Original commit message from CVS:
* gst/rtsp/gstrtspsrc.c: (gst_rtspsrc_send), (gst_rtspsrc_open),
(gst_rtspsrc_close), (gst_rtspsrc_play), (gst_rtspsrc_pause),
(gst_rtspsrc_change_state):
* gst/rtsp/gstrtspsrc.h:
* gst/rtsp/rtspdefs.c: (rtsp_method_as_text), (rtsp_find_method):
* gst/rtsp/rtspdefs.h:
* gst/rtsp/rtsptransport.c: (rtsp_transport_parse):
Handle RTSP defaults better.
Issue OPTIONS request to figure out what we are allowed to do.
Make the methods a bitfield so we can easily collect supported
options.
Fix rtsp_find_method.
Do proper RTSP connection shutdown.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_stream),
(gst_avi_demux_stream_header):
* gst/qtdemux/qtdemux.c: (gst_qtdemux_change_state),
(gst_qtdemux_add_stream), (qtdemux_parse_tree):
Uncomment metadata and codec-name handling.
Original commit message from CVS:
2005-08-16 Andy Wingo <wingo@pobox.com>
* ext/raw1394/gstdv1394src.c (gst_dv1394src_iso_receive): Note
license info in the source code -- was only in the commit log
before.
* ext/dv/gstdvdec.h:
* ext/dv/gstdvdec.c: Only decodes systemstream=FALSE dv video --
old pipelines using dvdec should probably have a dvdemux first.
* ext/dv/gstdvdemux.h:
* ext/dv/gstdvdemux.c: Split out from dvdec, chunks the incoming
systemstream=TRUE data into frames, sets caps data, and spits out
PCM audio in addition to systemstream=FALSE video frames. Operates
in chain mode only for now; should make a getrange version as
well.
* ext/dv/gstdv.c: New file, registers the libgstdv plugin.
* ext/dv/Makefile.am: Library name changed to libgstdv. Split
dvdec into dvdemux and dvdec.
Original commit message from CVS:
* ext/mad/Makefile.am:
* gst/avi/Makefile.am:
* gst/effectv/Makefile.am:
* gst/udp/Makefile.am:
* gst/wavparse/Makefile.am:
Use -lgstfoo-@GST_MAJORMINOR@ instead of -lgstfoo-0.9
Original commit message from CVS:
* ext/jpeg/gstjpegdec.c: (gst_jpeg_dec_decode_indirect),
(gst_jpeg_dec_decode_direct), (gst_jpeg_dec_chain):
Fix decoding of pictures with certain uneven or unaligned
widths where jpeglib needs more horizontal padding than our
I420 buffers provide, resulting in blocky artifacts at the
left side of the picture (#164176).
Also make use of our shiny new GST_ROUND_N() macros.
Original commit message from CVS:
* ext/jpeg/gstjpegdec.c: (gst_jpeg_dec_init), (gst_jpeg_dec_chain),
(gst_jpeg_dec_change_state):
* ext/jpeg/gstjpegdec.h:
Fix crashes/invalid memory access for pictures that have a height
that is not a multiple of 16 (or rather: v_samp_factor * DCTSIZE).
Also fix the state change function for downwards state changes
(need to chain up to parent before destroying our resources, to
make sure pads get deactivated and our chain function isn't
running and using those very same resources in another thread).
The jpeg line buffer only needs to be v_samp_factor*DCTSIZE lines
per plane, not picture_height lines; allocate that on the stack.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_stream_headers),
(gst_wavparse_stream_data):
Add some fixes from 0.8 branch: allow 24/32bps songs and
blockalign samples to the header-specified size, if any
(#311070); error out on channels==0 or bitrate==0
(#309043, #304588).
Original commit message from CVS:
* ext/faad/gstfaad.c: (gst_faad_class_init), (gst_faad_setcaps):
Add debug category, remove Close() call that made it crash
whenever reusing, renegotiating or anything; Close() actually
free()s the handle and should only be called on READY->NULL.
* gst/qtdemux/qtdemux.c: (gst_qtdemux_loop_header):
Actually set caps on buffer (in addition to pad), also.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_stream_header):
Fix AVI header parsing: add missing break statement after
GST_RIFF_INFO_LIST parsing code; gst_riff_read_chunk() has
already advanced the avi->offset, no need to do it twice
(fixes MovieOfMovies.avi).
Original commit message from CVS:
* ext/jpeg/gstjpegdec.c: (gst_jpeg_dec_init),
(gst_jpeg_dec_setcaps), (gst_jpeg_dec_chain),
(gst_jpeg_dec_change_state):
* ext/jpeg/gstjpegdec.h:
Make mjpeg actually work and skip jpeg data parsing if we
know that the input is packetized (ie. each input buffer
is exactly one jpeg frame).
Original commit message from CVS:
* ext/mad/gstmad.c: (gst_mad_init), (gst_mad_chain):
It'd be nice if I could listen to my mp3 files, so send out an
initial discont, as the sink apparently wants.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_handle_src_event),
(gst_avi_demux_handle_seek):
Fix seeking (or, well, fix threading issue where a variable was
set before a lock was taken and was already unset before that
same lock was taken and was thus no longer in existance when it
actually had to be used).
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_process_next_entry):
Mixing binary and logical operators is not going to work; fix
position-querying in Totem.
Original commit message from CVS:
* ext/faad/gstfaad.c: (gst_faad_base_init), (gst_faad_class_init),
(gst_faad_init), (gst_faad_setcaps), (gst_faad_srcgetcaps),
(gst_faad_event), (gst_faad_update_caps), (gst_faad_chain),
(gst_faad_change_state):
* ext/faad/gstfaad.h:
Fix negotiation (#310932) and miscellaneous other stuff. Probably
still needs some more work.
