IRC Logs

*******************************************************************
[03:01] <BBB> bbl
[03:01] BBB ([email protected]) left irc: "Client exiting"
[03:03] <mathrick> night
[03:03] Nick change: mathrick -> mathrick|sleep
[03:04] foser ([email protected]) left irc: "[ I want to believe ]"
[03:09] Rotty ([email protected]) left irc: Remote closed the connection
[03:19] abo ([email protected]) left irc: Read error: 113 (No route to host)
[03:48] <LeRoutier> good night ppl
[03:48] LeRoutier ([email protected]) left irc: "Leaving"
[04:23] jcsston ([email protected]) joined #gstreamer.
[04:32] <NFusi0n> Is it just me or is alsasink sorta flaky?
[05:13] jcsston ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[05:14] thaytoo ([email protected]) got netsplit.
[05:14] thaytoo ([email protected]) returned to #gstreamer.
[05:26] <thaytoo> NFusi0n: it has trouble with a few sound cards that need fixing, but it's fine in general
[05:44] <NFusi0n> thaytoo: cs46xx, it skips a lot for me, and i have to specify device=hw:0 otherwise it will segfault gst-player
[05:44] <NFusi0n> oss plays smooth for me
[05:49] sub_pop ([email protected]) left irc: "Client exiting"
[05:55] sub_pop ([email protected]) joined #gstreamer.
[06:02] jcsston ([email protected]) joined #gstreamer.
[06:42] spyder482 ([email protected]) left irc: "Leaving"
[07:29] sxpert_work ([email protected]) left irc: Read error: 110 (Connection timed out)
[07:29] lilo ([email protected]) left irc: "brb"
[07:29] lilo ([email protected]) joined #gstreamer.
[07:31] jcsston` ([email protected]) joined #gstreamer.
[07:44] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[08:26] sxpert_work ([email protected]) joined #gstreamer.
[08:37] Nick change: mathrick|sleep -> mathrick
[08:37] <mathrick> re
[08:54] <ds> er
[08:55] <mathrick> ds: nah, no Clooney here
[08:56] <ds> hmmm, rhytmbox made MacWorld UK
[08:57] <mathrick> nice, so MacWorld covers non-mac world too?
[08:57] <ds> http://www.macworld.co.uk/news/main_news.cfm?NewsID=8209
[08:58] robla_ ([email protected]) joined #gstreamer.
[08:59] <mathrick> cool, pretty decent note
[09:02] Action: bitshifter is glad to find that the awkward phrase "Gnome Unix and Linux desktop suite and development platform" is unique to that article
[09:16] <mathrick> bitshifter: why? it's true, after all
[09:16] <mathrick> although maybe not the most obvious
[09:17] <bitshifter> mathrick: yeah, but it sounds like bureaucrat lingo, something a committee has come up with as a compromise at 5.30 in the morning
[09:20] <mathrick> bitshifter: not really, rather like it was taken from full name, with "for unix and linux" glued together
[09:21] Action: bitshifter quickly goes to get himself some coffee before someone has the opportunity to challenge him to come up with something catchier
[09:27] <mathrick> bitshifter: I was tempted to do so, but decided to spare you this time ;)
[09:33] thaytoo ([email protected]) left irc: "hometime"
[09:34] NFusi0n ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[09:41] wheels ([email protected]) left irc: "work"
[09:56] Rotty ([email protected]) joined #gstreamer.
[10:06] thomasvs ([email protected]) left irc: Client Quit
[10:16] thaytan ([email protected]) joined #gstreamer.
[10:23] sub_pop ([email protected]) left irc: "Client exiting"
[10:25] thomasvs ([email protected]) joined #gstreamer.
[10:29] dolphy ([email protected]) joined #gstreamer.
[10:31] kmaraas ([email protected]) left irc: Client Quit
[10:44] jdahlin ([email protected]) joined #gstreamer.
[10:45] Nick change: mathrick -> mathrick|Uni
[10:55] Nick change: jcsston` -> jcsston
[10:56] markey ([email protected]) joined #gstreamer.
[11:15] markey ([email protected]) left irc: "leaving"
[11:15] jdahlin ([email protected]) left irc: "Leaving"
[11:16] jdahlin ([email protected]) joined #gstreamer.
[11:16] jdahlin_ ([email protected]) joined #gstreamer.
[11:17] jdahlin ([email protected]) left irc: Client Quit
[11:18] Nick change: jdahlin_ -> jdahlin
[11:25] kmaraas ([email protected]) joined #gstreamer.
[11:26] abo ([email protected]) joined #gstreamer.
[11:27] abo ([email protected]) left #gstreamer ("Leaving").
[11:33] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) joined #gstreamer.
[11:49] Shoragan ([email protected]) joined #gstreamer.
[11:56] steve_b ([email protected]) left irc: Remote closed the connection
[11:58] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[12:02] sublett ([email protected]) joined #gstreamer.
[12:20] iain ([email protected]) joined #gstreamer.
[12:42] markey ([email protected]) joined #gstreamer.
[12:56] Nick change: markey -> markey|afk
[13:00] <jdahlin> ** ERROR **: file ../../../testsuite/refcounting/element_pad.c: line 45 (main): assertion failed: (GST_PAD_IS_LINKED (gst_element_get_pad (element2, "src")))
[13:00] <jdahlin> aborting...
[13:00] <jdahlin> XFAIL: element_pad
[13:01] <jdahlin> I wonder what broke that, probably Benjamins changes to GstBin
[13:01] <jdahlin> oh, the actual error is: FAIL: bin
[13:15] ensonic ([email protected]) joined #gstreamer.
[13:44] foser ([email protected]) joined #gstreamer.
[13:57] <ensonic> with current cvs I get in "gstreamer/gst" :  *** Keine Regel vorhanden, um das Target »gstthreaddummy.c«,
[13:59] Action: ensonic thinks the highest goel for a the release is to get gst running at all :-(
[13:59] <ensonic> goel=goal
[13:59] <thomasvs> ensonic: you sure you updated correctly ?
[14:00] <thomasvs> ensonic: and have you ran autoregen.sh ?
[14:19] Company ([email protected]) joined #gstreamer.
[14:20] <Company> good morning angels
[14:27] <ensonic> thomasvs, I am now using a SuseLinux9/i686 - because I don't want to be stuck due to Solaris ;-), it is a fresh cvs co; make clean; autoreconf; make all
[14:28] <ensonic> thomasvs, I try autoregen.sh
[14:28] <ensonic> imho, make should detect that it is neccesary to rerun any of the autotools
[14:31] <Company> make does detect that
[14:32] <Company> if you run with --enable-maintainer-mode, which autogen does automatically
[14:32] <Company> make does not detect that you changed CPUs to a different architecture though ;)
[14:32] <iain> Company: good news...the destruction of germany will not happen anymore
[14:33] <Company> iain: he suicided himself?
[14:34] <iain> Company: no, they're going to split up...
[14:34] <iain> Company: she is losing faith in long distance relationships
[14:35] <ensonic> Company, ;-)
[14:36] <Company> iain: so you're just happy that germany is very far away
[14:37] <iain> Company: well...I would be, but in the course of the conversation where I found that out she also told me "It would be much easier if you and I felt the same way, but we don't. I never wanted you to get hurt"
[14:37] <iain> Company: which given everything, I figured it translated into "I'm going to be single soon, but don't think you're going to get any"
[14:43] <Company> iain: oh, you're currently the "good friend"?
[14:45] <ensonic> thomasvs: you've been right, make didn't detected that some Makefiel has changed
[14:45] sublett ([email protected]) left irc: "Why are all the really good songs short?"
[14:49] Nick change: mathrick|Uni -> mathrick
[14:55] markey|afk ([email protected]) left irc: "bbl"
[15:05] jimmy_dean ([email protected]) left irc: Remote closed the connection
[15:05] <iain> Company: yeah
[15:06] <iain> Company: I went out with her Friday night, Saturday night, Sunday night and tomorrow night...and if thats only "good friend" I'm not sure I want to know what boyfriends get to do...
[15:11] kmaraas ([email protected]) left irc: Read error: 110 (Connection timed out)
[15:16] <Company> iain: tough - though i managed to get my girlfriend even though we were best friends
[15:39] <iain> Company: I always think thats the best way...I'll just have to make sure I dont freak out
[15:41] jonathang ([email protected]) joined #gstreamer.
[15:49] jonathang ([email protected]) left irc: "Leaving"
[15:53] bilboed ([email protected]) left irc: "Client exiting"
[15:54] md` ([email protected]) left irc: 
[15:55] ensonic ([email protected]) left irc: Read error: 113 (No route to host)
[16:04] md` ([email protected]) joined #gstreamer.
[16:17] jimmy_dean ([email protected]) joined #gstreamer.
[16:25] <Company> hm
[16:26] <Company> everyone should have written his own scheduler
[16:26] <jdahlin> is it that fun?
[16:28] jonathang ([email protected]) joined #gstreamer.
[16:28] <Company> no, but it teaches you a lot and allows you to easily do modifications
[16:30] <Company> it's not unfunny though
[17:00] pb_ (~pb@2002:3e03:42cc:1:20b:6aff:fe19:f290) left irc: "bbl"
[17:09] <Company> ds-work: pls ping me when you wake up
[17:09] <Company> ds-work: got a question about GST_CAPS_ANY
[17:10] sublett ([email protected]) joined #gstreamer.
[17:12] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[17:12] mxpxpod ([email protected]) joined #gstreamer.
[17:14] bilboed_school ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[17:15] Action: thomasvs wonders why he can't play theora video in the player
[17:15] bilboed_school ([email protected]) joined #gstreamer.
[17:17] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[17:20] <thomasvs> hm, it doesn't play anything at all.  ok, what's up
[17:20] LeRoutier ([email protected]) joined #gstreamer.
[17:20] <LeRoutier> hello
[17:23] ensonic ([email protected]) joined #gstreamer.
