IRC Logs

*******************************************************************
[03:03] steveb_ ([email protected]) left irc: Read error: 110 (Connection timed out)
[03:05] foser ([email protected]) left irc: "[ I want to believe ]"
[03:15] sub_pop ([email protected]) joined #gstreamer.
[03:17] sub_pop ([email protected]) left irc: Remote closed the connection
[03:18] sub_pop ([email protected]) joined #gstreamer.
[03:26] <thaytoo> ds-work: I'm talking about P200 level spec, with possibly a more decent spec as the server
[03:26] BBB ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[03:26] BBB ([email protected]) joined #gstreamer.
[03:26] BBB_ ([email protected]) joined #gstreamer.
[03:27] <ds-work> thaytoo: that should be fine
[03:27] <ds-work> thaytoo: if they have mmx, it should be able to decode MPEG
[03:27] <thaytoo> a low-res mpeg, sure
[03:27] <ds-work> thaytoo: otherwise, you could decode MPEG on the server, and push out raw frames
[03:28] <ds-work> if that's too much bandwidth, you could jpeg compress the portions of the image
[03:28] BBB ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[03:28] BBB__ ([email protected]) joined #gstreamer.
[03:28] sub_pop ([email protected]) left irc: "Client exiting"
[03:28] <ds-work> ... ! jpegenc ! jpegdec ! ... works well now
[03:28] <ds-work> even with resizing :)
[03:28] <thaytoo> yeah, that or rtjpeg, mebbe
[03:28] <thaytoo> (jpeg is still reasonably heavy @ 25-30FPS
[03:28] <thaytoo> )
[03:29] BBB__ ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[03:29] <ds-work> rtjpeg is basically the same as jpeg
[03:29] <ds-work> except for headers and such
[03:30] BBB_ ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[03:30] <thaytoo> didn't know that
[03:30] BBB ([email protected]) joined #gstreamer.
[03:31] BBB_ ([email protected]) joined #gstreamer.
[03:32] <thaytoo> wow 4953 CVS commits since I subscribed 10 Dec 2002
[03:38] sub_pop ([email protected]) joined #gstreamer.
[03:44] <Misirlou> thaytoo: That's 4953 too few!
[03:45] mxpxpod ([email protected]) joined #gstreamer.
[03:48] BBB ([email protected]) left irc: Read error: 110 (Connection timed out)
[03:54] Lagreca ([email protected]) left #gstreamer.
[03:54] <thaytoo> Misirlou: well, get committing then :)
[03:56] <taaz> ls
[03:56] <taaz> oops
[04:10] BBB_ ([email protected]) left irc: Read error: 110 (Connection timed out)
[04:21] Nick change: jcsston` -> jcsston
[04:25] Nick change: spyder482 -> spyder_homework
[04:30] ProtectR ([email protected]) left irc: "Leaving"
[04:40] <danb> i'm trying to compile the latest cvs and having some trouble
[04:41] <danb> gst-plugins/gst-libs/gst/idct/mmx32idct.c seems to be pretty messed up, with many things commented-out that it still tries to use
[04:41] <danb> i'm not sure if that data might come in through headers (i doubt that it does)
[04:42] <danb> and the gstreamer module built fine
[04:42] <danb> should i just check-out a gst-plugins module from a few days ago, or is there a better solution?
[04:43] <mxpxpod> why would I be getting this error with rhythmbox HEAD and gstreamer 0.7.6:
[04:43] <mxpxpod> ** (rhythmbox:19911): CRITICAL **: how to remove plugins?
[04:43] <mxpxpod>  
[04:43] <mxpxpod> (rhythmbox:19911): GStreamer-WARNING **: could not load valid typefind function for feature 'video/x-msvideo'
[04:43] <mxpxpod>  
[04:43] Last message repeated 1 time(s).
[04:43] <mxpxpod> ** (rhythmbox:19911): CRITICAL **: how to remove plugins?
[04:43] <mxpxpod>  
[04:43] <mxpxpod> (rhythmbox:19911): GStreamer-WARNING **: could not load valid typefind function for feature 'audio/x-flac'
[04:43] <mxpxpod> and then it does that critical thing for every plugin
[04:44] <thaytoo> mxpxpod: it looks like your install is message up, or rhythmbox needs recompiling, although I don't know of any recent ABI change that would require that
[04:44] <mxpxpod> message up?
[04:45] <thaytoo> heh. messed up
[04:45] <mxpxpod> :D
[04:45] <mxpxpod> hmm
[04:45] <thaytoo> perhaps old plugin libs that need deleting from the install dir?
[04:45] mathrick|sleep ([email protected]) left irc: Read error: 60 (Operation timed out)
[04:45] <thaytoo> does gst-register run ok?
[04:45] <mxpxpod> thaytoo: yup
[04:45] mathrick|sleep ([email protected]) joined #gstreamer.
[04:46] <thaytoo> danb: I don't think that idct has changed since december
[04:46] <danb> thaytoo: hum
[04:46] <danb> thaytoo: i'm going to try 0.7.6, see what's different in the build process'
[04:47] <thaytoo> mxpxpod: and gst-inspect too?
[04:47] <thaytoo> how recently did you build rhythmbox?
[04:47] <thaytoo> danb: what's the actual error?
[04:47] <mxpxpod> thaytoo: a couple of minutes ago
[04:47] <mxpxpod> thaytoo: and gst-inspect works fine
[04:48] <danb> thaytoo: let me regen...
[04:48] harshy ([email protected]) joined #gstreamer.
[04:49] <thaytoo> hi harshy 
[04:49] <harshy> hey thaytoo
[04:49] <danb> thaytoo: starts with: mmx32idct.c:209:17: mmx.h: No such file or directory
[04:51] <danb> thaytoo: there are many "implicit declaration"s and "undeclared"s following that, and if i patch it up to find mmx.h, only the first error seems to go away
[04:51] <danb> thaytoo: and, inspecting the file, it seems all the undeclared errors are a result of the declarations/definitions being commented out in the file...
[04:51] Nick change: spyder_homework -> spyder482
[04:54] <danb> for instance, search for RND_FRW_ROW in http://freedesktop.org/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst-libs/gst/idct/mmx32idct.c?rev=1.3&content-type=text/vnd.viewcvs-markup
[04:59] <thaytoo> danb: I guess it's because I don't have libmmx
[05:00] <thaytoo> (so it works for me)
[05:02] <danb> thaytoo: ic
[05:09] sub_pop ([email protected]) left irc: "Client exiting"
[05:24] mathrick|sleep ([email protected]) left irc: Read error: 60 (Operation timed out)
[05:27] harshy ([email protected]) left irc: "I quit for now"
[05:28] mathrick|sleep ([email protected]) joined #gstreamer.
[05:31] harshy ([email protected]) joined #gstreamer.
[05:33] walters ([email protected]) joined #gstreamer.
[05:45] spyder482 ([email protected]) left irc: "Leaving"
[05:57] <danb> thaytoo: that did it, thanks
[05:57] <danb> but now gst-python is being funny
[06:01] <thaytoo> how so?
[06:01] <thaytoo> not that I've compiled it recently
[06:02] <thaytoo> there's a bug report about removing libmmx
[06:03] <thaytoo> ... usage
[06:03] <thaytoo> and delegating to liboil instead
[06:05] <danb> liboil? hmm. didn't do anything with that...
[06:05] <danb> um, the configuration didn't pick up include flags for pygtk for some reason
[06:06] <danb> PYGTK_CFLAGS isn't getting written into the makefile...
[06:06] <thaytoo>  liboil is ds' project 
[06:06] <thaytoo> "Optimised Inner Loops"
[06:07] <thaytoo> it's only CVS so far
[06:07] <danb> hmm
[06:07] <danb> meh
[06:07] <danb> i'm not looking for much performance right now. just want to program with gst-python
[06:09] <danb> stupid pos
[06:09] Action: danb resumes frustration
[06:09] <danb> theme of the day...
[06:09] <danb> how different is gst-python-1.0 from the current cvs?
[06:09] <danb> i'm questioning the value of trying to use the cvs at this point
[06:10] <danb> err, 0.1.0, i guess
[06:11] <taaz> vastly different
[06:11] <danb> f
[06:11] <taaz> there is little code left after the fluendo crew got their hands on it :)
[06:12] <danb> ic
[06:12] <taaz> little original code that is
[06:12] <danb> so it's not worth using if i want to continue programming against the current gst-python?
[06:13] <taaz> 0.1.0 probably doesn't work with recent 0.6 and jdahlin kills 0.6 support in cvs and i think thomasvs did scary cvs surgery so i'm not sure if you can check out a version that works with 0.6 anymore.
[06:13] <danb> ok
[06:13] <taaz> your best bet is a recent 0.7.x and cvs gst-python
[06:13] jcsston ([email protected]) left irc: "VDubMOD watchout TCME is coming!"
[06:13] <danb> well, i'm planning to build against the gstreamer-0.7.6 tarballs
[06:13] <danb> k
[06:14] <danb> i think the current cvs gst-python requires >=0.7.5 anyway
[06:14] <danb> so, yeah
[06:14] <taaz> that might work... has gstpy been updated for the 0.8 switch?
[06:14] <danb> *shrug*
[06:15] jcsston ([email protected]) joined #gstreamer.
