Original commit message from CVS:
2004-01-01 Ronald Bultje <rbultje@ronald.bitfreak.net>
* configure.ac:
Fix configure check for mpeg2enc. We need 1.6.1.93 instead of
1.6.1.92, since the pkg-config file of 1.6.1.92 is borked and
it therefore uses the wrong include paths. Too bad... Note
that 1.6.1.93 is not release yet. ;).
Also add a check for mplex, which is now using the lib'ified
mplex from mjpegtools, too.
* ext/ffmpeg/gstffmpegcodecmap.c:
Add codec_tag for 3ivx/xvid. For xvid, this should fix playback
issues. I don't think ffmpeg handles 3ivx correctly, so this
probably won't work. But it won't hurt either.
* ext/ffmpeg/gstffmpegdec.c: (gst_ffmpegdec_connect),
(gst_ffmpegdec_chain):
* ext/ffmpeg/gstffmpegenc.c: (gst_ffmpegenc_connect),
(gst_ffmpegenc_chain_audio):
Fix memleak in audio encoding. Close codec if open fails, this
calls the cleanup routines so we can re-use the context.
* ext/mpeg2enc/gstmpeg2enc.cc:
Fix pad template names/types, fix memory issue with getcaps().
* ext/mpeg2enc/gstmpeg2encoder.cc:
* ext/mpeg2enc/gstmpeg2encoder.hh:
Fix compile issue with new caps system (const thingy).
* ext/mpeg2enc/gstmpeg2encpicturereader.cc:
* ext/mpeg2enc/gstmpeg2encpicturereader.hh:
We read a first frame right on initing, so that we have a caps
when we init the output. This caps is cached in padprivate and
read as first frame.
* ext/mplex/Makefile.am:
* ext/mplex/gstmplex.cc:
* ext/mplex/gstmplex.h:
* ext/mplex/gstmplex.hh:
* ext/mplex/gstmplexibitstream.cc:
* ext/mplex/gstmplexibitstream.hh:
* ext/mplex/gstmplexjob.cc:
* ext/mplex/gstmplexjob.hh:
* ext/mplex/gstmplexoutputstream.cc:
* ext/mplex/gstmplexoutputstream.hh:
We wrap mjpegtools mplex. So I rewrote the plugin. The old plugin
had issues, didn't do capsnego, supported only a subset of the
mplex features and required a mplex fork in our local CVS. Plus
that it worked agaist a very old mplex version. Rewriting was
faster than updating it.
* gst-libs/ext/Makefile.am:
* gst-libs/ext/mplex/INSTRUCT:
* gst-libs/ext/mplex/Makefile.am:
* gst-libs/ext/mplex/README:
* gst-libs/ext/mplex/TODO:
* gst-libs/ext/mplex/ac3strm_in.cc:
* gst-libs/ext/mplex/audiostrm.hh:
* gst-libs/ext/mplex/audiostrm_out.cc:
* gst-libs/ext/mplex/aunit.hh:
* gst-libs/ext/mplex/bits.cc:
* gst-libs/ext/mplex/bits.hh:
* gst-libs/ext/mplex/buffer.cc:
* gst-libs/ext/mplex/buffer.hh:
* gst-libs/ext/mplex/fastintfns.h:
* gst-libs/ext/mplex/format_codes.h:
* gst-libs/ext/mplex/inputstrm.cc:
* gst-libs/ext/mplex/inputstrm.hh:
* gst-libs/ext/mplex/lpcmstrm_in.cc:
* gst-libs/ext/mplex/mjpeg_logging.cc:
* gst-libs/ext/mplex/mjpeg_logging.h:
* gst-libs/ext/mplex/mjpeg_types.h:
* gst-libs/ext/mplex/mpastrm_in.cc:
* gst-libs/ext/mplex/mpegconsts.cc:
* gst-libs/ext/mplex/mpegconsts.h:
* gst-libs/ext/mplex/mplexconsts.hh:
* gst-libs/ext/mplex/multplex.cc:
* gst-libs/ext/mplex/outputstream.hh:
* gst-libs/ext/mplex/padstrm.cc:
* gst-libs/ext/mplex/padstrm.hh:
* gst-libs/ext/mplex/stillsstream.cc:
* gst-libs/ext/mplex/stillsstream.hh:
* gst-libs/ext/mplex/systems.cc:
* gst-libs/ext/mplex/systems.hh:
* gst-libs/ext/mplex/vector.cc:
* gst-libs/ext/mplex/vector.hh:
* gst-libs/ext/mplex/videostrm.hh:
* gst-libs/ext/mplex/videostrm_in.cc:
* gst-libs/ext/mplex/videostrm_out.cc:
* gst-libs/ext/mplex/yuv4mpeg.cc:
* gst-libs/ext/mplex/yuv4mpeg.h:
* gst-libs/ext/mplex/yuv4mpeg_intern.h:
* gst-libs/ext/mplex/yuv4mpeg_ratio.cc:
We don't fork mjpegtools' mplex in our CVS anymore.