[17:29] trow ([email protected]) joined #gstreamer.
[17:39] <thomasvs> if a link is scheduled by calling _link, then later on when setting to PLAYING the link is found to not work, how should the app check for this ?
[17:42] <Company> you'll get an element error that tells you that nego failed
[17:44] <mathrick> bbl
[17:44] mathrick ([email protected]) left irc: "Client exiting"
[17:47] BBB ([email protected]) joined #gstreamer.
[17:47] Action: BBB waves
[17:47] <BBB> so it's 5:45 there in Spain?
[17:47] <BBB> (pm)
[17:48] Action: BBB pokes thomasvs
[17:48] <jdahlin> no, 5:42
[17:48] Action: jdahlin runs
[17:50] <BBB> is he in? :)
[17:50] <jdahlin> yes
[17:52] <thomasvs> BBB: shoot
[17:53] Action: BBB shoots a ffmpeg gun @ thomasvs
[17:53] Action: BBB is a true US {ju,ya}nkee
[17:54] <Company> hum, there's a bug somewhere
[17:54] Action: Company doesn't care atm
[17:55] <BBB> "somewhere"
[17:55] Action: BBB thinks company is so very right
[17:55] <BBB> you know what?
[17:55] <BBB> I think there's more than one bug somewhere!
[17:56] <Company> bah
[17:56] <BBB> couldn't resist ;)
[17:56] <Company> but i've fixed "allow pause to release device" with rb
[17:56] <Company> "fixed" would look better
[17:57] <thomasvs> BBB: as soon as I can finally play something again with gst, I'll look at ffmpeg
[17:57] <thomasvs> BBB: in the meantime, it'd be nice if you'd tell me what version exactly to update ffmpeg to
[17:58] <BBB> current CVS, as always
[17:59] <BBB> but I want to be able to update it myself ;)
[17:59] <BBB> so let's do a crashtest course for me, ok?
[17:59] <BBB> you surely have more important things to do then updating ffmpeg every time I find a bug fixed in their CVS
[17:59] <BBB> s/then/than/
[18:00] <ensonic> is 'mad' of current cvs working for anyone?
[18:01] <thomasvs> BBB: for the first time to figure out how to do it, I really do first need to figure it out myself
[18:01] <thomasvs> BBB: also, "current" is the wrong answer.  we need a known good working copy.
[18:01] <thomasvs> biab
[18:01] <ensonic> gst-launch-0.8 filesrc location=~/download/03-Drowning.mp3 ! spider ! esdsink : (process:16205): GStreamer-CRITICAL **: Factory for `mad' has no type
[18:01] <ensonic> Speicherzugriffsfehler
[18:01] thomasvs ([email protected]) left irc: "Client exiting"
[18:01] <Company> ensonic: gst-register helps
[18:01] <ensonic> Company: I've done that
[18:02] <Company> after getting that assertion?
[18:03] <ensonic> I have a suse9 with gst-0.6 here, I'vwe just checked out and installed current cvs of gstreamer and as root run gst-register-0.8, the I've executed the above pipeline and it crashes 
[18:04] <ensonic> Company: starnge is, when I run gst-editor I have all plugins twice, should I remove the registry manually and rebuild it?
[18:05] Action: Company thinks the current registry code sucks beyond repair
[18:05] <BBB> talking about that
[18:05] Action: BBB adds random stuff to his profiles bug
[18:05] Action: Company won't fix it though unless someone allows him to break ABI compat
[18:06] thomasvs ([email protected]) joined #gstreamer.
[18:06] <ensonic> gst-register says : rebuilding user_registry (/root/.gstreamer-0.8/registry.xml)
[18:06] <ensonic> how can I say it should rebuild the system wide registry?
[18:06] <BBB> run it as root and user
[18:06] <Company> it should rebuild the global registry, too
[18:06] <BBB> that mostly helps
[18:07] <ensonic> okay, that helps a bit, but now I get :
[18:07] <ensonic> > gst-launch-0.8 filesrc location=~/download/03-Drowning.mp3 ! spider ! esdsink RUNNING pipeline ...
[18:07] <ensonic> ERROR        GST_PADS(16256) gstpad.c(2394):gst_pad_set_explicit_caps:<mad0> (null)
[18:07] <ensonic> ERROR        GST_PADS(16256) gstpad.c(2394):gst_pad_set_explicit_caps:<mad0> failed to negotiate (try_set_caps returned REFUSED)ERROR: from element /pipeline0/spider0/mad0: Internal GStreamer error: pad problem.  File a bug.
[18:07] <ensonic> Additional debug info:
[18:07] <ensonic> gstpad.c(2394): gst_pad_set_explicit_caps: /pipeline0/spider0/mad0:
[18:07] <ensonic> failed to negotiate (try_set_caps returned REFUSED)
[18:07] <ensonic> ERROR       scheduler(16256) gstoptimalscheduler.c(2287):gst_opt_scheduler_iterate:<GstOptScheduler@0x8058f48> in error state
[18:07] <ensonic> Execution ended after 1 iterations (sum 34585000 ns, average 34585000 ns, min 34585000 ns, max 34585000 ns).
[18:08] <Company> that error handler should at least tell what caps it tries to set
[18:10] <BBB> esdsink is picky
[18:10] <BBB> add audioconvert/audioscale
[18:11] ChrisHJW ([email protected]) joined #gstreamer.
[18:13] <ensonic> BBB, thats why there is the spider inbetween
[18:13] <ensonic> BBB: gst-launch-0.8 filesrc location=~/download/03-Drowning.mp3 ! mad ! audioconvert ! esdsink
[18:13] <Company> spider can't handle audioscale afaik
[18:13] <ensonic> BBB: causes the same error
[18:13] kwm ([email protected]) joined #gstreamer.
[18:14] bilboed ([email protected]) joined #gstreamer.
[18:14] <ensonic> Company: that's it, so this works - gst-launch-0.8 filesrc location=~/download/03-Drowning.mp3 ! mad ! audioconvert ! esdsink
[18:15] <ensonic> thanks
[18:16] <BBB> Company: it can, it works for gst-player
[18:16] <BBB> I think
[18:17] <Company> BBB: i'm pretty sure it either inserts audioscale or audioconvert, no matter what the problem is
[18:20] Action: BBB doesn't know
[18:20] <BBB> I just know totem works for me ;)
[18:21] <Company> it works for me, too - but i don't use the gst backend :p
[18:23] Action: Company prefers gst-player
[18:24] <BBB> gmbl :)
[18:24] <BBB> ass
[18:24] Action: BBB tries to make totem rock
[18:24] Action: Company does the same for rhythmbox
[18:24] Action: BBB shakes hands and proposes a partnership where they try together to make both rock
[18:25] yippi ([email protected]) joined #gstreamer.
[18:26] lupusBE ([email protected]) joined #gstreamer.
[18:28] kmaraas ([email protected]) joined #gstreamer.
[18:36] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[18:36] robla_ ([email protected]) left irc: "Client exiting"
[18:37] foser ([email protected]) left irc: "[ I want to believe ]"
[18:40] jcsston ([email protected]) joined #gstreamer.
[18:41] <thomasvs> gah.  why won't my gstplay play stuff...
[18:41] Action: thomasvs adds debugging to player lib
[18:44] <Company> my gst-player plays stuff
[18:45] <ensonic> thomasvs: did you have time to look at the adder?, it still seems to be broken
[18:45] <ensonic> gst-launch-0.8 sinesrc freq=1000 ! adder ! audioconvert ! esdsink sinesrc freq=500 ! adder0.sink1
[18:49] <thomasvs> ensonic: no, I just got back
[18:58] Misirlou ([email protected]) left irc: Read error: 60 (Operation timed out)
[19:02] Misirlou ([email protected]) joined #gstreamer.
[19:07] <ensonic> thomasvs: is there any chance that 'adder' gets a free seat in you todo-top-ten ;-) ?
[19:08] kmaraas ([email protected]) left irc: Remote closed the connection
[19:09] kmaraas ([email protected]) joined #gstreamer.
[19:17] BBB ([email protected]) joined #gstreamer.
[19:18] <thomasvs> ensonic: yeah, it's going to be checked soon :)
[19:19] Action: ensonic has been worked on a slave ship in one of his former lifes
[19:21] Action: BBB wonders what to write next
[19:21] <BBB> thomasvs: please look at the encodingprofile patches I provided
[19:21] <BBB> thomasvs: they're a good first reason to branch off either 0.8.x or 0.9.x
[19:21] <thomasvs> BBB: ?
[19:21] <BBB> in bugzilla
[19:21] pb_ (~pb@2002:5160:4550:1:240:95ff:fe30:aa84) joined #gstreamer.
[19:21] <thomasvs> BBB: I'm not psychic :)
[19:22] Action: BBB looks up
[19:22] <BBB> did I mention that bugzilla is slower?
[19:22] <BBB> http://bugzilla.gnome.org/show_bug.cgi?id=118142
[19:22] <ensonic> BBB: what about the adder ;-)
[19:22] Action: BBB points to audio experts in here
[19:22] <BBB> I'm only video
[19:26] <alley_cat> BBB: any plans to support audio stream switching in totem with gst backend? :)
[19:26] <alley_cat> it's one of the few things i'm missing (subtitle support being the biggest miss)
[19:26] <BBB> no actual plans, but yes, the end goal is to support that
[19:26] <dolphy> we should be able to do that one day or another :)
[19:26] <BBB> I'd suggest to ask dolphy to help witht hat too
[19:26] <BBB> speaking of the devil ;)
[19:27] <dolphy> devil ? me? :)
[19:27] <BBB> but I'm currently only focussing on actual media playback
[19:27] <dolphy> you kidding
[19:27] <dolphy> i m like a white angel
[19:27] <BBB> LeRoutier, for example, did a crashtest for a huge series of movies to identify where we detect media types wrong (or not at all)
[19:27] Action: BBB pets dolphy
[19:28] <BBB> I want more such tests of unknown media trown at totem, and fix each nd any bug that results
[19:28] <BBB> only then will I care about new features such as visualization, subtitles or ...