[06:15] <danb> jdahlin said earlier to use cvs stuff. i'm using 0.7.6 gst because i got sick of trying to build the cvs gst code and figured 0.7.6 would be Good Enough
[06:16] <danb> maybe i'll just hack some CFLAGS into my env...
[06:16] <danb> err: s/some/some more/
[06:16] <thaytoo> I would not expect gst-python 0.1 to work against 0.7 at all
[06:17] <thaytoo> s/would/do/
[06:17] <danb> yeah, i'm not going to use 0.1
[06:17] <danb> it's the cvs stuff i'm talking about
[06:17] <thaytoo> oki
[06:18] <taaz> not only can you not expect it to work, it's flat out will not ;)
[06:18] <danb> hehe
[06:18] <danb> taaz: good. less reason for me to quit on building the cvs stuff
[06:18] <taaz> eh, too many negatives in that statement.. i'm not sure what i said ;)
[06:18] <taaz> just use cvs when in doubt.
[06:19] <danb> hehe, yeah
[06:19] jcsston ([email protected]) left irc: Client Quit
[06:21] <danb> the problem seems to be that the configure script isn't setting-up enough environment in the makefiles...
[06:22] jcsston ([email protected]) joined #gstreamer.
[06:22] <danb> but i'm trying to get around that by setting up the environment myself
[06:28] mathrick|sleep ([email protected]) left irc: Read error: 60 (Operation timed out)
[06:32] mathrick|sleep ([email protected]) joined #gstreamer.
[06:39] harshy ([email protected]) left irc: "I quit for now"
[06:40] <mxpxpod> is anyone around that can help me debug some gstreamer 0.7.6 problems?
[06:40] <mxpxpod> it can't seem to find plugins from programs
[06:41] <walters> mxpxpod: did you run gst-register-0.7?
[06:41] <walters> or maybe -0.8 now
[06:41] <mxpxpod> walters: yup
[06:42] <mxpxpod> walters: and gst-launch-0.8 works fine
[06:42] <mxpxpod> but any program that uses gst is crapping out
[06:42] <walters> mxpxpod: with the pipeline from the rb readme?
[06:44] <mxpxpod> walters: hmm, no
[06:44] <mxpxpod> RUNNING pipeline ...
[06:44] <mxpxpod>  
[06:44] <mxpxpod> ** (process:25291): CRITICAL **: how to remove plugins?
[06:44] <mxpxpod>  
[06:44] <mxpxpod> (process:25291): GStreamer-CRITICAL **: Factory for `flacdec' has no type
[06:44] <walters> what's the error?
[06:44] <mxpxpod> Segmentation fault
[06:44] <walters> hmm.
[06:44] <walters> can you delete your installation tree, reinstall, and reregister?
[06:44] <mxpxpod> I've done that twice now
[06:44] <walters> ok
[06:45] <mxpxpod> oh, and gstreamer 0.7.6 uses /system/gstreamer/0.8/default/audiosink
[06:45] <walters> yeah
[06:46] <taaz> eh... this 0.7+0.8 thing is annoying yo
[06:46] <mxpxpod> so I changed the command in the readme to reflect that when I ran it
[06:46] <walters> did you build with --disable-plugin-dir?
[06:46] <mxpxpod> what does that do?
[06:46] <taaz> i'm looking at creating debs but a 0.7 package cant be installing 0.8 stuff
[06:46] <mxpxpod> I've never gotten a definitive answer
[06:46] <walters> mxpxpod: well, it makes gst not assume you're going to be running from the build tree
[06:47] <mxpxpod> oh really
[06:47] <walters> i never really understood how running from the build tree was supposed to work anyways
[06:47] <walters> so i always pass --disable-plugin-dir
[06:47] <mxpxpod> heh
[06:47] <mxpxpod> okey dokey
[06:47] <walters> er
[06:47] <walters> --disable-plugin-builddir
[06:47] <mxpxpod> to both gst and gst-plugins?
[06:48] <walters> i always pass it to both, i think it only makes sense for gst
[06:48] <walters> but it doesn't hurt
[06:48] <mxpxpod> ok
[06:49] <walters> here's how i build both core and plugins:
[06:49] <walters> ./autogen.sh -- --disable-plugin-builddir --prefix=/build/gstreamer-0.7 --with-gconf-source='xml:readwrite:/build/rb/etc/gconf/' --with-gconf-schema-file-dir=/build/rb/etc/gconf
[06:49] <mxpxpod> walters: ok, let me try rebuilding
[06:49] <mxpxpod> good thing I have ccache
[07:01] <mxpxpod> walters: is the xml:readwrite: thing necesary?
[07:02] <mxpxpod> and does it need to be readwrite?
[07:02] <walters> yes and yes
[07:02] <walters> http://www.gnome.org/projects/gconf/
[07:03] thaytoo ([email protected]) left irc: "home"
[07:03] <danb> i've built gstreamer-0.7.6, gst-plugins-0.7.6, and gst-python (cvs), but keep getting this error when i try to run the gst-python examples: ImportError: ../gst/.libs/_gstmodule.so: undefined symbol: _gst_bin_type
[07:03] <mxpxpod> walters: do I want my /etc/gconf to be readwrite?
[07:04] <danb> i've set PYTHONPATH to include the root of the gst-python build tree (because the docs said to)
[07:04] <walters> mxpxpod: if you're installing rb into a separate prefix, yes
[07:05] <danb> it looks like python detecting that the linker failed to link _gstmodule, so i also tried exporting LD_LIBRARY_PATH to contain the directory where the appropriate gstreamer and gst-plugins .so files seem to reside, but it doesn't improve things
[07:06] <danb> any ideas?
[07:06] <mxpxpod> walters: and if I'm not?
[07:07] <walters> mxpxpod: then you need to install as root
[07:07] <mxpxpod> walters: ok
[07:08] Nick change: kh_out -> khalek
[07:11] <mxpxpod> walters: same results :(
[07:11] <mxpxpod> even with --disable-plugin-builddir
[07:11] <walters> mxpxpod: with flacdec?
[07:11] <mxpxpod> yup
[07:11] <walters> mxpxpod: do you have a ~/.gstreamer?
[07:11] <walters> mxpxpod: are you running gst-register-0.7 as root?
[07:11] <mxpxpod> I ran it as root and as my regular user
[07:12] <walters> ok, generally, i would just always run gst-register-0.7 as root
[07:13] <mxpxpod> ok, so should I remove my ~/.gstreamer-0.8?
[07:13] <walters> i would yes
[07:14] <mxpxpod> whoa
[07:14] <mxpxpod> that did it
[07:14] <walters> i suspect what happened is that your ~ cache was pointing at an older installation 
[07:14] <mxpxpod> strange
[07:14] <mxpxpod> thanks a lot
[07:15] <walters> i never run gst-register as non-root
[07:15] <walters> sure, np
[07:15] Action: mxpxpod hugs walters platonically
[07:16] <mxpxpod> now sound-juicer needs to use 0.7.6 and I'll be set!
[07:20] <taaz> walters: arch vs indep configure flags in cdbs: how?
[07:22] <walters> taaz: ah, at the moment, you kind of have to do it "manually" by including autotools-vars.mk and having separate things in configure-indep:: and configure-arch::
[07:22] <walters> iirc
[07:24] <taaz> eh.. this is getting too tricky.  i think i'll punt on gst core using cdbs for now
[07:24] <taaz> for today at least
[07:39] <taaz> so can anyone else with a relatively up-to-date debian box build the docs?  jadetex has been eating 100% cpu for 20 minutes or more and done nothing.  i'm wondering if it will finish at all...
[07:40] Action: taaz sighs...
[07:40] markey ([email protected]) joined #gstreamer.
[07:43] <dilinger> taaz: sure
[07:43] <dilinger> taaz: you working on 0.8 debian packages?
[07:43] <taaz> yeah
[07:44] <taaz> "sure" they build?  why would that not be the case for my system?
[07:44] <dilinger> taaz: i meant sure i can try building the docs
[07:49] <taaz> i haven't been able to build them from cvs for a month or two.  i figure something in my env is just really broken but i have no clue how to to debug it
[07:49] <dilinger> i'm going to try 0.7.6 as soon as build-dep is done
[07:51] Nick change: khalek -> kh_out
[07:57] <dilinger> taaz: it failed
[07:57] <dilinger> maybe i should make everything
[07:58] <dilinger> taaz: what subdir is it in?
[08:00] <taaz> pwg
[08:00] <taaz> mine fails on a docbook2ps command.  which says it's using sgml catalogs for some reason
[08:00] <dilinger> it worked for me in pwg
[08:01] <dilinger> it generated html docs only, though
[08:01] <dilinger> i assume it's generating ps for you?
[08:02] <dilinger> configure:21635: Will output HTML documentation
[08:02] <dilinger> configure:21655: Will not output PS documentation
[08:02] <dilinger> configure:21667: Will not output PDF documentation
[08:02] <taaz> no, that's the problem ;)
[08:02] <dilinger> heh, well, attempting to generate ps, at least
[08:02] <taaz> it probably depends on having some tools installed
[08:06] <dilinger> ok, after installing netpbm and a few other packages, it'll now create pdf and ps docs as well
[08:10] <dilinger> docbook2ps worked for me
[08:10] <taaz> this crap brings me to new levels of frustration
[08:10] <dilinger> cd build && docbook2ps -o .. pwg.xml
[08:11] <taaz> does it say its using /etc/sgml/catalog?