* gst/avi/gstavidemux.c: (gst_avi_demux_src_getcaps),
(gst_avi_demux_add_stream):
* gst/avi/gstavidemux.h:
Add getcaps() function for proper caps nego. This makes some
parts of AVI playback/reading work.
* sys/ximage/ximagesink.c: (gst_ximagesink_sinkconnect):
Resize window on new capsnego. This is probably wrong, but
I'm still committing it because with current capsnego, the
first successfull capsnego is auto-fixated, therefore rounded
down to the lowest values in the caps. this results in a 16x16
XWindow that is not reized when real capsnego finishes.
Dave, I see more cases of this, do you know a proper solution?
* tools/gst-launch-ext.in:
Fix MPEG-4 AAC (Apple iPod/iTunes) file commandline.
Original commit message from CVS:
2003-12-22 Benjamin Otte <in7y118@public.uni-hamburg.de>
* gst/qtdemux/qtdemux.c: (plugin_init):
qtdemux requires bytestream
Original commit message from CVS:
2003-12-21 Ronald Bultje <rbultje@ronald.bitfreak.net>
* configure.ac:
Improve mpeg2enc detection. This is for distributions that do
ship mjpegtools, but without mpeg2enc. Also does object check
for might there ever be ABI incompatibility.
* ext/mpeg2enc/gstmpeg2enc.cc:
Add Andrew as second maintainer (he's helping me), and also add
an error if no caps was set. This happens if I pull before capsnego
and that's something I should solve sometime else.
* gst/matroska/matroska-demux.c:
(gst_matroska_demux_parse_blockgroup):
Fix time parsing.
* gst/matroska/matroska-mux.c: (gst_matroska_mux_audio_pad_link),
(gst_matroska_mux_track_header):
Add caps to templates.
* gst/mpegaudioparse/gstmpegaudioparse.c: (mp3_sink_factory):
Add mpegversion=1 to prevent confusion with MPEG/AAC.
* gst/mpegstream/gstmpegdemux.c:
Remove layer since it causes warnings about unfixed caps.
* gst/videotestsrc/gstvideotestsrc.c: (gst_videotestsrc_get):
Fix obvious typo (we error out if caps were set, we should of
course error out if *no* caps were set).
* sys/oss/gstosselement.c: (gst_osselement_convert):
Fix format conversion, we confused bits/bytes.
* sys/oss/gstosselement.h:
Improve documentation for 'bps'.
* sys/v4l/TODO:
Remove stuff about plugins that need removing - this was done
ages ago.
* sys/v4l/gstv4lmjpegsrc.c: (gst_v4lmjpegsrc_init),
(gst_v4lmjpegsrc_src_convert), (gst_v4lmjpegsrc_src_query):
* sys/v4l/gstv4lsrc.c: (gst_v4lsrc_init), (gst_v4lsrc_src_convert),
(gst_v4lsrc_src_query):
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_init),
(gst_v4l2src_src_convert), (gst_v4l2src_src_query):
Add get_query_types(), get_formats() and query() functions.
Original commit message from CVS:
Sorry Dave... Add mpegversion=1 to mp3 caps everywhere so that the autoplugger uses mad and not faad for mp3 decoding. This should fix mp3 playback.