[19:28] <BBB> I guess dolphy will work the other way around, so he's your best bet :)
[19:29] Uraeus ([email protected]) joined #gstreamer.
[19:29] <alley_cat> hmm, just noticed totem with xine backend doesn't suppoert audio stream switching for those files either ...
[19:29] <Uraeus> g'day
[19:29] <BBB> alley_cat: even then, we should
[19:29] <alley_cat> BBB: agreed :)
[19:29] <alley_cat> i can choose the stream with gst-launch, so it shouldn't be too hard to support it in totem :)
[19:29] <BBB> hi Uraeus
[19:29] <BBB> alley_cat: DVD has the same issue
[19:30] <thomasvs> later guys
[19:30] <BBB> we need some generic way of selecting a stream from a series of equal pads
[19:30] <Uraeus> BBB: I think the audio plugin used with mkv files has caps issues
[19:30] <BBB> pad grouping or so
[19:30] <Uraeus> bye thomasvs :)
[19:30] <LeRoutier> but for bivix et subtitles, even if not actually supported, would it breaks because of too numerous/unknown streams or would it only play one audio stream and ignore subs ?
[19:30] <BBB> Uraeus: that can be any
[19:30] walters ([email protected]) joined #gstreamer.
[19:31] <dolphy> BBB: the other way around ?
[19:31] <dolphy> BBB: are you kidding :)
[19:31] <Uraeus> BBB: the star wars clip
[19:31] <Uraeus> hi walters
[19:32] <dolphy> ok enough for today
[19:32] dolphy ([email protected]) left irc: "Network down, IP Packets delivered via UPS"
[19:33] <Uraeus> hmm both thomasvs and dolphy leaving, I am starting to wonder if team barcelona is full of slackers :)
[19:33] <BBB> Uraeus: we can't downchannel 5:1 to 2:0 sound yet
[19:33] <BBB> so osssink doesn't play it
[19:33] <BBB> feel free to open a bug report, we don't have one yet
[19:33] <walters> yo Uraeus
[19:33] <BBB> I'd love to see a bug for each failing movie
[19:34] <BBB> and someone should make a tracker
[19:34] <jdahlin> Uraeus: yeah, damn slackers
[19:34] <BBB> so all 'my media X Y Z doesn't play' bugs are tracked
[19:34] <BBB> LeRoutier: currently, it simply plays the first sound clip
[19:34] <Uraeus> jdahlin: talking about slacking, fixed my g-s-r bug yet? :)
[19:35] <BBB> Uraeus: is there such a bug yet?
[19:35] <LeRoutier> BBB: for the media sample in my FTP, i'll give the path but not the full FTP uri (i don't really want anyone seeing bugzilla to know a ftp user/pass on my server
[19:35] <BBB> Uraeus: for the starwars clip?
[19:35] <BBB> ok, fine with me
[19:36] <BBB> then please add my email address to each as assigner, plus add a quote that I know the FTP login
[19:36] <BBB> people willing to work on the bugs should be able to do so
[19:36] <LeRoutier> BBB: feel free to give user+pass to any gst people you trust and that could help resolving it
[19:36] <Uraeus> BBB: I create a startwars bug :)
[19:36] <BBB> so I can then send files to those persons individually
[19:36] <BBB> Uraeus: url?
[19:36] <BBB> LeRoutier: sure :)
[19:36] ChrisHJW_log ([email protected]) joined #gstreamer.
[19:37] <Company> BBB: open your tracker bug
[19:37] <BBB> k
[19:37] <jdahlin> Uraeus: bla bla bla
[19:38] <Company> BBB: and start by adding the spiderman trailer bug to it :)
[19:38] <Uraeus> BBB: I have created it yet, but I do so today, you have the star wars clip yourself right?
[19:41] sub_pop ([email protected]) joined #gstreamer.
[19:41] bilboed ([email protected]) left irc: Remote closed the connection
[19:43] <BBB> http://bugzilla.gnome.org/show_bug.cgi?id=138435
[19:43] <BBB> Uraeus: yes, but provide a link nevertheless
[19:43] <BBB> Company: feel free to make it depend on it yourself :)
[19:45] wheels ([email protected]) joined #gstreamer.
[19:45] jcsston ([email protected]) left irc: Read error: 60 (Operation timed out)
[19:46] <Company> BBB: done
[19:46] <BBB> now, the cool thing is that if we have several tens of such bugs, we can easily ask relatively new people to hack on that
[19:46] <BBB> because they're usually fairly easy to solve
[19:46] <BBB> Company: ty
[19:47] <BBB> Company: and if you want, add yourself CC to the tracker... it'd be cool if all developers did that so we can give this some priority in the next few months :)
[19:47] <BBB> oh
[19:47] <BBB> wait
[19:47] <BBB> you did
[19:47] <BBB> d'oh
[19:47] <Company> yeah
[19:47] <BBB> nevermind ;)
[19:47] Action: Company likes replacing mplayer with gst-player some day
[19:48] thomasvs ([email protected]) left irc: No route to host
[19:49] Action: Company wonders if he should poke wheels wrt kiosrc and event loops again
[19:50] <ensonic> cu, tomorrow
[19:50] ensonic ([email protected]) left #gstreamer ("Leaving").
[19:50] <Uraeus> jdahlin: I sold my appartment today, got 1 070 000 NOK for it :)
[19:50] <jdahlin> Uraeus: great
[19:50] <jdahlin> Uraeus: you should have taken my offer though
[19:50] <BBB> where are you going to live?
[19:51] <BBB> au?
[19:51] <jdahlin> 10 SEK is clearly more than 1M NOK
[19:51] <Uraeus> BBB: for the next few months, my mothers place, then I will travel then world, and then hopefully Barcelona :)
[19:51] <Uraeus> s/then/the/
[19:51] <iain> 100,000quid? nice
[19:51] <Uraeus> iain: yeah, actually a little more as you should divide by 8
[19:52] <Uraeus> iain: of course the bank will take half of it :)
[19:52] <iain> that'd buy a very nice house in N. Ireland
[19:52] <iain> or half an apartment on the riverfront in belfast, but you don't want to live there :)
[19:54] <iain> if Ximian people are "monkies" because of their logo, does that mean GStreamer people are slugs?
[19:54] <iain> the sluggies
[19:55] <Uraeus> yup, we are the slimiest hackers in the community :)
[19:58] <iain> if sluggy.com was still funny, I'd try to get a mention on the site
[19:58] <iain> but I've not read it in a year or so
[20:00] <LeRoutier> bbl, lunch
[20:00] Nick change: LeRoutier -> LeRaway
[20:02] BBB ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[20:03] <wheels> Uraeus: So it's true that you get your CVS account deactivated if word gets around that you've bathed recently?  :-)
[20:04] <Uraeus> wheels: yeah, we have a special community set up to do suprise controls, kinda like the people checking for drug abuse in sports :)
[20:07] ChrisHJW_log ([email protected]) left irc: Read error: 60 (Operation timed out)
[20:08] yippi ([email protected]) left irc: Remote closed the connection
[20:09] <wheels> Uraeus: I mean -- just imagine if random developers started showering -- next thing you know they'd be up in arms about everyone else stinking and whatnot; just think of the impact on the community.
[20:10] <Uraeus> yeah, some of them could even encouter situations where they where allowed to breed if we allowed them to clean themselves
[20:11] <iain> uh oh...I'm just about to go and shower...but I want my CVS account kept
[20:11] <wheels> Uraeus: Well, and its widely known that copulation and the associated baggage is a cheif distractor from coding.
[20:11] <Company> wheels: where is the kio source code?
[20:11] Action: Company can't find it in webcvs
[20:11] <wheels> Company: kdelibs/kio
[20:12] <Company> thx
[20:14] Uraeus ([email protected]) left irc: "Client exiting"
[20:15] walters ([email protected]) got netsplit.
[20:15] lupusBE ([email protected]) got netsplit.
[20:15] jonathang ([email protected]) got netsplit.
[20:15] Rotty ([email protected]) got netsplit.
[20:15] alley_cat ([email protected]) got netsplit.
[20:15] taaz ([email protected]) got netsplit.
[20:15] danb ([email protected]) got netsplit.
[20:16] walters ([email protected]) returned to #gstreamer.
[20:16] lupusBE ([email protected]) returned to #gstreamer.
[20:16] jonathang ([email protected]) returned to #gstreamer.
[20:16] Rotty ([email protected]) returned to #gstreamer.
[20:16] alley_cat ([email protected]) returned to #gstreamer.
[20:16] taaz ([email protected]) returned to #gstreamer.
[20:16] danb ([email protected]) returned to #gstreamer.
[20:18] Kaetzchen ([email protected]) joined #gstreamer.
[20:18] alley_cat ([email protected]) left irc: Nick collision from services.
[20:18] Nick change: Kaetzchen -> alley_cat
[20:19] BBB ([email protected]) joined #gstreamer.
[20:19] <BBB> *poof* and gone was internet
[20:20] Nick change: LeRaway -> LeRoutier
[20:20] markey ([email protected]) joined #gstreamer.
[20:22] md`` ([email protected]) joined #gstreamer.
[20:23] <LeRoutier> oh, the "novell will switch to qt" just reached slashdot
[20:24] <markey> about 1 week delayed
[20:24] <markey> woohoo
[20:24] <jdahlin> Regarding Novell and Qt, read http://slashdot.org/comments.pl?sid=102104&cid=8705159
[20:24] Action: BBB checks date
[20:24] <BBB> too close
[20:24] <BBB> ;)
[20:25] sjoerd ([email protected]) left irc: Remote closed the connection
[20:28] md` ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[20:29] lupusBE ([email protected]) left irc: "Bezig met verlaten"
[20:30] jdahlin ([email protected]) left irc: "Leaving"
[20:34] Rotund ([email protected]) joined #gstreamer.
