protected]) joined #gstreamer.
[01:54] <iain> wheels: yeah :)
[01:55] jdahlin ([email protected]) left irc: "Leaving"
[01:55] <ds-work> iain: great, another sound daemon that doesn't support non-standard hardware
[01:57] <iain> I don't quite understand what freedombox network is
[01:58] <wheels> iain: It's a global network of both FreedomBox users.
[01:58] foser ([email protected]) left irc: "[ I want to believe ]"
[01:59] <iain> is that it?
[01:59] <wheels> Well, since I just made that up I wouldn't count it as authoritative.  :-)
[02:00] <iain> okay
[02:00] <walters> garr.
[02:01] Action: walters gets annoyed at waiting 10 minutes for valgrind'd rb to load his library only to lose it to a valgrind segfault
[02:04] Rotty ([email protected]) left irc: Read error: 60 (Operation timed out)
[02:04] walters ([email protected]) left irc: "out"
[02:07] <Company> http://www.freedombox.info/quotes.html
[02:07] <LeRoutier> "30-day demo" ?
[02:07] <LeRoutier> hum, commercial product so
[02:08] <LeRoutier> seems there is some voice recognition controls
[02:10] Misirlou ([email protected]) joined #gstreamer.
[02:11] iain ([email protected]) left irc: "!"
[02:15] smoke ([email protected]) got netsplit.
[02:20] <LeRoutier> hadess wrote about usound in his blog
[02:21] smoke ([email protected]) got lost in the net-split.
[02:21] <LeRoutier> there : http://www.advogato.org/person/hadess/diary.html?start=356
[02:23] Action: wheels notes that a "box" is not a very good metaphor for freedom.
[02:38] <LeRoutier> good night ppl
[02:38] LeRoutier ([email protected]) left irc: "Leaving"


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