[08:11] <dilinger> yes
[08:12] <dilinger> (flood)
[08:12] <dilinger> *** Generating PS output ***
[08:12] <dilinger> cd build && docbook2ps -o .. pwg.xml
[08:12] <dilinger> Using catalogs: /etc/sgml/catalog
[08:12] <dilinger> Using stylesheet: /usr/share/sgml/docbook/utils-0.6.13/docbook-utils.dsl#print
[08:12] <dilinger> Working on: /home/dilinger/src/gstreamer-0.7.6/docs/pwg/build/pwg.xml
[08:12] <dilinger> jade:/usr/share/sgml/declaration/xml.dcl:31:27:W: characters in the document character set with numbers exceeding 65535 not supported
[08:12] <dilinger> x -i print -d /usr/share/sgml/docbook/utils-0.6.13/docbook-utils.dsl#print -V paper-type=A4 /usr/share/sgml/declaration/xml.dcl /home/dilinger/src/gstreamer-0.7.6/docs/pwg/build/pwg.xml
[08:12] <dilinger> Done.
[08:13] <taaz> at the jade msg mine sits and spins
[08:14] <dilinger> what kind of machine?
[08:16] <taaz> black case.  maybe 20" high.  not very loud.
[08:16] <taaz> what does it matter? ;)
[08:16] <mxpxpod> walters: thanks again
[08:16] <taaz> it's a x86 2.8GHz 1G ram box with latest debs
[08:17] <dilinger> taaz: maybe try building it within a pbuilder chroot, see if it's some weird interaction w/ stale config files or external packages?
[08:17] <taaz> eh... i found one issue
[08:18] <taaz> XML_CATALOG_FILES point to /etc/xml/catalog.. which i guess confused the sgml code or something.  now it dies in a random tex error instead of halting
[08:19] <taaz> i'm not sure if this is an improvement or not ;)
[08:19] <dilinger> i've always been a fan of error messages
[08:19] <dilinger> versus hanging
[08:22] <taaz> i said a _tex_ error.  nothing a human could read. ;)
[08:23] <dilinger> it could be a 9000 column c++ stl template error; i'd still take it over a hang w/ no usable output  ;p
[08:27] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[08:28] walters ([email protected]) left irc: "Off to build shareholder value"
[08:33] mx|ibook ([email protected]) joined #gstreamer.
[08:42] mx|ibook ([email protected]) left irc: "Leaving"
[08:49] Nick change: kh_out -> khalek
[09:15] thaytan ([email protected]) joined #gstreamer.
[09:22] Nick change: mathrick|sleep -> mathrick|Uni
[09:34] <taaz> dilinger: i lied.  it still hangs.  but if i break in the middle the run make again it gives bogus errors
[10:00] sjoerd ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[10:05] sjoerd ([email protected]) joined #gstreamer.
[10:07] thomasvs ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:13] Rotty ([email protected]) joined #gstreamer.
[10:16] sublett ([email protected]) joined #gstreamer.
[10:19] alley_cat ([email protected]) joined #gstreamer.
[10:23] markey ([email protected]) left irc: "leaving"
[10:26] kmaraas ([email protected]) left irc: "Leaving"
[10:38] bilboed_school ([email protected]) joined #gstreamer.
[10:41] Shoragan ([email protected]) joined #gstreamer.
[10:43] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[10:48] thomasvs ([email protected]) joined #gstreamer.
[10:51] mxpxpod ([email protected]) joined #gstreamer.
[10:51] <danb> i've built gst-python, but i'm having trouble using it (like, loading the module in python)
[10:51] <danb> i built and installed gstreamer, gst-plugins, and gst-python from cvs
[10:52] <danb> anyone around that's made this work?
[10:53] <thaytan> jdahlin, mainly
[10:53] <thaytan> and taaz
[10:53] <thaytan> but jdahlin lately
[10:53] <danb> yeah
[10:53] jdahlin ([email protected]) joined #gstreamer.
[10:53] <thaytan> oh look :)
[10:53] <danb> speak of the devil
[10:53] <danb> ;)
[10:54] <danb> jdahlin: do you have some time to help me out?
[10:54] <jdahlin> morning
[10:54] <jdahlin> danb: sure, shot
[10:54] <danb> g'morning
[10:55] <danb> jdahlin: i've built gstreamer, gst-plugins, and gst-python all from cvs, but the gst python module fails to load with this error: ImportError: /usr/local/lib/python2.3/site-packages/gst/_gstmodule.so: undefined symbol: _gst_bin_type
[10:55] <jdahlin> danb: hmm, let me have a look
[10:55] <jdahlin> I have some uncommitted stuff in my tree
[10:56] <danb> and when i `ldd gst/_gstmodule.so`, it only shows linux-gate, libpthread, libc, and ld-linux, which i figure is the problem
[10:56] <danb> ok
[10:57] <danb> yeah, the Makefile.am and configure.in that i got from cvs were still trying to build against 0.7 (which i changed)
[10:58] <jdahlin> danb: weird, I have no idea why it tries to access _gst_bin_type
[10:58] <jdahlin> danb: you're using latest cvs right?
[10:59] <danb> jdahlin: within the last number of hours, yeah
[11:00] <thaytan> oh look :)
[11:01] <thaytan> (as in, oh look what happens when you hit up-enter)
[11:02] <danb> jdahlin: are you going to be able to commit your tree soon?
[11:02] <jdahlin> danb: I just did, but it should not be related
[11:02] <danb> ic
[11:03] <jdahlin> I should try to build it against 0.8
[11:03] <danb> ah. what are you currently building against?
[11:04] KoRnouille ([email protected]) joined #gstreamer.
[11:08] markey ([email protected]) joined #gstreamer.
[11:08] <jdahlin> danb: builds fine here with 0.8
[11:08] <danb> hmm
[11:08] <jdahlin> except gstplay, but that's not necessary atm
[11:09] <danb> what about GST_MAJORMINOR=0.8
[11:09] <danb> err
[11:09] <danb> what about GST_MAJORMINOR=0.7 in configure.in?
[11:10] <danb> or maybe it's a different file...
[11:10] <jdahlin> I'll fix that in a sec
[11:11] <jdahlin> But try to change it to 0.8
[11:11] <jdahlin> Which is what my commit will do
[11:11] <danb> yeah, i have
[11:12] <danb> hmm, now i'm getting: make[2]: *** No rule to make target `gstinterfaces.defs', needed by `gstinterfaces.c. Stop.
[11:12] <danb> not sure how that happened
[11:13] <danb> let me start from a clean tree again...
[11:13] <jdahlin> no
[11:13] <jdahlin> wait :)
[11:13] <danb> hehe
[11:14] <danb> yeah, same error from a clean tree
[11:14] <danb> tell me when...
[11:14] <jdahlin> 2 min
[11:14] <danb> np
[11:14] Action: jdahlin waits for cvs roundtrips
[11:15] <danb> roundtrips?
[11:15] <jdahlin> takes a few seconds to commit stuff
[11:15] <danb> ah
[11:15] <jdahlin> should be okay now
[11:15] <danb> k
[11:15] <jdahlin> thomasvs: woho, syncmail broke!
[11:18] <danb> jdahlin: IOError: [Errno 2] No such file or directory: 'xoverlay.override'
[11:18] <danb> during make, doing something with codegen.py
[11:18] <danb> want the full text block in priv?
[11:19] <jdahlin> danb: nope, that's good enough
[11:19] Action: jdahlin is stupid
[11:20] <danb> k
[11:20] <jdahlin> danb: could you please cvs up and retry?
[11:20] <danb> ya
[11:22] <danb> same error, was i too quick?
[11:23] Nick change: markey -> markey|afk
[11:27] <jdahlin> danb: yeah, several minutes, now it's really commited
[11:27] <danb> jdahlin: k
[11:28] Action: jdahlin does a make distcheck
[11:28] <danb> jdahlin: hmm, `cvs diff` shows nothing new
[11:28] <jdahlin> danb: you need to do a cvs up?
[11:28] <danb> ah, that updated xoverlay.override
[11:29] <danb> k, built
[11:29] <danb> jdahlin: still the same undefined symbol error...
[11:30] <jdahlin> danb: hmm :/.
[11:30] <jdahlin> Can't really see why
[11:31] <danb> yeah
[11:31] <danb> what version of gst were you building against before?
[11:31] <jdahlin> danb, 0.7.5 and 0.7.6
[11:32] <danb> hmm
[11:32] <danb> i might try those tomorrow or something
[11:32] <danb> it's getting late here
[11:33] <jdahlin> You might have a broken build, but I'm not sure
[11:33] <danb> *shrug*
[11:35] <danb> i've been messing with it all day
[11:35] <danb> kind of tired of it at this point
[11:35] <danb> i'll try again later
[11:35] <jdahlin> thanks
[11:35] Action: jdahlin tries to figure out why
[11:36] <jdahlin> Can you run the rest of the gstreamer programs?
[11:36] <danb> hehe, that would probably be a good thing to test
[11:36] <danb> let me check
[11:38] <danb> yeah
[11:39] pb_ ([email protected]) joined #gstreamer.
[11:39] <danb> had to register the plugins, but `gst-launch filesrc ... ! spider ! osssink` worked
[11:40] <jdahlin> weird.