[20:34] <Rotund> okay.  I have a couple questions
[20:34] <Company> Rotund: go ahead
[20:34] <Rotund> I'm having a problem building "libgstinterfaces-0.8.1a"
[20:34] <Rotund> no rule to make target
[20:35] <Rotund> needed by oss and alsa
[20:35] <Rotund> is there something in the makefile for it?
[20:35] <Company> yeah
[20:35] <Rotund> I mean configure script
[20:36] <Company> it should be built in gst-libs/gst/ when you build gst-plugins
[20:36] <Rotund> hmmm.
[20:36] <Rotund> Gentoo splits up the plugins package into separarte packages
[20:36] <Rotund> build scripts... whatever you wanna call it
[20:38] <LeRoutier> Rotund, bad boy, you unmasked gnome 2.6 ?!
[20:38] <Rotund> it IS being built when I do the "big gst-plugins"
[20:38] <Rotund> =)
[20:38] <Company> pkgconfig's gstreamer-interfaces-0.8 is needed
[20:39] <BBB> er
[20:39] <BBB> rotund
[20:39] <Rotund> It's not building it when just building the alsa one
[20:39] <BBB> gentoo questions belong on the gentoo fora
[20:39] <Rotund> or the oss one
[20:39] <BBB> not in developer channels
[20:39] <Rotund> sorry.  I'm wondering if it's not an issue in the configure script
[20:39] <BBB> ask foser if you want to, he's the gentoo build maintainer for the gstreamer-related packages
[20:39] <BBB> no, it's a gentoo bug
[20:39] <BBB> not a gst bug
[20:39] <LeRoutier> Rotund, it is gentoo specific. go to #gentoo or ask foser when he's back (gentoo gnome master)
[20:39] <Rotund> okay.  sorry.
[20:40] <BBB> np :)
[20:42] <Company> waargh
[20:43] <Company> kde is complicated
[20:44] jcsston ([email protected]) joined #gstreamer.
[20:47] <ds-work> http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=banana+slugs&btnG=Google+Search
[20:47] <ds-work> Company: !
[20:48] <LeRoutier> Rotund, you meant /usr/lib/libgstinterfaces-0.8.la not .1a i suppose
[20:50] <LeRoutier> Rotund, anyway, usually, masked packages aren't supported, you'd better not fill a bug report about it, or you'll be shot
[20:50] <Company> ds-work: i think we should not link ANY caps to any other caps
[20:50] <Company> ds-work: or we should make filesrc emit a different caps
[20:52] <Company> ds-work: i think it's better if we define semantics for the GstFormats and offset fields of "unknown" streams, while an ANY streeam shouldn't have those
[20:52] <ds-work> Company: yeah
[20:53] <ds-work> Company: I wanted to get rid of ANY, but couldn't figure out how without a lot more work
[20:53] <LeRoutier> bbl
[20:53] LeRoutier ([email protected]) left irc: "Leaving"
[20:53] <ds-work> of course, we're halfway there, since filesrc ! filesink will negotiate to application/octet-stream
[20:54] <ds-work> but ANY is still useful for identity
[20:54] <Company> yeah
[20:54] <ds-work> but only in the pad template
[20:55] <Company> but the stuff that filesrc emits and the stuff identity accepts are different
[20:55] <Company> we should have a capsrewriter element that allows changing caps at will
[20:55] markey ([email protected]) left irc: "bbl"
[20:56] <ds-work> yes
[20:57] <ds-work> imo, iaiw, we should always use filesrc ! typefind ! mpegdemux, etc.
[20:57] <ds-work> that way, filesrc can always use application/octet-stream
[20:57] <Company> iaiw?
[20:57] <ds-work> iaiw = in an ideal world
[20:58] <Company> woot, google changed layout
[21:00] <Company> ds-work: btw, please use --gst-scheduler=entrygthread to find problems with switching the core to forced-discont-before-buffer
[21:02] <ds-work> can that be added easily to other schedulers?
[21:02] <Company> ds-work: it should be added to the core
[21:02] <ds-work> Company: btw, I agree about the registry stuff -- I've had my eye on rewriting it for a while
[21:03] <ds-work> Company: ah, ok.  Is this something that may cause bugs (hint hint)?
[21:03] <Company> ds-work: i've already fixed one bug
[21:03] Action: BBB freaks out and decides it's time to propose branching off gstreamer-0.8.x
[21:03] <Company> ds-work: that's why i hacked that into the scheduler only
[21:04] <Company> we don't branch off gstreamer-0.8.x
[21:04] <ds-work> BBB: I disagree.  If we can add stuff carefully, I think it's a good thing
[21:04] Action: BBB is a scared-ass
[21:04] <BBB> what's your plan then?
[21:04] <ds-work> any chance we can switch to arch?
[21:05] <BBB> dunno, what's that?
[21:05] <Company> i don't think arch would solve that problem
[21:05] <wheels> BBB: CVS for weenies.  :-)
[21:05] <Company> we use cvs HEAD in the future
[21:05] Nick change: md`` -> md`
[21:05] <Company> and do releases from cvs HEAD
[21:06] <Company> unless we need to clean a brown paperbag, then we release from the branch of last release
[21:06] <BBB> hm, right
[21:07] <walters> which problem?
[21:07] <BBB> didn't we have some mp3 file bugs too?
[21:07] <BBB> that should go into the trackerbug?
[21:07] <walters> branching?
[21:07] <Company> walters: the problem of having a seperate HEAD and 0.8 branch
[21:08] <walters> Company: well, arch doesn't have a formalized idea of HEAD really.  you would just have a 0.9 branch and a 0.8 branch, like rb does.
[21:08] <ds-work> do we want to release 0.10 in september?
[21:08] <BBB> probably not
[21:08] <BBB> we can do 0.8.x a while longer
[21:08] <Company> i don't
[21:09] <Company> i want to branch off 0.9 in september
[21:09] <BBB> 0.8.x?
[21:10] <Company> walters: yeah, but i don't want an 0.9 branch yet
[21:10] <Company> walters: i want to have the developers eat their own released dogfood for a while
[21:10] <walters> Company: sounds reasonable to me.
[21:12] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[21:12] <BBB> hey, I'm happy with 0.8.x right now ;)
[21:12] <BBB> at least people are fixing bugs now :-p
[21:13] <Company> yeah, that's why i don't want to branch ;)
[21:13] yippi ([email protected]) joined #gstreamer.
[21:13] <BBB> the problem is that my profiles stuff should go into non-0.8.x for a while before being backported
[21:13] <BBB> since I might want to change API
[21:13] apoc ([email protected]) joined #gstreamer.
[21:14] <Company> enclose the headers in #ifdef GST_I_KNOW_THIS_MAY_CHANGE_RANDOMLY or something
[21:14] <apoc> hi
[21:14] <BBB> heh :)
[21:15] <ds-work> BBB: that's why we need arch :)
[21:15] <Company> BBB: can't you just put it into sandbox?
[21:16] <Company> (apart from the fact that a profiles lib does not belong in gst-plugins but in gst-media-lib or something similar anyway)
[21:16] <Company> well, apart from allowing plugins to define their own profiles...
[21:16] <Company> oh well
[21:16] jcsston ([email protected]) joined #gstreamer.
[21:20] bilboed_school ([email protected]) left irc: "Leaving"
[21:20] bilboed_school ([email protected]) joined #gstreamer.
[21:23] teuf ([email protected]) joined #gstreamer.
[21:23] jcsston` ([email protected]) joined #gstreamer.
[21:25] <BBB> Company: I'm not sure
[21:25] <BBB> Company: it's similar stuff as metadata, really
[21:25] <BBB> that's why it's a separate lib (not in gstinterfaces or gst-core)
[21:25] <BBB> but still in gst-plugins
[21:25] <BBB> I'm not sure where it belongs
[21:26] <BBB> I'll put the UI in gst-sandbox, that's where it's currently on my HD ;)
[21:26] <Company> it's not really that similar to metadata
[21:27] <Company> it's just a List<GProperty> => string mapping
[21:27] <BBB> true
[21:27] <BBB> but in 'impact' or 'level'
[21:30] <Company> you could just ship all of them in an xml file and wouldn't need to require gst-plugins at all
[21:30] <BBB> that'd suck if I want to add external projects
[21:31] <BBB> like profiles provided with the monkeyaudio encoder or the shorten encoder
[21:31] jcsston ([email protected]) left irc: Read error: 60 (Operation timed out)
[21:32] <Company> you _could_ do it that way (kinda like gconf)
[21:32] <Company> i didn't say you should ;)
[21:32] <Company> qt sucks
[21:33] BBB ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[21:34] BBB ([email protected]) joined #gstreamer.
[21:34] <BBB> what was the last thing I said?
[21:34] <Company> <BBB> like profiles provided with the monkeyaudio encoder or the shorten encoder
[21:34] <Company> <-- jcsston has quit (Read error: 60 (Operation timed out))
[21:34] <Company> <Company> you _could_ do it that way (kinda like gconf)
[21:34] <Company> <Company> i didn't say you should ;)
[21:34] <BBB> ok, so we think alike
[21:34] <BBB> I tried saying this:
[21:34] <BBB> http://bugzilla.gnome.org/buglist.cgi?product=GStreamer&component=gstreamer+%28c
[21:34] <BBB> hm
[21:34] <BBB> not that
[21:35] <Rotty> ds-work: is arch support planned?
[21:35] <BBB> (14:29:58) BBB: that's why I want this element-specific stuff in the elements themselves
[21:35] <ds-work> Rotty: support?
[21:37] mxpxpod ([email protected]) joined #gstreamer.
[21:38] mathrick ([email protected]) joined #gstreamer.
[21:40] <mathrick> moin
[21:40] <BBB> hi
[21:40] Action: BBB back to work
[21:40] <BBB> bbl
[21:40] BBB ([email protected]) left #gstreamer.