Original commit message from CVS:
Riff, EBML, fourcc etc. work. Not fully finished, but better than
what we used to have and definately worth a first broad testing.
I've revived rifflib. Rifflib used to be a bytestream-for-riff, which
just dup'ed bytestream. I've rewritten rifflib to be a modern riff-
chunk parser that uses bytestream fully, plus adds some extra functions
so that riff file parsing becomes extremely easy. It also contains some
small usability functions for strh/strf and metadata parsing. Note that
it doesn't use the new tagging yet, that's a TODO.
Avidemux has been rewritten to use this. I think we all agreed that
avidemux was pretty much a big mess, which is because it used all
sort of bytestream magic all around the place. It was just ugly.
This is a lot nicer, very complete and safe. I think this is far more
robust than what the old avidemux could ever have been. Of course, it
might contain bugs, please let me know.
EBML writing has also been implemented. This is useful for matroska.
I'm intending to modify avidemux (with a riffwriter) similarly. Maybe
I'll change wavparse/-enc too to use rifflib.
Lastly, several plugins have been modified to use rifflib's fourcc
parsing instead of their own. this puts fourcc parsing in one central
place, which should make it a lot simpler to add new fourccs. We might
want to move this to its own lib instead of rifflib.
Enjoy!
Original commit message from CVS:
tagging stuff and build fixes. In detail:
- make gdk-pixbuf loader work when distchecking
- fix invalid syntax in ffmpeg Makefile. wildcards for EXTRA_DIST are not allowed. This broke builds where distdir != srcdir
- fix ffmpeg cvs grabbing when srcdir != distdir
- new id3tag plugin for id3 tag reading/writing (uses mad's libid3tag)
- mad and libid3tag require mad/libid3tag v0.15. Fixed configure to require that
- added ogg demuxer in ext/ogg. The demuxer does not handle events yet. Especially getting seeking right will require some effort or code copying from libvorbis.
- added raw vorbis detection to typefinding. oggdemux requires a typefind function to detect its contents.
- tags plugin in gst/tags. Provides API in <gst/tags/gsttagediting.h>. API includes tag matching GStreamer <=> ID3 and GStreamer <=> vorbis and writing/reading vorbiscomments or ID3v1 tags. Also included is a simple vorbiscomment reader/writer. Writing will not really work though until someone writes oggmux.
- various build fixes. Mostly missing (DIST)CLEANFILES.
- vorbisenc handles tag writing.
Now it's YOUR turn to fix and write more plugins that handle writing/reading of tags. :)
Original commit message from CVS:
Remove all config.h includes from header files, add it to each source file and remove duplicate config.h includes from several source files
Original commit message from CVS:
Add a local copy of riff.h as we don't use rifflib anymore.
Rewrite the main loop to use bytestreams instead of rifflib. Make it a loopbased
filter.
Handle metadata, cues and labels as well
Original commit message from CVS:
merge TYPEFIND branch. Major changes:
- totally reworked type(find) system
- all typefind functions are in gst/typefind now
- more typefind functions then before
- some plugins might fail to compile now because I don't have them installed and they
a) require bytestream or
b) haven't had their typefind fixed.
Please fix those plugins and put the typefind functions into gst/typefind if they don't have dependencies
Original commit message from CVS:
Add a parser for metadata
(demux_metadata): Given the buffer containing the metadata, look through it and
get the info out of it.
(wav_new_chunk_callback): Change the if statement to a switch statement.
Handle GST_RIFF_TAG_LIST by changing what the type of list it is and parsing out
metadata if it is "INFO".
Original commit message from CVS:
New typefind system:
* bytestream is now part of the core
* all plugins have been modified to use this new typefind system
* asf typefinding added
* mpeg video stream typefiding removed because it's broken
* duplicate typefind entries removed
* extra id3 typefinding added, because we've seen 4 types of files
(riff/wav, flac, vorbis, mp3) with id3 headers and each of these needs
to work. Instead, I've added an id3 element and let it redo typefiding
after the id3 header. this needs a hack because spider only typefinds
once. We can remove this hack once spider supports multiple typefinds.