[11:43] <danb> jdahlin: what do you get from `ldd .../gst/_gstmodule.so`?
[11:46] <danb> oh wow
[11:46] <danb> yeah, my build is missing a few
[11:47] <jdahlin> I wonder how configure could pass
[11:47] <jdahlin> what system are you on?
[11:47] <danb> gentoo
[11:48] <danb> so, i think i *have* all the deps. i think the dynamic libs just weren't linked against them properly or something
[11:51] crazney ([email protected]) joined #gstreamer.
[11:51] <crazney> anyone know where I might find some example source code for writing a gstreamer source?
[11:51] <thomasvs> crazney: any of the source plugins
[11:51] <thomasvs> crazney: depends on what you want to write
[11:57] <crazney> thomasvs: ok.. so say i have a proprietary streaming protocol, and currently my crappy implementation of it just writes it to a pipe.
[11:57] <crazney> but i kinda want to make some nice source plugin that buffers and all that shit. not really sure how best to approach it
[11:59] kmaraas ([email protected]) joined #gstreamer.
[12:07] smoke ([email protected]) joined #gstreamer.
[12:14] pb_ ([email protected]) left irc: Remote closed the connection
[12:18] <thomasvs> crazney: I guess looking at gnomevfssrc might help, it has some streaming code
[12:22] pb_ ([email protected]) joined #gstreamer.
[12:32] smoke_ ([email protected]) joined #gstreamer.
[12:32] smoke_ ([email protected]) left irc: Client Quit
[12:36] <thomasvs> jdahlin: ok, so throw gst-ffmpeg in the buildbot
[12:39] Action: jdahlin battles rsync
[12:43] lilo ([email protected]) got netsplit.
[12:43] kmaraas ([email protected]) got netsplit.
[12:43] Rotty ([email protected]) got netsplit.
[12:43] crazney ([email protected]) got netsplit.
[12:43] thomasvs ([email protected]) got netsplit.
[12:43] khalek ([email protected]) got netsplit.
[12:43] sxpert ([email protected]) got netsplit.
[12:43] Misirlou ([email protected]) got netsplit.
[12:43] steveb__ ([email protected]) got netsplit.
[12:43] hyriand ([email protected]) got netsplit.
[12:43] snikkt ([email protected]) got netsplit.
[12:43] sjoerd ([email protected]) got netsplit.
[12:43] sublett ([email protected]) got netsplit.
[12:43] bitshifter ([email protected]) got netsplit.
[12:43] ds ([email protected]) got netsplit.
[12:43] pb_ ([email protected]) got netsplit.
[12:43] bilboed ([email protected]) got netsplit.
[12:43] ds-work ([email protected]) got netsplit.
[12:43] grub_booter ([email protected]) got netsplit.
[12:43] sri ([[email protected]) got netsplit.
[12:43] Jara[zZ] ([email protected]) got netsplit.
[12:43] Shoragan ([email protected]) got netsplit.
[12:43] sxpert_work ([email protected]) got netsplit.
[12:43] markey|afk ([email protected]) got netsplit.
[12:43] yippi ([email protected]) got netsplit.
[12:43] camh ([email protected]) got netsplit.
[12:43] jimmy_dean ([email protected]) got netsplit.
[12:43] bluejay ([email protected]) got netsplit.
[12:43] thaytan ([email protected]) got netsplit.
[12:43] mathrick|Uni ([email protected]) got netsplit.
[12:43] danb ([email protected]) got netsplit.
[12:43] thies ([email protected]) got netsplit.
[12:43] tromey ([email protected]) got netsplit.
[12:43] jdahlin ([email protected]) got netsplit.
[12:43] bilboed_school ([email protected]) got netsplit.
[12:43] md` ([email protected]) got netsplit.
[12:43] wtay_ ([email protected]) got netsplit.
[12:43] shawarma ([email protected]) got netsplit.
[12:43] KoRnouille ([email protected]) got netsplit.
[12:43] coredump ([email protected]) got netsplit.
[12:43] sri_ ([email protected]) got netsplit.
[12:43] taaz ([email protected]) got netsplit.
[12:43] dilinger ([email protected]) got netsplit.
[12:46] pb_ ([email protected]) returned to #gstreamer.
[12:46] kmaraas ([email protected]) returned to #gstreamer.
[12:46] crazney ([email protected]) returned to #gstreamer.
[12:46] markey|afk ([email protected]) returned to #gstreamer.
[12:46] KoRnouille ([email protected]) returned to #gstreamer.
[12:46] jdahlin ([email protected]) returned to #gstreamer.
[12:46] thomasvs ([email protected]) returned to #gstreamer.
[12:46] Shoragan ([email protected]) returned to #gstreamer.
[12:46] bilboed_school ([email protected]) returned to #gstreamer.
[12:46] sublett ([email protected]) returned to #gstreamer.
[12:46] Rotty ([email protected]) returned to #gstreamer.
[12:46] sjoerd ([email protected]) returned to #gstreamer.
[12:46] thaytan ([email protected]) returned to #gstreamer.
[12:46] mathrick|Uni ([email protected]) returned to #gstreamer.
[12:46] steveb__ ([email protected]) returned to #gstreamer.
[12:46] bitshifter ([email protected]) returned to #gstreamer.
[12:46] lilo ([email protected]) returned to #gstreamer.
[12:46] yippi ([email protected]) returned to #gstreamer.
[12:46] bilboed ([email protected]) returned to #gstreamer.
[12:46] camh ([email protected]) returned to #gstreamer.
[12:46] sxpert_work ([email protected]) returned to #gstreamer.
[12:46] hyriand ([email protected]) returned to #gstreamer.
[12:46] snikkt ([email protected]) returned to #gstreamer.
[12:46] coredump ([email protected]) returned to #gstreamer.
[12:46] thies ([email protected]) returned to #gstreamer.
[12:46] khalek ([email protected]) returned to #gstreamer.
[12:46] danb ([email protected]) returned to #gstreamer.
[12:46] sri ([[email protected]) returned to #gstreamer.
[12:46] grub_booter ([email protected]) returned to #gstreamer.
[12:46] sri_ ([email protected]) returned to #gstreamer.
[12:46] ds-work ([email protected]) returned to #gstreamer.
[12:46] md` ([email protected]) returned to #gstreamer.
[12:46] ds ([email protected]) returned to #gstreamer.
[12:46] Misirlou ([email protected]) returned to #gstreamer.
[12:46] jimmy_dean ([email protected]) returned to #gstreamer.
[12:46] tromey ([email protected]) returned to #gstreamer.
[12:46] wtay_ ([email protected]) returned to #gstreamer.
[12:46] sxpert ([email protected]) returned to #gstreamer.
[12:46] Jara[zZ] ([email protected]) returned to #gstreamer.
[12:46] bluejay ([email protected]) returned to #gstreamer.
[12:46] dilinger ([email protected]) returned to #gstreamer.
[12:46] taaz ([email protected]) returned to #gstreamer.
[12:46] shawarma ([email protected]) returned to #gstreamer.
[13:03] Nick change: markey|afk -> markey
[13:09] jonathang ([email protected]) joined #gstreamer.
[13:27] spyder482 ([email protected]) joined #gstreamer.
[13:27] Nick change: spyder482 -> spyder_away
[13:55] LeRoutier ([email protected]) joined #gstreamer.
[13:56] <LeRoutier> hi
[14:01] sublett ([email protected]) left irc: "Nice of you to put me on the money"
[14:15] <thaytan> evening
[14:18] <LeRoutier> hi thaytan
[14:18] arana ([email protected]) joined #gstreamer.
[14:19] <LeRoutier> hi arana
[14:23] <thaytan> thomasvs: awake?
[14:29] sublett ([email protected]) joined #gstreamer.
[14:34] spyder_away ([email protected]) left irc: Read error: 110 (Connection timed out)
[14:42] kmaraas ([email protected]) left irc: "Leaving"
[14:43] markey ([email protected]) left irc: "leaving"
[14:45] mathrick|Uni ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[14:45] mathrick ([email protected]) joined #gstreamer.
[14:47] kmaraas ([email protected]) joined #gstreamer.
[14:49] haiku ([email protected]) joined #gstreamer.
[14:49] <haiku> hello
[14:49] <haiku> bilboed salut :)
[14:50] <thaytan> salut, haiku
[14:55] <bilboed> haiku: lu
[14:55] smoke ([email protected]) got netsplit.
[15:01] smoke ([email protected]) got lost in the net-split.
[15:03] mathrick ([email protected]) left irc: Remote closed the connection
[15:03] <haiku> anybody use the hoontech dsp24 soundcard ?
[15:09] <LeRoutier> sounds like a pro card, no ?
[15:11] <haiku> yup
[15:12] <haiku> i have some problems with it under alsa, i cannot get low latency (buffer size < to 2048)
[15:19] sublett ([email protected]) left irc: "Nice of you to put me on the money"
[15:21] <alley_cat> thomasvs: are the unversioned binaries installed by gstreamer 0.6.x and 0.7.x identical?
[15:23] <LeRoutier> alley_cat, i don't think he's around
[15:25] Uraeus ([email protected]) joined #gstreamer.
[15:25] <Uraeus> hello
[15:25] <LeRoutier> hello Uraeus
[15:29] pb_ ([email protected]) left irc: "bbl"
[15:30] <Uraeus> ds, ds-work: around?