[21:43] Nick change: taaz -> taaz-gone
[21:44] <Rotund> actually, the Gentoo bug is sorta a bug in the Makefile
[21:46] <Rotund> also, (if anyone cares) the Makefiles refer to variables as $(top_builddir) and ${top_builddir}
[21:46] <Rotund> both are valid, just a pain
[21:53] <Company> *$?&*$&*&* qt
[21:54] <mathrick> quicktime or trolltech's child?
[21:54] <Company> trolltech
[21:55] <mathrick> why are you touching that? :)
[21:55] <Company> either i don't grok C++ at all or it's just not possible to run multithreaded
[21:55] <Company> mathrick: kiosrc
[21:56] <mathrick> ahh, right. So you decided to use qt as mainloop provider?
[21:56] <Company> no
[21:56] <Company> i decided to run my own event loop for the kioslave i'm using
[21:56] <Company> unfortunately the concept of "my own" is unknown to qt
[21:57] <Company> it's either global or not at all
[21:57] <mathrick> eh?
[21:58] <Company> qt happily keeps a list of all pending events for you
[21:58] <Company> in the qapplication object
[21:58] <Company> unfortunately there is only one qapplication object
[21:58] <mathrick> oh, and you cannot convince it otherwise?
[21:59] <Company> unless you tell me how to duplicate global variables i don't think i can
[21:59] Action: mathrick looks at Qt's docs
[21:59] <Rotund> Company: what is the problem?  can't access events from a second thread?
[22:00] <Company> Rotund: can't add events to a different eventloop
[22:00] <Rotund> ahh.
[22:00] <Rotund> may I ask what kioslave is?  (use GNOME)
[22:01] <Company> can't make parts of kde add events to a different event loop
[22:01] <Company> kioslaves are the equivalent of gnomevfs modules
[22:01] <Rotund> okay.  what sort of events do you need?
[22:02] <Company> i need the kioslave i'm using to put the events it produces into a different event loop
[22:02] <Rotund> okay.  so pump an event
[22:02] <Company> the kioslave pumps events, not me...
[22:03] <Rotund> wait.  is pump add or remove?
[22:03] <Rotund> I mean add an event
[22:03] <Company> yeah
[22:03] <mathrick> Company: why do you need them in different loop?
[22:03] <mathrick> ie, what is that other loop?
[22:03] <Company> mathrick: there's the so called "main event loop"
[22:04] <Company> mathrick: that one may or may not exist, depending on gstreamer being used by a qt app or a gst app
[22:04] Nick change: mxpxpod -> mx|gone
[22:04] <Company> hmmm
[22:04] <mathrick> hmm, which is when?
[22:05] Action: mathrick lost a little bit
[22:05] <Company> yeah, it's not that easy at all ;)
[22:05] <Rotund> okay.  can you describe what you want to kioslave to do?
[22:06] markey ([email protected]) joined #gstreamer.
[22:06] <Company> the kioslave does stuff
[22:06] <Company> whenever the kioslave has data available, it puts an event in the main event queue
[22:07] <mathrick> k
[22:07] <Company> then, when the kde main event loop runs, those events are processed
[22:07] <Company> unfortunately we don't know if a kde main event loop runs
[22:07] <mathrick> are these two main loops one?
[22:08] <mathrick> one kioslave puts events in and kde's one?
[22:08] <Rotund> okay.  is the kioslave for gstreamer?
[22:08] <Company> mathrick: no, it's just one event loop
[22:08] <Company> mathrick: the thing called "main event loop" by kde guys
[22:08] <Company> ...is called...
[22:09] <mathrick> Company: so, kioslave puts its events into different loop that kde's one?
[22:09] <Rotund> you are supposed to use signals and slots to pass events
[22:09] <Rotund> I think
[22:09] <Company> mathrick: no, it uses the same
[22:09] <mathrick> Company: k, that was my initial question ;)
[22:10] <Company> mathrick: i'd like it to use a different one, but i have no idea how
[22:10] <mathrick> Company: so, problem arises when we don't have kde handy?
[22:10] <mathrick> or, wait, maybe it's the opposite
[22:10] <Company> mathrick: it's one of those cases ;)
[22:11] <Rotund> okay.  So what does the kioslave access?  I want to know what the kioslave does
[22:11] <mathrick> hmm, /me thinks we always want to put events into different than kde's loop, wether it's running or not
[22:11] <mathrick> because it's gstreamer who needs to get these events, not kde
[22:11] <mathrick> it == kde
[22:12] <Company> mathrick: yeah
[22:12] <Company> mathrick: but unfortunately, kde events are _always_ put into the kde queue
[22:12] Action: Company convinces wheels and markey to use Gtkmm in the future
[22:13] <markey> gtkmm?
[22:13] <markey> what's it
[22:13] <Company> gtk c++ bindings
[22:13] <markey> omg
[22:13] <markey> never
[22:13] <markey> :)
[22:13] <markey> take qt from my dead cold hands
[22:13] <mathrick> " For any GUI application that uses Qt, there is precisely one QApplication object, no matter whether the application has 0, 1, 2 or more windows at any time."
[22:14] <Company> mathrick found the problem
[22:14] <mathrick> Company: wait a sec, lemme see further
[22:14] Action: Company prepares to take qt from people's cold dead hands
[22:15] <Company> qt has more global variables than mikmod
[22:15] <Company> and mikmod is bad already
[22:15] <mathrick> http://doc.trolltech.com/3.2/qeventloop.html#details
[22:16] <mathrick> http://doc.trolltech.com/3.2/qapplication.html#enter_loop
[22:16] <markey> we'll get it running somehow
[22:16] <markey> I might go insane in the process
[22:16] <markey> but hey
[22:16] <Company> mathrick: that doesn't solve the problem...
[22:16] <markey> it's worth it ;)
[22:16] <Company> markey: a true KDE guy speaking :p
[22:17] <markey> tim jansen admitted he gave up on the plugin in terror
[22:17] <Company> hehe
[22:17] <markey> and basically forgot about gst for a while
[22:17] <mathrick> in second, read the eventLoop() description, I found no other link than by enter_loop() :)
[22:18] <Company> mathrick: i've found webcvs.kde.org/cgi-bin/cvsweb.cgi/qt-copy/src/kernel to be better :p
[22:18] Action: Company is a true "who needs docs anyway" guy ;)
[22:18] <markey> qt docs are so nice you want to read them even at night
[22:19] harshy ([email protected]) joined #gstreamer.
[22:20] <mathrick> "There can only be one event loop object. The QEventLoop is usually constructed 
[22:20] <mathrick>     by calling QApplication::eventLoop(). To create your own event loop object create 
[22:20] <mathrick>     it before you instantiate the QApplication object."
[22:20] <mathrick> hmm, not good
[22:20] Action: Company shoots
[22:22] <mathrick> looks like there needs to be some clever subclassing for external mainloop interaction
[22:25] <mathrick> Company: what about subclassing it to take GMainLoop, and forward any incoming event there?
[22:26] <Company> mathrick: and if the app uses KDE? ;)
[22:26] <mathrick> well, that means the position of one and only loop is already taken, right?
[22:27] <Company> mathrick: apart from that it doesn't work anyway
[22:27] <mathrick> why?
[22:28] <Company> because you can only start changing stuff in the plugin's plugin_init function
[22:28] <Company> and then it's too late already...
[22:28] spyder482 ([email protected]) joined #gstreamer.
[22:29] <mathrick> why? if we're not kde app, there's no QEventLoop yet, right?
[22:29] <mathrick> but if we are, well, then suckage happens
[22:29] <Company> yeah
[22:29] <Company> hm
[22:29] <Company> unlesss...
[22:29] Action: Company needs to test something
[22:30] <mathrick> markey: does KDE implement its own QEventLoop subclass ?
[22:30] <mathrick> markey: or uses stock Qt's one?
[22:30] <markey> I'd have to look it up
[22:30] <markey> let's see
[22:31] <markey> QApplication's eventloop
[22:32] <markey> KApplication derives from it
[22:32] <mathrick> and doesn't provide its own loop?
[22:32] <markey> no
[22:32] <markey> it doesn't work that way
[22:32] <markey> there's one event loop
[22:33] <mathrick> good, one more potential contact point avoided
[22:33] <markey> from QApplication
[22:33] <mathrick> I know, but kde may have wanted it to do something more spiffy than usual
[22:33] <mathrick> then it would have subclassed it
[22:34] <markey> right now I don't see another solution as the one tim already tried
[22:34] <mathrick> uh-huh, looks like my system exploded somewhat :\
[22:34] <mathrick> I installed udev from debian package, but I still run 2.4
[22:35] <mathrick> at least, I don't know of any other reason why suddenly my sound module stopped to load on startup
[22:40] <Rotund> okay.  to decode an ogg from gst-launch-0.8, what are the middle parts I need? (like vorbisdec)
[22:40] <mathrick> hmm, I wonder. Why didn't it autoload module for my card? It was always doing so
[22:41] <jonathang> mathrick, maybe it has to do with discover or hotplug?
[22:42] <mathrick> jonathang: maybe, I have discover, yes. But I didn't uninstall it, AFAIK
[22:42] Action: mathrick doesn't remember how his system works
[22:43] <jonathang> mathrick, discover has been updated to v2 lately... If I were you, I'd add the module's name in /etc/modules
[22:44] <mathrick> jonathang: yep, probably good idea. I need to find out if it's indeed a module :)
[22:45] <Company> Rotund: filesrc ! oggdemux ! vorbisdec ! audioconvert ! osssink
[22:45] <jonathang> mathrick, which sound card do you have?
[22:45] <mathrick> C-Media PCI
[22:45] <mathrick> Discovering hardware: cmpci ignore usb-ohci ne2k-pci
[22:45] <mathrick> Skipping cmpci; assuming it is compiled into the kernel.
[22:45] <mathrick> hmm
[22:45] <mathrick> but there IS cmpci module
[22:45] <jonathang> huh
[22:46] <jonathang> if you modprobe it, is it ok?