* with all this, mp3 typefinding is semi-rewritten
* id3 typefinding in flac/vorbis is removed, it's no longer needed
* fixed spider and gst-typefind to use this, too.
* Other general cleanups
Original commit message from CVS:
Set keyframe flag only when it is a keyframe. This will break quite some formats because many encoders dont set this flag correctly, but I'll fix that as I encounter them. divx5 works correctly now. ;).
Original commit message from CVS:
In the intfloat conversion elements, there were bugs when the float buffer was
bigger than the int buffer (in frames). That's now fixed, some style issues are
"fixed" (subjectively), mallocs are removed from int2float's processing loop,
and the default buffer size was raised to 256 frames to give better performance in
the default case.
Wavenc was modified to not set an event handler on its sink pad. It deals
with EOS in its chain function instead.
Original commit message from CVS:
Fix padding (2 bytes instead of 4), improve index offset calculation, fix some minor other issues and make avimux actually work with mp3 audio
Original commit message from CVS:
Fixes to make it compile without GNOME, and with a modern (>= 0.3)
version of GStreamer. Now that I got it compiled, I want to delete
it.
Original commit message from CVS:
Added #include of config.h so that they symbol "VERSION" could be found.
Also removed GST_CAT_EVENT from a GST_DEBUG line to reflect recent changes
in the debug logic. Now this plugin compiles.
Original commit message from CVS:
New mimetypes gone into effect today - this commit changes all old mimetypes over to the new mimetypes spec as described in the previous commit's document. Note: some plugins will break, some pipelines will break, expect HEAD to be broken or at least not 100% working for a few days, but don't forget to report bugs
Original commit message from CVS:
compatibility fix for new GST_DEBUG stuff.
Includes fixes for missing includes for config.h and unistd.h
I only ensured for plugins I can build that they work, so if some of them are still broken, you gotta fix them yourselves unfortunately.
Original commit message from CVS:
This adds width/height properties to qtdemux, so that it connects to ffdec_*... still doesn't work, because the buffer allocation in ffdec_ allocs too small buffers (edge emulation failure or so?), which causes a segfault. I'm working on that
Original commit message from CVS:
Handle compressed headers. Fix inappropriate use of bytestream_flush().
Code cleanup. Added getcaps and _link functions for src pads. Extract
and set the size,rate,channels correctly. Fix some of the caps to
agree with avidemux and/or ffmpeg.
Original commit message from CVS:
Actually push the event over to the next element instead of leaving it into eternity - thanks for Joshua for detecting
Original commit message from CVS:
Fix avimux (somewhat). Note: the EOS event still doesn't get through. This is completely braindead, I have no clue why, but setting this thing to PAUSE should do the trick too. EOS fix will come later on...
Original commit message from CVS:
Made the SWAP and PREPARE_3D_LINE macros work with gints rather than
using typeof(), since typeof() is a gcc-extension and does not work
with other compilers. This is okay since every place these macros
are used, gints are passed in. I renamed SWAP to SWAP_INT to reflect
that it is not so generic.
Original commit message from CVS:
Corrected the configure.ac so it actually works. Updated some c files
so that they build on Solaris. This mostly involved supporting ISO
style variable-argument macros.
Original commit message from CVS:
Updated autogen.sh/configure.ac and various Makefiles to make the
configure script set up all gcc specific compiler arguments, rather
than hardcoding them in the Makefile.am files
Original commit message from CVS:
Fix typo (incorrect pointer dereference). Change the magic number
for little-endian streams to match with /etc/magic.
Original commit message from CVS:
bugfixes:
- seek correctly on SEEK_METHOD_END
- don't emit a warning when mp3's in wav's have no width information
- use BYTES format on discontinuous events and omit timestamp when invalid (mp3 anyone?)
Original commit message from CVS:
fixes to wavparse:
- endianness is G_LITTLE_ENDIAN and not G_BYTE_ORDER
- support the law formats
- error out on unknown format, don't try to go on (fixes#110516)
- check buffer size before typefinding