[15:41] mxpxpod ([email protected]) left irc: "To quit or not to quit.. wait... was that a question?"
[15:48] <LeRoutier> is there anyone here that have speex 1.0.x and can try the following gst-plugins patch ? http://bugzilla.gnome.org/showattachment.cgi?attach_id=25560
[15:49] markey ([email protected]) joined #gstreamer.
[15:50] markey ([email protected]) left irc: Client Quit
[15:50] <Uraeus> LeRoutier: sorry, no speex here 
[15:51] smoke ([email protected]) joined #gstreamer.
[15:52] lin_matt|work ([email protected]) joined #gstreamer.
[15:59] teuf ([email protected]) joined #gstreamer.
[16:00] <thomasvs> back
[16:00] <Uraeus> hi thomasvs
[16:01] <Uraeus> thomasvs: how is the mood in Barcelona atm considering the bombings in Madrid?
[16:02] <thomasvs> people are demonstrating and convening everywhere
[16:02] <thomasvs> we are making plans to bomb the basques
[16:03] <LeRoutier> hello teuf, thomasvs
[16:05] <thomasvs> hi Leroutier
[16:08] Nick change: khalek -> kh_zZz
[16:09] <alley_cat> hmm, error in gstreamer 0.6.5 make install
[16:09] <alley_cat> cp -pr  /usr/share/doc/gstreamer-0.6.5/faq
[16:09] <alley_cat> cp: missing destination file
[16:12] <thomasvs> alley_cat: yeah, you need the doc tools required to build the docs in 0.6.5
[16:13] <alley_cat> what are those, gtk-doc and ?
[16:13] <alley_cat> gtk-doc is installed
[16:13] <kh_zZz> cool gstreamer is the first result in google for multimedia framework
[16:14] <thomasvs> the whole shebang - openjade, xmlto, docbook2, stylesheets, and so on
[16:14] <alley_cat> hmm, the only thing i don't have is xmlto
[16:18] <Uraeus> hmm, CVS rb seems to crash, probably due to some issues with how it use the new fileselector
[16:18] BBB ([email protected]) joined #gstreamer.
[16:19] BBB ([email protected]) left irc: Client Quit
[16:19] BBB ([email protected]) joined #gstreamer.
[16:20] Action: BBB kicks gaim
[16:20] <BBB> hi all :)
[16:20] <BBB> thomasvs: I didn't test gst-ffmpeg.new yet, will do soon
[16:20] <Uraeus> hi BB
[16:20] <Uraeus> B
[16:20] <LeRoutier> hi ronald
[16:20] <BBB> did you add a README on howto use it?
[16:20] <BBB> hi Uraeus
[16:21] <bitshifter> silly question maybe, but: I'm looking at the deinterlace plugin and wondering if the plugin can make the assumption that it can directly write into GST_BUFFER_DATA(buf) in the chain function? Couldn't it be some kind of read-only buffer (e.g. mmaped)?
[16:25] kmaraas ([email protected]) left irc: Client Quit
[16:27] Nick change: LeRoutier -> LeRaway
[16:28] <thomasvs> BBB: I did
[16:28] <thomasvs> so test it right now, and give me the source to that colorspace element RIGHT NOW
[16:33] <thomasvs> yippi
[16:33] Action: thomasvs got the translation project to agree on translating gstreamer
[16:33] <jdahlin> thomasvs: will you need to import the po files to cvs?
[16:34] <jdahlin> eg, gnome cvs
[16:34] <thomasvs> jdahlin: they send me .po files
[16:34] <thomasvs> or rather, web links
[16:34] <thomasvs> jdahlin: not the gnome one
[16:34] <thomasvs> jdahlin: the official gnu one
[16:34] <jdahlin> thomasvs: oh
[16:34] <thomasvs> it'll be messy to maintain
[16:34] <thomasvs> but still better than trying to do it ourselves :)
[16:34] iain ([email protected]) joined #gstreamer.
[16:35] <thomasvs> iain: so have you tried media-info ?
[16:35] <iain> I'm having issues with marlin and gstreamer not being on the best of terms at the moment
[16:35] <iain> thomasvs: but my only problems with it was that it was only available in CVS and not in the rpms
[16:36] <iain> thomasvs: so Uraeus using rpms couldn't build marlin
[16:37] <thomasvs> right - I'm going to put it in then
[16:38] jcsston ([email protected]) joined #gstreamer.
[16:42] <BBB> bitshifter: there's a function to check
[16:42] <BBB> bitshifter: and another to copy on read-only
[16:42] <BBB> bitshifter: something like gst_buffer_copy_on_write()
[16:42] <BBB> call that, the output buffer is mostly the input buffer, unless it was readonly in which case it's a copy
[16:43] <BBB> thomasvs: sorry, not at home
[16:43] <BBB> thomasvs: is it ok if I send it tonight?
[16:44] <BBB> I forgot to send it yesterday - I wasn't at home yesterday night (we went out to a really cool bar with a blues band; the singer was also singing on broadway, she was beyond excellent)
[16:44] <bitshifter> BBB: thanks. The element doesn't seem to do that - so I was wondering if I'm missing something
[16:44] <BBB> instantly fell in love :P
[16:44] <BBB> bitshifter: bug :)
[16:45] <BBB> thomasvs: can't you get FDO to do something?
[16:45] <BBB> I mean
[16:45] <BBB> isn't FDO having a project-wide problem here?
[16:46] <BBB> why doesn't FDO just give them FDO-wide CVS xs?
[16:47] <thomasvs> BBB: FDO to do what ?
[16:49] <BBB> to give translators (e.g. the GNOME translation team) FDO-wide CVS xs so they can do all this
[16:51] <thomasvs> BBB: we don't work with the gnome translators for this
[16:51] <BBB> (and GNOME isn't politically correct, so you could form a FDO translation team which consist of GNOME and KDE translators, sometyhing like that)
[16:51] <thomasvs> BBB: THE gnome translators only work with stuff in gnome cvs
[16:51] <thomasvs> BBB: most fdo projects use the international translation project
[16:51] <thomasvs> BBB: it doesn't work through cvs, for any of the projects, and it has a few decent reasons for that
[16:51] <thomasvs> BBB: anyway, it doesn't really matter :) I had to find us some translators, and I found them.
[16:52] <BBB> cool
[16:52] <BBB> :)
[16:56] foser ([email protected]) joined #gstreamer.
[16:57] Action: lin_matt|work is away: fixing things stupid people breaak
[16:59] teuf ([email protected]) left irc: "Client exiting"
[17:02] <BBB> thomasvs: care to change default scheduler to basicgthread?
[17:02] <BBB> I'm sick of opt
[17:02] <BBB> and MPEG plays better anyway in basic
[17:03] <Uraeus> BBB: don't we risk breaking other stuff then?
[17:03] <BBB> nah
[17:03] <BBB> opt implements only half of a scheduler
[17:04] Nick change: LeRaway -> LeRoutier
[17:05] <thomasvs> BBB: no, I don't want to change schedulers so late
[17:05] <thomasvs> we changed it away from basicgthread last time because it was better
[17:05] <BBB> even if MPEG can not be played back with opt?
[17:06] <BBB> no
[17:06] <BBB> we never had basicgthread by then
[17:06] <BBB> we changed from basicomega
[17:06] <BBB> basicgthread was introduced far later
[17:07] <jdahlin> BBB, would be nice with bug reports
[17:07] <jdahlin> against the opt scheduler, so we know what's broken
[17:07] <BBB> MPEG playback, I've mentioned that several times and sent several emails to the list about it
[17:07] <jdahlin> test cases even better
[17:07] <BBB> I can explain the problem in detail in an email if you wish
[17:07] <jdahlin> BBB, sure, I'm interested
[17:07] <jdahlin> BBB, send it to [email protected]
[17:12] sublett ([email protected]) joined #gstreamer.
[17:14] Action: lin_matt|work is back (gone 00:16:41)
[17:22] mxpxpod ([email protected]) joined #gstreamer.
[17:24] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[17:25] <BBB> jdahlin: sent
[17:26] <jdahlin> BBB, thanks
[17:26] yippi ([email protected]) left irc: Remote closed the connection
[17:29] markey ([email protected]) joined #gstreamer.
[17:48] haiku ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[17:50] <BBB> jdahlin: any thoughts?
[17:55] <jdahlin> BBB, not yet, trying to figure out what a bufpen is
[17:56] kwm ([email protected]) joined #gstreamer.
[17:57] Action: bitshifter realises he has just been staring at videotestsrc ! vertigoTV ! ximagesink for at least 5 minutes
[17:57] <BBB> jdahlin: a list of buffers that's poending between two elements in a basic scheduler
[17:58] trow ([email protected]) joined #gstreamer.
[17:59] <thomasvs> BBB: so how come I can play mpeg fine in totem/gst-player ?
[17:59] Nick change: Uraeus -> Ura_gym
[18:00] <jdahlin> oh, en norrman!
[18:00] Action: jdahlin looks at Ura_gym
[18:00] <Ura_gym> hei på dej grabben, nu måste jag løpa for at hinna helsestudiot
[18:01] <jdahlin> helsestudiot :)
[18:02] Action: iain gets his ever handy norge/english Ordbok
[18:02] mxpxpod ([email protected]) left irc: "Leaving"
[18:04] spyder482 ([email protected]) joined #gstreamer.