[22:46] <Company> mathrick: use alsa
[22:46] <jonathang> Company, you're soooooo right :)
[22:46] <mathrick> jonathang: not sure, my mixer app still complains
[22:46] <mathrick> Company: I will, when I upgrade to 2.6
[22:47] <jonathang> mathrick, with debian, it's really easy to use alsa, wether you have a 2.4 or 2.6 kernel
[22:47] <mathrick> but then, my custom compiled 2.6 dumped stacktrace on startup, which wasn't entirely desirable
[22:47] <Company> mathrick: i'm using a cmedia pci as my main soundcard (with alsa)
[22:48] <mathrick> jonathang: I had alsa earlier, but then something gone wrong when changing kernel or mobo, and then I changed it
[22:48] <mathrick> I don't remember the reason too well
[22:48] <jonathang> mathrick, time to try again ;)
[22:49] <mathrick> hell, it's high time to compile 2.6
[22:49] <jonathang> right :)
[22:49] <jonathang> and to remove discover :)
[22:49] <mathrick> I just need to copy debian's .config, it's proven to not crash :)
[22:49] <jonathang> unless you're changing your hardware every two days :)
[22:49] <mathrick> but I still have no clue why my own .config caused it to crash
[22:50] <mathrick> well, no, but I like having automagic here and there
[22:50] <jonathang> yes, but warning with alsa
[22:50] <mathrick> it makes me feel warm and fuzzy ;)
[22:50] <mathrick> why?
[22:50] <jonathang> discover and hotplug load oss modules
[22:50] <mathrick> ah, yes, I remember
[22:50] <jonathang> and then /etc/init.d/alsa is confused... :(
[22:51] <mathrick> I even added ignore line for my old card
[22:51] <mathrick> it's a pity discover can't be convinced otherwise
[22:51] <jonathang> the solution is to either configure discover correctly...
[22:51] <jonathang> or to remove it :)
[22:51] <mathrick> define 'correctly' :)
[22:52] <jonathang> i don't know... i removed it!
[22:53] <mathrick> hehe
[22:54] <mathrick> time for my usual kernel rant - I want stable drivers ABI, dammit!
[22:54] BBB ([email protected]) joined #gstreamer.
[22:55] <mathrick> wb BBB
[22:55] <BBB> ty
[22:56] <jonathang> good night everybody
[22:56] jonathang ([email protected]) left irc: "Leaving"
[22:57] Action: BBB kicks mathrick with the magic words mms
[22:57] <BBB> O-)
[22:57] <Company> use the xine wrapper :p
[22:57] <BBB> nohohohoho
[22:58] <BBB> I don't even have xinelib on my system
[22:58] <Company> it works for mms :)
[22:58] <BBB> I'm quite sure you only tested mp3, mms and qdm2 ;)
[22:59] <alley_cat> apropos xine, a question that came up recently: is it possible to have gstreamer framebuffer output through the xine plugin?
[22:59] <Company> not even mp3
[22:59] <mathrick> what is qdm2?
[22:59] <Company> but i tested cdda!
[22:59] <BBB> wee
[23:00] <BBB> xinesrc_cdda ! osssink
[23:00] <BBB> :D
[23:00] <Company> yup, that works
[23:00] <BBB> mea culpa
[23:00] <Company> xinesrc_cdda ! xineaudioout_oss should work, too
[23:00] Action: BBB writes some random muxer documentation
[23:01] <BBB> after that N-to-N, which is fairly simple because nobody uses it so it's all unknown ;)
[23:01] <BBB> and then the interesting stuff!
[23:01] <BBB> autopluggers and managers
[23:01] <Company> but i gave up, because wrapping is kinda boring ;)
[23:01] <BBB> d'oh
[23:01] <BBB> just cut out mms from it
[23:01] <BBB> that's all I care of
[23:01] <Company> i have mms :p
[23:01] <BBB> pff
[23:02] <Company> (i'm not sure asfdemux can handle it in all cases though)
[23:02] <Company> which was the reason i started chunklib ;)
[23:02] <BBB> anyone have opinions on libgstplay implementing typefind functions for m3u (mp3 playlist) and asx (asf xml description)?
[23:02] <Company> uhm
[23:02] <Company> that should be done by the typefind plugin
[23:03] <BBB> should it?
[23:03] <BBB> that's what I'm unsure about
[23:03] <BBB> since they contain playlists, and libgstplay handles those, not gstreamer itself
[23:03] Action: BBB is in doubt
[23:04] <mathrick> hmm, culprit #1: CONFIG_SOFTWARE_SUSPEND=y
[23:05] <Company> a lib should never implement a typefind function
[23:05] Action: mathrick tries w/o SWSUSP
[23:06] <Company> we need a way to handle playlists somehow though
[23:07] jdahlin ([email protected]) joined #gstreamer.
[23:07] jdahlin_ ([email protected]) joined #gstreamer.
[23:07] <iain> playlistsrc
[23:07] <Company> i'd like to have an asxdecode plugin that parses the asx file, connects to the stream and handles it (via gstreamer plugins)
[23:08] <Company> so it'd need to be a bin
[23:08] <Company> would be useful for rb and iradio streams, too
[23:08] jdahlin_ ([email protected]) left irc: Client Quit
[23:08] <Rotund> hmmm.  I can't seem to get oggs to work
[23:09] <Rotund> does this make sense? gst-launch-0.8 filesrc location=14-amen.ogg ! oggdemux ! vorbisdec ! audioconvert ! alsasink
[23:09] <Rotund> wait.  alsasink isn't working =)
[23:09] <Rotund> sorry
[23:10] <Rotund> Is alsasink supposed to work w/ alsalib 1.0?
[23:10] <Company> it does here
[23:10] <Company> 1.03b to be exact
[23:10] <Rotund> okay.  I have it working on MP3s but not oggs
[23:11] <mathrick> hrmpf, anyone knows what is the difference between swsusp and suspend-to-disk?
[23:11] BBB ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[23:11] <Rotund> ** ERROR **: file gstoptimalscheduler.c: line 752 (destroy_group): assertion failed: (group->group_links == NULL)
[23:12] Action: Company blames the scheduler
[23:12] <Company> try --gst-scheduler=basicgthread
[23:12] <Rotund> same thing works if I change to osssink
[23:12] <Rotund> =)
[23:12] <Rotund> works
[23:12] <iain> Company: whats your new scheduler called?
[23:13] <Rotund> I love you guys.  Rhythmbox works now =)
[23:13] <Company> iain: entry{gthread,omega}
[23:15] BBB ([email protected]) joined #gstreamer.
[23:20] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[23:22] Shoragan ([email protected]) left irc: "Leaving"
[23:26] jdahlin ([email protected]) left irc: "Leaving"
[23:32] <walters> anyone have a sample divx video?
[23:32] <teuf> I've got a few 700MB samples ;)
[23:32] <walters> mmm. big :)
[23:33] <teuf> walters: maybe http://www.divx.com/movies/?src=toptab_movies_from_/index.php
[23:34] <walters> teuf: looks like they cost $
[23:34] <teuf> ah crap
[23:34] <teuf> sorry
[23:35] <teuf> walters: http://jimmac.musichall.cz/gimp2demos.php then
[23:36] <walters> teuf: sweet, thanks
[23:37] Nick change: mx|gone -> mxpxpod
[23:39] kwm ([email protected]) left irc: "Leaving"
[23:41] Nick change: jcsston` -> jcsston
[23:42] BBB ([email protected]) joined #gstreamer.
[23:46] Action: BBB wonders what people are doing
[23:49] <Company> cursing CPP atm
[23:49] <Misirlou> BBB BBB BBB BBB
[23:54] <mathrick> configuring kernel
[23:54] <iain> watching invader zim
[23:55] <Misirlou> Building GStreamer and Rhythmbox!
[23:58] yippi ([email protected]) left irc: Remote closed the connection
[23:59] Nick change: apoc -> apocz
[00:00] --- Tue Mar 30 2004
[00:02] <Company> lalala
[00:02] <Company> did i mention compiling c++ takes ages
[00:02] <Company> ?
[00:03] <markey> _bloody_ ages
[00:03] yippi ([email protected]) joined #gstreamer.
[00:04] Action: mathrick compiles 2.6.4
[00:06] Action: Company suspects KDE apps need a KApplication and not a QApplication
[00:06] <BBB> Misirlou: yes?
[00:07] <Misirlou> BBB: nothing
[00:08] <BBB> :)
[00:10] Action: Company dislikes the fact that gstreamer rebuilds the registry every time i modify a plugin
[00:10] <Misirlou> BAD COMPANY!
[00:11] <Misirlou> /me text is in the third person, not first!
[00:11] Action: Misirlou slaps Company around a bit with a large trout
[00:12] <Misirlou> Having to add --disable-plugin-builddir is a PITA.
[00:13] <ds-work> Misirlou: we're going to change that soon
[00:13] <ds-work> Misirlou: just need to set up some sort of test, so that it doesn't break
[00:13] <ds-work> nobody uses it
[00:14] <Misirlou> Nobody uses --enable-plugin-builddir or nobody uses --disable-plugin-builddir?
[00:14] <ds-work> nobody uses --enable-plugin-builddir
[00:14] <Company> gngngngngngngn
[00:15] <BBB> someone really looks frustrated here
[00:15] Action: mathrick likes clean compile output when building 2.6 kernels
[00:16] <Company> i'm writing C++ code for kde...
[00:16] <Company> now where do those guys hide the KApplication docs?
[00:17] <BBB> why? :P
[00:17] <Company> kiosrc
[00:19] <Company> but my app just crashes with a sigill
[00:21] <Company> oh, now it wants KCmdLineArgs
[00:21] thaytan ([email protected]) left irc: Read error: 113 (No route to host)
[00:22] <mathrick> sounds almost like hungarian notation :)
[00:22] Action: mathrick wonders why StupidStudlyCaps are so popular amongst C++ crowd
[00:23] <Company> BecauseInTheirOpinionItJustLooksBetter_than_using_underscores
[00:24] ChrisHJW ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[00:24] ChrisHJW ([email protected]) joined #gstreamer.