[18:08] <thomasvs> BBB: do you know if container formats like avi or rm have patent issues ?
[18:16] jonathang ([email protected]) left irc: "Leaving"
[18:23] earthworm ([email protected]) joined #gstreamer.
[18:25] arana ([email protected]) left irc: Read error: 113 (No route to host)
[18:29] <thomasvs> BBB: could you PLEASE try gst-ffmpeg.new ? I want to move it
[18:30] lin_matt|work ([email protected]) left irc: "I like core dumps"
[18:34] spyder482 ([email protected]) left irc: Read error: 113 (No route to host)
[18:34] <BBB> thomasvs: tonight... I'm not at home, remember?
[18:35] <BBB> thomasvs: some MPEG movies play fine... others have issues
[18:35] <thomasvs> BBB: well, I want to move gst-ffmpeg.new to gst-ffmpeg :)
[18:35] <thomasvs> do you have any local stuff uncommited ?
[18:35] <BBB> some MPEG movies simply have time-aligned elementary streams
[18:35] <BBB> others don't
[18:35] <BBB> no
[18:35] <BBB> I did a small commit yesterday
[18:35] <BBB> other stuff is all in there
[18:35] <thomasvs> I reorganized some patches, and added some stuff, like removing configure from the autotools patch, and adding an m4 dir and so on
[18:35] <BBB> oh, and I want to commit a new ffmpeg version again tonight, just so I can test how that works
[18:35] <BBB> ok?
[18:35] <thomasvs> basically, finetuning your already excellent work
[18:35] <thomasvs> ok, so I can just move it ?
[18:36] <BBB> sure
[18:36] <thomasvs> argh ,no :)
[18:36] <BBB> yes
[18:36] <thomasvs> no new ffmpeg verison yet
[18:36] <BBB> I need tyo test that ;)
[18:36] <thomasvs> I have to figure out how to do it cleanly
[18:36] <thomasvs> I know
[18:36] <thomasvs> but after the release please
[18:36] <thomasvs> it needs careful handling to be done correctly.
[18:36] <BBB> sure, sure
[18:36] <BBB> haha
[18:36] <BBB> I guess
[18:36] <thomasvs> It's not as easy as quilt pop -a, due to the cvs stuff
[18:36] <BBB> can you automate that in a script for me? :)
[18:36] <thomasvs> yeah, that's the goal
[18:36] <thomasvs> that's why I want to do the first updating
[18:36] <thomasvs> to make sure everything looks ok
[18:36] <BBB> if you automate *that* part, the other stuff is easy :)
[18:36] <thomasvs> but after the release :)
[18:36] <BBB> sure
[18:36] <BBB> :)
[18:37] <thomasvs> yeah, I know
[18:37] <thomasvs> when can I release gst-ffmpeg ?
[18:37] <thomasvs> and where can I find out what formats it plays ?
[18:37] <BBB> I'd like to add one small thing
[18:37] <BBB> probabl;y sunday
[18:37] <thomasvs> (btw, I left in the player, I suggest we enable it, because it's a good way of testing if ffmpeg plays the format)
[18:38] <BBB> ok
[18:38] <BBB> uninstalled, right
[18:39] Action: BBB noticed the Makefile.am hack
[18:39] <BBB> :P
[18:40] <thomasvs> which one ?
[18:40] <BBB> all-local:\n\t$(MAKE) -C ffmpeg all
[18:40] <BBB> and removing ffmpeg from the SUBDIRS
[18:40] <BBB> pretty fancy
[18:40] <BBB> didn't think of that myself
[18:40] <thomasvs> ah, yeah
[18:40] <thomasvs> neat trick
[18:41] <thomasvs> wonder when it will break
[18:41] <BBB> I'll let you know if I break it
[18:41] <thomasvs> I've done quite a few surgery on automake in the past :) pays off.  was a ten second hack
[18:41] Action: BBB will play with gst-ffmpeg tonight
[18:41] Action: thomasvs wonders how to make indent not do G_BEGIN_DECLS typedef enum on one line
[18:41] <BBB> my end goal is to playback all ffmpeg formats, even the ones that gst doesn't know about (types etc.)
[18:42] <BBB> eehw
[18:42] <BBB> submit a bug report :P
[18:44] <jdahlin> danb: So, your problems are now fixed, latest cvs should work as expected.
[18:49] <BBB> any chance one of you two wants to look at gnome sound recorder crashing if I record for a second time?
[18:50] <thomasvs> BBB: jdahlin claims he fixed that
[18:50] <BBB> oh... can you close the corresponding bug then? :)
[18:50] <jdahlin> I don't have the bug #
[18:50] Action: BBB looks up
[18:51] sub_pop ([email protected]) joined #gstreamer.
[18:53] <BBB> 129164
[18:53] <BBB> I just added a comment
[18:53] earthworm ([email protected]) left #gstreamer (""I allege that SCO is full of it." - Linus Torvalds").
[18:57] <jdahlin> BBB, good
[18:57] Action: jdahlin addes one and closes
[18:57] <jdahlin> adds
[18:58] Action: BBB sees loads of 'crassh on record' in 2.2 or 2.4 bugs
[18:58] <BBB> should I dup all those?
[18:58] <jdahlin> Yeah
[18:59] <jdahlin> or, but not on 129164
[18:59] <jdahlin> since that was clearly playing :)
[18:59] <thomasvs> BBB: how do I figure out what formats ffmpeg plays ?
[19:00] jcsston ([email protected]) joined #gstreamer.
[19:05] <thomasvs> BBB: how do  I make you talk without smacking you to bits ?
[19:07] <BBB> get me a job in programming
[19:07] <BBB> erm
[19:07] <BBB> ffmpeg
[19:07] <BBB> gst-inspect 
[19:07] <BBB> | grep ff
[19:07] <BBB> basically
[19:07] <BBB> or ffmpeg -formats
[19:07] <BBB> ffplay -formats should work too
[19:12] <LeRoutier> bbl
[19:12] LeRoutier ([email protected]) left irc: "Leaving"
[19:14] jcsston ([email protected]) left irc: Read error: 60 (Operation timed out)
[19:15] <BBB> bbl
[19:15] BBB ([email protected]) left #gstreamer.
[19:21] teuf ([email protected]) joined #gstreamer.
[19:22] <jdahlin> ds, ping?
[19:25] pb_ ([email protected]) joined #gstreamer.
[19:28] teuf ([email protected]) left irc: Read error: 104 (Connection reset by peer)
[19:29] teuf ([email protected]) joined #gstreamer.
[19:40] Nick change: Ura_gym -> Uraeus
[19:46] bilboed ([email protected]) left irc: Read error: 110 (Connection timed out)
[19:46] spyder482 ([email protected]) joined #gstreamer.
[19:47] bilboed ([email protected]) joined #gstreamer.
[20:06] trow ([email protected]) left irc: "Leaving"
[20:07] BBB ([email protected]) joined #gstreamer.
[20:09] Action: BBB loves gstreamer if it works
[20:09] <jdahlin> Why is there _gst_foo_initialize() for all random types?
[20:09] <jdahlin> it's just so much craaaack
[20:09] <BBB> random?
[20:10] <jdahlin> caps, taglist, structure and a few others
[20:10] <BBB> they're called by gst_init(), right?
[20:10] <jdahlin> yes
[20:10] <BBB> caps/taglist/structure: dunno
[20:10] <jdahlin> but they should be removed, in favor for gst_foo_get_type()
[20:10] <BBB> for the others (bin, element): you need those defined before using them, so you can create them using gst_element_factory_make()
[20:10] Shoragan ([email protected]) left irc: "Leaving"
[20:11] Action: jdahlin is currently trying to clean that mess up
[20:11] <BBB> you're doing a great job, you know? :)
[20:11] <jdahlin> thanks
[20:11] <jdahlin> I hope david won't kill me
[20:11] <BBB> he won't... maybe benjamin will
[20:13] Shoragan ([email protected]) joined #gstreamer.
[20:13] <jdahlin> he can sends his complains after 0.8 <g>
[20:13] <ds-work> jdahlin: boing
[20:15] <jdahlin> ds-work, I'm currently cleaning up many of the headers and classes you've written
[20:15] <jdahlin> ds-work, is there a reason for the _gst_foo_type_initialize()
[20:15] <BBB> did I write any classes in core?
[20:15] Action: BBB doesn't think so
[20:15] <jdahlin> classes/structures
[20:15] Action: BBB randomly donkeytail-picks a subsystem that he'll rewrite for 0.9.x
[20:16] Action: BBB selects gstscheduler
[20:16] <BBB> :o
[20:16] spyder482 ([email protected]) left irc: Read error: 113 (No route to host)
[20:17] spyder482 ([email protected]) joined #gstreamer.
[20:18] <jdahlin> ds-work: and there is no reason marking structures/functions with a _ prefix if they're static
[20:18] <jdahlin> _* is *very* ugly
[20:19] <BBB> we only do that for internal nonstatic functions, don't we?
[20:19] <BBB> else it's a bug
[20:19] <jdahlin> It should only be used if they are used in other source files (eg objects)
[20:20] <jdahlin> at least, that's how it's done in gtk+ for instance
[20:20] <thomasvs> jdahlin: there are two good reasons; not forgetting that they are private, and afaik the symbols don't get exported either
[20:20] <BBB> static functions are never exported
[20:20] <jdahlin> thomasvs: no, as BBB says static means the function is not exported
[20:20] <jdahlin> and, inside an implementation of a class, it shouldn't matter for the caller if the function is private or not
[20:21] <jdahlin> perhaps one day you want to make it public
[20:21] <jdahlin> and then you have to go and rename everything
[20:21] <BBB> feel free to fix it :)
[20:21] Action: BBB off to experiments
[20:21] <BBB> will be back tonight
[20:21] <BBB> if you're all asleep
[20:21] <BBB> bye
[20:21] BBB ([email protected]) left #gstreamer.
[20:22] <thomasvs> we don't only have them static
[20:22] <teuf> solaris linker doesn't export symbols prefixed with _ when creating a library
[20:22] Shoragan ([email protected]) left irc: "Leaving"
[20:23] <jdahlin> teuf: for a library true, but what about between objects in a library?
[20:23] Uraeus ([email protected]) left irc: Remote closed the connection
[20:23] Shoragan ([email protected]) joined #gstreamer.
[20:23] Company ([email protected]) joined #gstreamer.
[20:23] <Company> good evening angels
[20:23] <teuf> jdahlin: the object file within the lib can reference any _ prefixed symbol from other object files who are also part of the lib
[20:24] <Company> oh, jdahlin is online :)
[20:24] <teuf> hi Company
[20:24] Uraeus ([email protected]) joined #gstreamer.
[20:25] <jdahlin> Company: hello
[20:25] <ds-work> jdahlin: I have no naming convention preference for internal functions
[20:25] <Uraeus> thomasvs: do you have any updated rpms ready for me to test?
[20:25] <jdahlin> ds-work: great, they I'll do it the gtk+ style
[20:26] <Company> jdahlin: can we (you with my help) figure out #134007 ?
[20:26] <Company> and the alsa bug belonging to it
[20:26] <thomasvs> Uraeus: they're all in the repo
[20:27] <thomasvs> later guys
[20:27] <spyder482> hmm, someone hilighted me?
[20:29] <jdahlin> Uraeus: did you see my fix in opt? :)
[20:29] <jdahlin> Company: I'm a bit busy at the moment :/
[20:29] <Uraeus> jdahlin: yeah I did, really happy to see you dig into the core 
[20:30] <Company> jdahlin: i won't force you to do it, but it's quite important and my soundcard doesn't trigger it...
[20:30] <Company> jdahlin: you can help me make iain look at it ;)
[20:31] sublett ([email protected]) left irc: "Nice of you to put me on the money"
[20:35] <jdahlin> It's nice to use buildbot to test your changes
[20:37] <jdahlin> My laptop is too slow to build around the clock
[20:40] <Company> ds-work: 
[20:40] <Company> -gst_alsa_timestamp_to_bytes (GstAlsa *this, GstClockTime time)
[20:40] <Company> +gst_alsa_timestamp_to_bytes (GstAlsa * this, GstClockTime time)
[20:40] <Company> ds-work: this is from indenting - can we somehow make it not put a space there?
[20:41] <Company> the other thing i don't like is that it reindents prototypes
[20:42] <Company> -    snd_pcm_access_mask_set (mask, this->mmap ? SND_PCM_ACCESS_MMAP_INTERLEAVED : SND_PCM_ACCESS_RW_INTERLEAVED);
[20:42] <Company> +    snd_pcm_access_mask_set (mask,
[20:42] <Company> +       this->
[20:42] <Company> +       mmap ? SND_PCM_ACCESS_MMAP_INTERLEAVED : SND_PCM_ACCESS_RW_INTERLEAVED);
[20:42] <Company> that's probably a bug in indent
[20:45] thomasvs ([email protected]) left irc: Client Quit
[20:47] <ds-work> personally, I don't like how our prototypes are indented
[20:49] <Company> i like it better if it looks like a table
[20:49] <Company> though i'm not the instance that decides that anyway
[20:50] <ds-work> the problem with indenting like a table is that there are too many exceptions, which make it look like crap in the end
[20:51] <Company> i think it looks like crap if it's not indented
[20:52] <Company> i can't easily figure out the number, type and order of arguments or the return values that way
[20:52] <Company> which is what i'm looking for most of the time
[20:55] sublett ([email protected]) joined #gstreamer.
[20:58] <ds-work> anyone want to write a C parser in perl?
[21:22] jcsston ([email protected]) joined #gstreamer.
[21:24] thomasvs ([email protected]) joined #gstreamer.
[21:25] <thomasvs> any of you l33thax0rs know how to interpret kernel panic led morse code ?
[21:27] <ds-work> yeah
[21:27] <thomasvs> where do I find the info ?
[21:27] <ds-work> btw, seeking is broken in rhytmbox
[21:27] <ds-work> thomasvs: pipe the output through ksymoops
[21:27] <thomasvs> ds-work: oh, you're supposed to write down the complete set of symbols ?
[21:27] <thomasvs> ds-work: is it long ?
[21:28] <ds-work> ah, it's really a panic?
[21:28] <ds-work> not just an oops?
[21:29] <thomasvs> ds-work: the machine is completely locked, and the leds are blinking morse code
[21:30] <ds-work> oh, that...
[21:30] <ds-work> no, I don't know how to interpret that
[21:31] <taaz> what does the morse code decode to?
[21:32] <ds-work> probably %pc at the time of the crash
[21:35] <iain> "Help, I'm fucked up!"
[21:36] <jdahlin> $ grep extern\ GType *.h
[21:36] <jdahlin> (nothing)
[21:36] <jdahlin> woho!
[21:36] <iain> where does gst# exist?
[21:37] <taaz> jdahlin: there was a reason for the _gst_foo_initialize at a point in the past.  it had something to do with omegas effort to make things efficient.  a few types were specially handled to avoid many type checks and that led to some special case of their init code
[21:37] <ds-work> what is the rationale behind hiding GTypes?
[21:37] <taaz> or something like that... you'd have to ask omega or wtay i think
[21:38] <jdahlin> ds-work: they shouldn't be exported
[21:38] <jdahlin> they are private
[21:38] <ds-work> why?
[21:38] <jdahlin> taaz: that's solved with G_GNUC_CONST, I'm quite sure
[21:39] <jdahlin> They're only used by the GST_TYPE_FOO macro
[21:39] <jdahlin> it's done like that in all other GObject libraries
[21:39] Uraeus ([email protected]) left irc: "Client exiting"
[21:39] <ds-work> yes, but _why_ is it done that way?
[21:40] <jdahlin> To try to avoid exporting as little as possible I guess?
[21:40] <jdahlin> we're already exporting gst_foo_get_type and _gst_foo_type right now
[21:40] <ds-work> well, you have to export a function instead
[21:40] <jdahlin> taaz, http://mail.gnome.org/archives/gtk-devel-list/2001-May/msg00174.html
[21:41] <jdahlin> hmm, 
[21:41] <taaz> jdahlin: i really can't recall the exact reasoning behind it.  i'm pretty sure that the type casting avoided all unneeded checks which required those types to be initialized before their use.  it was only a handful of types which were all non-GObject and intended to be very fast and low overhead to use
[21:42] <jdahlin> taaz: currently, for the objects with low overhead, you don't need to use the type macros at all (GstData, GstBuffer etc)
[21:43] <jdahlin> using G_GNUC_CONST seems to remove the overhead of that extra function call anyway
[21:44] <ds-work> one thing I'd like to do in 0.9 is create a GSubObject, which has the type properties of GObject, but much less overhead
[21:45] <jdahlin> ds-work: we should try to profile things first, to make sure that it's not unnecessary work
[21:45] <taaz> ds-work: that might be a nice addition to glib itself
[21:47] <ds-work> jdahlin: for things like GstBuffer, it's in our fast path, so any improvement is good
[21:48] <taaz> by the way, omega is alive in #tbcpdx trying to answer
[21:49] <taaz> 15:48 <omega> yeah, it registers types and puts them in globals
[21:49] <taaz> 15:48 <omega> instead of g_<object}_type being a function call every time
[21:52] <ds-work> if it's done correctly, I think the function call may be faster, or not much slower
[21:54] <jdahlin> hmm, I can't finish this tonight :/
[21:54] Action: jdahlin wanders home
[21:54] <taaz> omega suggests verifying the asm to make sure it does the proper thing
[21:55] <jdahlin> I will try
[21:55] jdahlin ([email protected]) left irc: "Leaving"
[21:58] Action: Company suggests not caring about optimizations until it's a real problem
[21:58] <ds-work> certainly not that one
[22:07] teuf ([email protected]) left irc: "Client exiting"
[22:10] <taaz> iirc, such optimizations did make a noticable difference in latency benchmarks at the time
[22:10] <taaz> probably more noticable on pdas or other slower boxen too
[22:10] bitshifter ([email protected]) left irc: "bye"
[22:14] bitshifter ([email protected]) joined #gstreamer.
[22:17] yippi ([email protected]) joined #gstreamer.
[22:20] smoke ([email protected]) got netsplit.
[22:21] <taaz> getting the docs to build on my sys is now officially a time-hole.
[22:22] <taaz> anyone have any suggestions at all why docbook2ps would invoke jadetex in such a way that it goes off into lala land forever?