[00:24] <mathrick> yeah, but why for function names too? This way you have one syntactic info lost
[00:24] <mathrick> because variableNames and functionNames look the same
[00:25] <Company> nonono
[00:25] <Company> function names start with a lowercase character
[00:26] <mathrick> and variables?
[00:26] <Company> hrm, right
[00:26] <Company> class names start uppercase
[00:26] <mathrick> and that's perfectly fine
[00:26] <Company> but there are no functions in C++ land, it's just members of objects
[00:26] <mathrick> blah
[00:27] <BBB> why would you need a KApplication to access KIO?
[00:27] <BBB> that sounds rather braindead
[00:27] <mathrick> myfoo.do_something() != myfoo.SomeSetting
[00:27] <Company> BBB: welcome to KDE land
[00:27] <mathrick> BBB: you missed the detailed discussion :)
[00:27] <BBB> I'm sure I did
[00:28] Action: BBB pets Company
[00:28] <mathrick> BBB: wanna backlog?
[00:28] Action: BBB won't fix it >:)
[00:28] <BBB> yeah
[00:29] <Company> kde dislikes "don't print everything you know to stdout or stderr" paradigm either
[00:29] <ds-work> BBB: there's a style of programming that takes object-oriented design, and intermingles all the classes so that you can't extract one without depending on OpenOffice
[00:30] <mathrick> BBB: got everything?
[00:30] <mathrick> hmm, no
[00:30] <ds-work> BBB: then there's another style with clear dependency layers
[00:31] <ds-work> (which is why having a gdk_pixbuf plugin for gstreamer and a gstreamer plugin for gdk_pixbuf is the height of insanity)
[00:32] steve_b ([email protected]) joined #gstreamer.
[00:32] <Company> in an ideal world gtk would depend on gstreamer
[00:32] <Company> and the gimp, too
[00:33] <ds-work> an ideal world?  Is that the one that's based on the principle of embrace and extend?
[00:33] <Company> no, that's the one where everyone understood that image parsing is a subset of video parsing
[00:34] <Company> well, at least animated pixmaps are videos
[00:34] <ds-work> that doesn't mean it should be done by gstreamer
[00:34] <mathrick> BBB: that's about everything
[00:34] <mathrick> hehe :)
[00:34] <mathrick> that means gstgst plugin, sort of :)
[00:34] <mathrick> why whould you call it ideal?
[00:34] Action: mathrick disagrees
[00:34] <mathrick> in an ideal world, we would have single binary running on bare metal
[00:34] <mathrick> but why should gtk depend on gimp?
[00:34] <mathrick> besides its name
[00:35] <Company> ds-work: well, i'm fine with a common superlib, too
[00:35] <Company> gimp should depend on gstreamer
[00:35] <Company> since it does animations
[00:35] <ds-work> the design of gegl is much more appropriate for image work
[00:35] <mathrick> well, this one could be argued, but not the other way
[00:35] <ds-work> and some mixture of gegl and gstreamer for animation work
[00:36] <Company> yeah
[00:36] <Company> combining gegl and gstreamer would be the video eitor people want
[00:36] <Company> and probably require fixing 1E22 bugs
[00:36] <Company> great
[00:36] <mathrick> you're being pessimistic
[00:37] <Company> now i say "no gui please" and Qt opens a connection to X for me
[00:37] <mathrick> btw, what about this gegl dev, he's been here once
[00:37] <Company> schweet
[00:37] <mathrick> is he dropping by here anymore?
[00:37] <Company> no clue
[00:37] jcsston ([email protected]) left irc: Read error: 60 (Operation timed out)
[00:38] <mathrick> Company: well, it's funny enough that ioslave requires you to do gui :)
[00:38] <Company> hey, kio has the prominent job.displayErrorDialog function
[00:38] <mathrick> _DIALOG_ ?
[00:39] <mathrick> ioslave does dialogs ??
[00:39] <Company> yeah, it does error dialogs for you
[00:39] Action: mathrick laughs
[00:39] sublett ([email protected]) left irc: "Why are all the really good songs short?"
[00:39] <BBB> so how does it work on dfb?
[00:40] <mathrick> it's qt magic, or doesn't at all
[00:40] <mathrick> either of these
[00:40] <Company> http://developer.kde.org/documentation/library/3.2-api/kio/html/classKIO_1_1Job.html#a7
[00:42] <BBB> sounds sweet
[00:43] <mathrick> QString KIO::Job::errorString() /* blah blah */ Do not call it if error() is not 0.
[00:43] <mathrick> ie, do not call errorstring when there is error?
[00:43] <BBB> this is rather evil
[00:43] <BBB> woa
[00:44] Action: BBB feels so sorry for company
[00:44] <BBB> such a good hacker... forever lost in the darkness of c++
[00:44] <BBB> ohwell
[00:44] <BBB> c'est la vie
[00:44] Action: BBB continues random character typing in gst-rec
[00:44] <BBB> er
[00:44] <BBB> programming
[00:45] <mathrick> hehe, and krowd actually calls glib bloated :)
[00:45] <mathrick> it's funny to see things @ the other side of fence
[00:45] Action: mathrick needs to try vim once, too
[00:45] <BBB> bbl
[00:46] Action: BBB goes homre
[00:46] <BBB> :)
[00:46] <iain> ERROR          id3tag( 5293) gstid3tag.c(701):gst_id3_tag_handle_event:<id3tag0> (null)
[00:46] BBB ([email protected]) left irc: "Download Gaim: http://gaim.sourceforge.net/"
[00:46] <iain> ERROR          id3tag( 5293) gstid3tag.c(701):gst_id3_tag_handle_event:<id3tag0> Seek during ID3v2 tag reading
[00:46] <iain> any idea?
[00:46] <Company> iain: using entrygthread?
[00:47] <iain> yes
[00:47] <Company> iain: updated id3tag today?
[00:47] <iain> no
[00:47] <Company> i've fixed that
[00:47] <iain> ok
[00:47] <iain> updating now
[00:48] teuf ([email protected]) left irc: "Client exiting"
[00:52] <mathrick> hmm, is initrd _required_ when I have it enabled in kernel?
[00:52] <mathrick> or can I leave it w/o initrd.img and it will just skip that?
[00:53] Action: Company doesn't know
[00:53] Action: Company knows kde now though
[00:53] <Company> i can now flame informed on /.
[00:55] <Company> "they don't seperate UI and backend" should be a pretty good flame
[00:56] <mathrick> nope, too informed
[00:56] <mathrick> KDE == bloat sounds better
[00:56] <Company> nah
[00:57] <Company> it's pretty good
[00:57] <markey> hmmm..
[00:57] <Company> doesn't say too much, but every stupid highschool student knows that you should do it
[00:59] <Company> though kiosrc will be the first element that displays an error dialog all by itself!
[01:00] <Company> guess what people will say when totem pops up a kde dialog ;)
[01:00] <mathrick> tee hee
[01:00] <mathrick> you should use the errorString(), really :)
[01:00] <Company> and password dialogs?
[01:01] <Company> :p
[01:01] <mathrick> hmm
[01:01] <mathrick> now, that's funny :>
[01:02] Action: mathrick relaxes with some asm code
[01:07] <Company> does kio have mms input?
[01:08] <Company> i could suggest kio instead of xine for BBB :)
[01:10] <mathrick> ok, point taken :)
[01:10] trow ([email protected]) left irc: "Leaving"
[01:11] markey ([email protected]) left irc: "leaving"
[01:13] <Company> mathrick: it's just nice to offer an equally good option to BBB
[01:13] <Company> mathrick: wasn't meant against you in any way
[01:14] <mathrick> Company: nono, that's my fault only
[01:15] <mathrick> if I were in any way normal, I would have finished it 3 times already :\
[01:15] <Misirlou> mathrick: I'm pretty sure that enabling initrd does not make it required.
[01:15] Action: mathrick hugs reegexps
[01:15] <mathrick> s/ee/e/
[01:16] <mathrick> Misirlou: ahh, that's good to hear
[01:16] <Misirlou> heheh
[01:16] <Misirlou> * mathrick hugs reegexps
[01:16] <Misirlou> <mathrick> s/ee/e/
[01:16] <Company> i so hate kde
[01:17] <Company> "dude, you need to convert this enum value to an int before using it here"
[01:17] <mathrick> Misirlou: yeah, that's why :)
[01:17] <Company> s/kde/kde and C\+\+/
[01:17] <Misirlou> MATHRICK CAN TOTALLY PREDICT THE FUTURE
[01:17] <mathrick> I just pretty-formatted 100+ lines of asm in 20s
[01:18] ChriHJW ([email protected]) joined #gstreamer.
[01:18] ChrisHJW ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[01:18] <mathrick> all thanks to regexps substitution
[01:18] <mathrick> btw, did I say writing asm using early borland C++ text control sucks? ;)
[01:19] <iain> no
[01:20] <mathrick> it does
[01:21] <Company> what do you write asm for?
[01:22] <mathrick> W-Machine
[01:22] <Company> what's that?
[01:22] <mathrick> W as in "Lecture" (in Polish), or prof. Wgrzyn (its creator)
[01:22] <mathrick> ahh, it's great fun @ my Uni :)
[01:23] <mathrick> it's super basic CPU you build youself from logic gates during lectures, then write microcode, and then asm
[01:23] <mathrick> gives you the most solid basics you can get
[01:23] <mathrick> and is incredible fun to do :D
[01:24] <iain> wow
[01:24] <Company> i so hate those stuff
[01:24] Action: mathrick loves that
[01:24] <Company> i want to do something useful, damnit
[01:24] <iain> I wish my uni had done interesting stuff
[01:24] <mathrick> I'm finally able to understand how it works, _really_
[01:24] <Company> you study x years and after that you throw away everything you did
[01:25] <mathrick> oh well, studies should learn you how to learn, that's the main purpose
[01:25] pb_ (~pb@2002:5160:4550:1:240:95ff:fe30:aa84) left irc: "Client exiting"
[01:25] thaytoo ([email protected]) joined #gstreamer.