[22:23] <taaz> i think i reinstalled all appropriate debian packages with no luck
[22:24] <ds-work> thaytan: thanks for fixing effecTV
[22:26] smoke ([email protected]) got lost in the net-split.
[22:29] Rotty ([email protected]) left irc: "Bye"
[22:33] yippi ([email protected]) left irc: Remote closed the connection
[22:35] <Company> ds-work: it's ok if i change XID to unsigned long in xoverlay, right?
[22:35] crazney ([email protected]) left irc: "Client exiting"
[22:36] <ds-work> Company: yeah
[22:36] Zeenix ([email protected]) joined #gstreamer.
[22:36] <Zeenix> hi
[22:39] yippi ([email protected]) joined #gstreamer.
[22:43] smoke ([email protected]) joined #gstreamer.
[22:45] steveb_ ([email protected]) joined #gstreamer.
[22:47] <taaz> out of curiosity, is XID always the same as a gulong or is it specified by bit size?
[22:48] Action: taaz too lazy to look ;)
[22:48] <Company> taaz: according to /usr/X11R6/include/X.h it's always unsigned long
[22:50] jimmy_dean ([email protected]) left irc: "Carrier Signal Lost"
[22:53] steveb__ ([email protected]) left irc: Read error: 60 (Operation timed out)
[22:54] 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 
[22:54] jimmy_dean ([email protected]) joined #gstreamer.
[22:55] mathrick ([email protected]) joined #gstreamer.
[22:56] <mathrick> yo
[22:57] Company ([email protected]) left irc: Remote closed the connection
[22:59] bitshifter ([email protected]) left irc: Remote closed the connection
[22:59] bitshifter ([email protected]) joined #gstreamer.
[23:01] teuf ([email protected]) joined #gstreamer.
[23:04] Action: mathrick wonders why it takes Evo several minutes to fetch "POP summary", when LIST command over telnet is a matter of 3 seconds for 109 incoming mails
[23:06] kwm ([email protected]) left irc: "Leaving"
[23:17] teuf ([email protected]) left irc: Read error: 54 (Connection reset by peer)
[23:18] teuf ([email protected]) joined #gstreamer.
[23:33] <ds-work> taaz: remember to vote
[23:34] <taaz> ds-work: that doesn't end today does it?
[23:35] <ds-work> taaz: any reason you're waiting for the last minute?
[23:36] <taaz> ds-work: because i haven't decided yet i guess
[23:36] <ds-work> just vote to get rid of non-free :)
[23:36] <taaz> ends the 21st... i've got some time to consider things ;)
[23:37] ProtectR ([email protected]) joined #gstreamer.
[23:40] <taaz> i use at least 1 non-free package.  though i think i didn't realize it was even in debian ;)  chntpw can be quite useful in acedemic world when sysadmins leave and vanish
[23:40] markey ([email protected]) left irc: "g'nite"
[23:41] <taaz> hmm eagle has a package too.  i use that as well.
[23:42] <taaz> speaking of non-free sort of things... what am i to do with gst-ffmpeg?  i still don't know if that code is allowed in debian or not.  it might not be as well accepted as a full package even though it's currently hiding inside of gst-plugins anway
[23:47] harshy ([email protected]) joined #gstreamer.
[23:48] <ds-work> that's one reason why having non-free hosted separately from Debian would be useful
[23:49] <Misirlou> Non-free is SUXXX DIE DIE DIE
[23:57] <taaz> i dunno... if non-free were already hosted somewhere else now and had all the bug tracking and authorization and so on that rest of project has the vote would be easy to many people.  without that _today_ i'm not so sure.
[23:57] <ds-work> it's a chicken and egg problem
[00:00] --- Sat Mar 13 2004
[00:00] <taaz> yes it is
[00:00] <ds-work> the problem that nobody bothers to talk about is that nobody cares to maintain non-free in Debian
[00:00] <taaz> well... not really
[00:00] <ds-work> just individual packages
[00:01] <taaz> it seems it's lower effor to maintain as is then if it were seperate.  but i don't know much about that either way...
[00:02] <ds-work> that's why maintainers of individual packages don't want non-free split off
[00:03] <taaz> or users of those packages
[00:03] <ds-work> users, it's easy
[00:03] <mathrick> nite
[00:04] <ds-work> I have over 10 repositories in my apt configuration
[00:04] <ds-work> 2 are Debian
[00:04] Nick change: mathrick -> mathrick|sleep
[00:05] <taaz> eh... pbuilder failed because fakeroot wasn't installed.  isn't that something installed by default? hmm.
[00:05] <ds-work> er, no
[00:06] <taaz> oh wait... i'm being dumb. nevermind
[00:06] <taaz> copy & paste bug ;)
[00:07] <taaz> so can i even maintain something like ffmpeg using debian resources?
[00:08] <taaz> specifically, is it a problem if i were to manage debs via an arch repository which is mirrored on arch.d.o?
[00:08] <ds-work> why not?
[00:08] <taaz> even though they may end up just being pushed to a debian repo on gst.net
[00:09] <taaz> i can foresee someone not wanting even the code to exist on debian servers
[00:09] <taaz> i'm overthinking this.  i'll wait for someone to actually complain ;)
[00:10] <taaz> manoj has tome in arch on there already...
[00:12] jcsston ([email protected]) left irc: Read error: 110 (Connection timed out)
[00:12] Misirlou ([email protected]) left irc: "asdf"
[00:13] <dilinger> taaz: i'd imagine debian would be fine w/ any free software being hosted w/ its resources
[00:13] <dilinger> oh, ffmpeg is non-free?
[00:13] <taaz> i dunno
[00:14] <dilinger> patent encumbered, or weird license?
[00:14] <taaz> more of ip issues than freeness
[00:15] Action: ds-work comments that jdahlin did a Very Bad Thing changing the API
[00:17] <taaz> what'd he change?
[00:18] <ds-work> gst_caps_is_simple() to GST_CAPS_IS_SIMPLE()
[00:23] <bitshifter> soon we'll be able to write whole plugins using nothing but macros ;)
[00:24] <ds-work> actually, we're strongly moving away from macros
[00:28] <taaz> soon we'll be able to write whole plugins using nothing but python
[00:28] <taaz> ok... maybe not
[00:34] jcsston ([email protected]) joined #gstreamer.
[00:46] sublett ([email protected]) left irc: "Nice of you to put me on the money"
[00:58] yippi ([email protected]) left irc: Remote closed the connection
[01:00] teuf ([email protected]) left irc: Read error: 110 (Connection timed out)
[01:00] <danb> w00t
[01:06] <ProtectR> Hum, I've got a question, not related to PCM, but I though you would maybe have the answer...
[01:07] <bluejay> danb: got gst-python working?
[01:07] <ProtectR> sorry, not realted to gstreamer
[01:07] <ProtectR> was about PCM
[01:07] <ProtectR> where could I find some good documetation (about mathematics behind the format)?
[01:08] <ds-work> ProtectR: uh, there's no mathematics behind PCM
[01:08] <ds-work> ProtectR: it's just evenly-sampled (in time) audio samples
[01:08] <ProtectR> well, I meanabout transformations...
[01:08] <ds-work> what kind of transformations?
[01:09] <ProtectR> well, about slowing a tempo, for example
[01:09] <ds-work> with or without changing pitch?
[01:09] <ProtectR> with
[01:10] <ds-work> there's a few notes in gst-plugins/gst-libs/gst/resample/README
[01:10] smoke ([email protected]) got netsplit.
[01:10] <ProtectR> ok, thanks
[01:11] <ds-work> otherwise, there are web pages that can be found using some of the key words in that README
[01:16] smoke ([email protected]) got lost in the net-split.
[01:16] <ProtectR> thx a lot
[01:17] <danb> bluejay: getting there...
[01:18] <danb> bluejay: jdahlin listened to me complain a lot yesterday and got some stuff fixed
[01:18] <danb> bluejay: the latest cvs commit seemed to fix the main problem (unresolved symbols in the .so files)
[01:19] <danb> bluejay: so now i need to actually try to use it and see what happens
[01:24] <ProtectR> ds-work: links are in the cvs i suppose? (cause I don't see them in the development package...)
[01:25] <ds-work> ProtectR: yeah, it probably doesn't get disted
[01:25] <ds-work> ProtectR: do you have Cvs checked out?
[01:25] mathrick|sleep ([email protected]) left irc: Read error: 60 (Operation timed out)
[01:25] <ProtectR> i'll get one
[01:25] thaytan ([email protected]) left irc: No route to host
[01:26] <ds-work> http://www.schleef.org/~ds/resample
[01:28] <ProtectR> ok, thx
[01:32] yippi ([email protected]) joined #gstreamer.
[01:39] mathrick|sleep ([email protected]) joined #gstreamer.
[01:56] <thomasvs> rock
[01:56] <thomasvs> I have a serbian .po file
[01:57] pb_ ([email protected]) left irc: "Client exiting"
[02:01] <ds-work> ugh, I hate having automake-1.4 on my system
[02:03] BBB ([email protected]) joined #gstreamer.
[02:10] ProtectR ([email protected]) left irc: Remote closed the connection
[02:23] Nick change: kh_zZz -> khalek
[02:27] sub_pop ([email protected]) left irc: "Client exiting"
[02:32] lilo ([email protected]) left irc: "Lost terminal"
[02:32] lilo ([email protected]) joined #gstreamer.


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