[01:25] <mathrick> every knowledge you get is replaceable with something else
[01:25] <mathrick> but skill is not
[01:26] <iain> ERROR    entrygthread(28081) entryscheduler.c(308):schedule_next_element:<GstEntryGThreadScheduler@0x829a2c8> having elements that must be scheduled, but nothing that can be scheduled
[01:26] <Company> yeah, but _please_ do something useful
[01:26] walters ([email protected]) left irc: "out"
[01:26] <iain> ERROR    entrygthread(28081) entryscheduler.c(1016):gst_entry_scheduler_iterate:<GstEntryGThreadScheduler@0x829a2c8> returning error because of element error
[01:26] <mathrick> but it's true, some of our classes are incredibly stupid, boring and waste of time
[01:26] <iain> yey :)
[01:26] <Company> iain: cute
[01:26] <iain> COBOL comes to mind
[01:26] <mathrick> Company: it's uber useful
[01:26] <iain> Company: thats with marlin's pipeline
[01:26] <Company> mathrick: only for learning, not for doing anything
[01:27] Action: mathrick thrills @ thought of using Rational's crap
[01:27] <mathrick> Company: sure, but I'm learning, am I not?
[01:27] <Company> iain: you should enable the #if 0'd parts and paste me the output in a query
[01:27] <Company> mathrick: yes
[01:27] <Company> mathrick: i learned by writing GSTreamer code - that was a lot more useful
[01:27] <mathrick> hmm, is "thrill" correct word for "shake in fear" >
[01:28] <mathrick> Company: sure, but it doesn't give you understanding how CPU works
[01:28] <mathrick> Company: and I find CS Basics (the class we're doing it in) incredibly useful
[01:28] <mathrick> Company: it's just it - basics I might have not come across otherwise
[01:28] <Company> mathrick: but it tought me how to learn
[01:29] <mathrick> Company: I agree, learning by coding is the best. But you need some basic basics anyhow
[01:29] <mathrick> and that's precisly what this class is all about
[01:29] <mathrick> that's why it's so fun to do
[01:30] <Company> yeah, it's better than reading
[01:31] <Company> or having someone talking in front of you
[01:31] <mathrick> it's like with maths - when you do some topology research, you're doing it yourself. But you were earlier introduced to basic math concepts, to have some starting point
[01:32] <Company> well, you don't need to know how a CPU works to start coding ;)
[01:32] <mathrick> it's not like you can jump in some abstract math discussion without completing elemtary course earlier
[01:32] <mathrick> Company: but you need it to understand asm correctly, IMO
[01:33] <mathrick> at least, I always felt I lacked something in this regard. No more
[01:33] <Company> i could certainly use some asm
[01:33] <mathrick> and it always troubled me, how is asm translated into something CPU does. And _how_ it does that
[01:34] <mathrick> that's the main point, I hate using something I don't understand completely
[01:34] <mathrick> I might lack some particular knowledge, but I must understand what I'm doing
[01:34] <Company> good luck reading the Linux, gcc and libc source code before starting on gstreamer plugins ;)
[01:34] <mathrick> otherwise, it just doesn't feel right
[01:34] <Company> oh and hf with autotools ;)
[01:35] <mathrick> nooooo, auto* doesn't count ;)
[01:35] Action: mathrick loathes auto*
[01:51] thaytan ([email protected]) joined #gstreamer.
[01:51] <mathrick> thaytan: are you home and @ work simultanously?
[01:52] harshy ([email protected]) left irc: "I quit for now"
[01:55] roberz ([email protected]) joined #gstreamer.
[01:55] <roberz> hi
[01:55] <mathrick> hi roberz
[01:55] <roberz> i have gstreamer installed and gstreamer-plugins 
[01:55] <roberz> i have recently installed gnome, i use gentoo
[01:55] <thaytan> mathrick: sort of
[01:56] <roberz> but when i double click on a media file it doesn't open
[01:56] <thaytan> thanks to the wonders of ssh
[01:56] <roberz> is not gstreamer a player ?
[01:56] <roberz> i am confused :S
[01:56] <mathrick> thaytan: ahh, ic :)
[01:56] <mathrick> roberz: nope, gstreamer is wonderful technology that allows you to have players easily :)
[01:56] <roberz> so 
[01:57] <roberz> tell me the name of a player please
[01:57] <roberz> that uses gstreamer
[01:57] <mathrick> roberz: totem, or gst-player
[01:57] <roberz> totem ? all right
[01:57] <roberz> thx
[01:58] <mathrick> roberz: you may need some magic to get it running if using cvs gstreamer and gnome
[01:58] <roberz> nope, i use stable packages
[01:58] <mathrick> roberz: ok, should be working mostly out-of-box then
[02:01] iain ([email protected]) left irc: "leaving"
[02:02] CompIBook ([email protected]) joined #gstreamer.
[02:07] <mathrick> I should probably create some LaTeX / DocBook templates for my lab reports
[02:07] <mathrick> doing it in OOo each time isn't too hax0r, and probably not super efficient either
[02:11] <CompIBook> yay
[02:11] <CompIBook> gst-launch kiosrc ! filesink works
[02:11] <CompIBook> wheels: ping
[02:12] <mathrick> CompIBook: wow, congrats
[02:12] <mathrick> CompIBook: how did you solve mainloop issue?
[02:12] <CompIBook> mathrick: if the app is a kde app, it runs the mainloop
[02:13] <CompIBook> mathrick: if it's not a kde app, we run the mainloop in our element from time to time
[02:13] md` ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[02:13] <mathrick> hmm, not sure I get it correctly
[02:13] Action: mathrick thinks about it
[02:14] <CompIBook> there's a QEventLoop::processOneEvent function
[02:14] <CompIBook> and processEvents which processes all pending events
[02:15] <mathrick> k, I think I understand
[02:16] <mathrick> so, it turns out that solution was to basically not care about being KDE or not :)
[02:16] <mathrick> sometimes thinking ahead is bad ;)
[02:16] <CompIBook> no, the solution was to handle both cases with an if :)
[02:16] md` ([email protected]) joined #gstreamer.
[02:17] <mathrick> eh?
[02:17] <mathrick> if (kde) ?
[02:17] <CompIBook> yeah
[02:17] roberz ([email protected]) left irc: "Leaving"
[02:18] <CompIBook> it's a bit (in fact very) ugly, but works
[02:18] <mathrick> but won't just running processEvents DTRT ?
[02:18] <CompIBook> uhm no
[02:18] <mathrick> it shouldn't do much harm if it was run already
[02:18] <CompIBook> running processEvents from the kde main loop and our thread would lead to some ugly races and so on
[02:19] <mathrick> hmm, dunno
[02:19] <CompIBook> apart from that we should process gui events in  kiosrc
[02:19] <CompIBook> shouldn't
[02:19] <mathrick> well, we can alway showDialog() ;)
[02:19] <mathrick> and how do you check for kde?
[02:19] <CompIBook> global variables
[02:20] <CompIBook> the magic qApp object
[02:20] <CompIBook> i do that in plugin_init, so you must create your application before calling gst_init
[02:20] <mathrick> does it return whether mainloop was already created?
[02:21] <CompIBook> creating the QApplication object creates the mainloop and sets qApp = self
[02:22] <mathrick> so qApp is global var?
[02:22] <CompIBook> if i were as crazy as ds-work i'd now download juk source and make it use kiosrc
[02:22] <CompIBook> i'm not crazy though
[02:22] <CompIBook> mathrick: yeah
[02:22] <mathrick> yuck!
[02:23] <mathrick> talk about "elegant C++" here
[02:23] <mathrick> if NMM is of the same kind of elegance, I don't wanna see it :P
[02:24] <CompIBook> kde is not elegant
[02:24] <CompIBook> it's simple
[02:24] <mathrick> simple?
[02:24] <mathrick> I wouldn't say so
[02:24] <CompIBook> yeah
[02:24] <ds-work> isn't simple elegant?
[02:24] <CompIBook> if you have an error with your kio object, just call kio.showErrorDialog
[02:25] <CompIBook> ds-work: not for every meaning of simple
[02:27] <CompIBook> how does kde do autoconf voodoo?
[02:27] <mathrick> it's more of being simpleton than simple, IMHO
[02:29] thaytoo ([email protected]) left irc: Read error: 113 (No route to host)
[02:30] Nick change: crazney_ -> crazney
[02:32] <CompIBook> another cool feature of kde: don't ever use something standard
[02:32] <ds-work> "The author even explains how he then fit it in to a Red Hat box and used a sensor to tell the bootloader which OS to run based on which box it is in."
[02:32] <ds-work> lol
[02:32] BBB ([email protected]) joined #gstreamer.
[02:33] <CompIBook> huh?
[02:33] <mathrick> heh, something on /. ?
[02:33] <ds-work> yep
[02:33] crazney ([email protected]) left irc: "Client exiting"
[02:38] <CompIBook> so, i have to test for KDE with AC_CHECK_LIB now?
[02:41] <CompIBook> well, i could alternatively switch to importing an admin dir, rename configure.ac to configure.in.in, make autogen run make -f Makefile.cvs and wait for thomas' mail to the list...
[02:42] BBB_ ([email protected]) joined #gstreamer.
[02:43] BBB ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[02:44] BBB_ ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[02:44] jcsston ([email protected]) joined #gstreamer.
[02:49] thaytan ([email protected]) left irc: Read error: 110 (Connection timed out)
[02:52] md` ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[02:57] arana ([email protected]) joined #gstreamer.
[02:59] CompIBook ([email protected]) left irc: "Client exiting"


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