Original commit message from CVS:
* ext/jpeg/gstjpegdec.c: (gst_jpeg_dec_init),
(gst_jpeg_dec_setcaps), (gst_jpeg_dec_chain):
Add setcaps() function (for mjpeg).
Original commit message from CVS:
2005-08-08 Andy Wingo <wingo@pobox.com>
* ext/esd/esdsink.c (gst_esdsink_getcaps): Seems that wierd
va_list caps setting function was borked. Fixed esdsink.
Original commit message from CVS:
2005-08-08 Andy Wingo <wingo@pobox.com>
* sys/oss/gstosssink.c (gst_oss_sink_open, gst_oss_sink_close)
(gst_oss_sink_prepare, gst_oss_sink_unprepare): Update for newer
audiosink api.
* ext/raw1394/gstdv1394src.c (gst_dv1394src_get_property)
(gst_dv1394src_set_property): Style. All about the style.
* ext/esd/esdsink.c (gst_esdsink_getcaps): Return specific caps
only if in READY or higher (i.e., if _open() has been called.)
(gst_esdsink_open, gst_esdsink_close, gst_esdsink_prepare)
(gst_esdsink_unprepare): Update for audiosink changes.
(gst_esdsink_change_state): Die!
Original commit message from CVS:
2005-08-04 Andy Wingo <wingo@pobox.com>
* sys/oss/gstossaudio.c (plugin_init): Second-class citizen.
* gst/videobox/gstvideobox.c (gst_video_box_get_size): Update for
API changes.
* configure.ac (DEFAULT_AUDIOSINK, DEFAULT_VIDEOSINK): Set to
autoaudiosink and autovideosink.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_reset),
(gst_avi_demux_parse_stream), (gst_avi_demux_process_next_entry):
You need to allocatate (len+1) characters to store a len size string.
Also don't stop the processing task if the output pad is not linked.
Original commit message from CVS:
* configure.ac:
* ext/Makefile.am:
* ext/libpng/Makefile.am:
* ext/libpng/gstpng.c:
* ext/libpng/gstpngenc.c:
Ported pngenc , still have to port pngdec...
Original commit message from CVS:
* ext/mad/gstmad.c: (gst_mad_src_event):
First try forwarding events, makes seeking in AVI files with mp3
audio work again.
Original commit message from CVS:
2005-07-19 Edgard N. A. G. Lima <edgard.lima@indt.org.br>
* configure.ac
* ext/Makefile.am
* ext/amrnb/amrnbdec.c
* ext/amrnb/amrnbenc.c
* ext/amrnb/amrnbparse.c
* ext/faad/gstfaad.c
* ext/mpeg2dec/gstmpeg2dec.c
Ported amrnb, faad, mpeg2dec to 0.9
Original commit message from CVS:
* configure.ac:
* gst/wavparse/gstwavparse.c:
* gst/wavparse/gstwavparse.h:
* gst/wavparse/Makefile.am:
Ported wavparse to 0.9 . Playing, seeking and state changes work.
Could need more loving on the headers though.
Original commit message from CVS:
2005-07-19 Andy Wingo <wingo@pobox.com>
* ext/dv/gstdvdec.c (gst_dvdec_decode_video): Set the proper
framerate on the outbound buffer.
Original commit message from CVS:
2005-07-19 Andy Wingo <wingo@pobox.com>
* ext/dv/gstdvdec.c (gst_dvdec_decode_video): Don't clobber
alloc_buffer's return value.
(gst_dvdec_decode_frame): Handle unlinked pads with grace and
agility.
Original commit message from CVS:
* gst/realmedia/rmdemux.c: (gst_rmdemux_sink_event),
(gst_rmdemux_change_state), (gst_rmdemux_chain),
(gst_rmdemux_get_stream_by_id), (gst_rmdemux_send_event),
(gst_rmdemux_add_stream):
Send discont event before pushing first buffer.
Original commit message from CVS:
2005-07-16 Philippe Khalaf <burger@speedy.org>
* gst/fdsrc/gstfdsrc.c:
* gst/fdsrc/gstfdsrc.h:
* gst/fdsrc/Makefile.am:
Moved fdsrc 0.9 port from gstreamer/gst/elements to here.
Original commit message from CVS:
* gst/level/gstlevel.c: (gst_level_transform):
* gst/videobox/gstvideobox.c: (gst_video_box_class_init),
(gst_video_box_get_size), (gst_video_box_transform):
Port to new base class.
Original commit message from CVS:
2005-07-08 Andy Wingo <wingo@pobox.com>
* gst/avi/Makefile.am (libgstavi_la_CFLAGS): No gettext hacks, the
defines come from config.h.
* autogen.sh: Run autopoint, etc.
* Makefile.am (DIST_SUBDIRS, SUBDIRS): Go into po/.
* configure.ac: Add gettext stuff.
Original commit message from CVS:
* gst/videobox/gstvideobox.c: (gst_video_box_init),
(gst_video_box_transform_caps), (gst_video_box_set_caps):
Logic was reversed. Needs some more fixes in the transform
function to include AYUV output.
Moved AYUV as prefered format.
Original commit message from CVS:
* gst/base/gstbasesrc.c: (gst_base_src_get_range),
(gst_base_src_default_negotiate), (gst_base_src_negotiate):
Allow subclasses to implement their own negotiation.
Original commit message from CVS:
2005-07-05 Andy Wingo <wingo@pobox.com>
* gst/videobox/gstvideobox.c: Clean up, port to 0.9, use
BaseTransform.
* gst/videobox/Makefile.am: Link to base libs, include
plugins-base cflags, dist the README.
* configure.ac (GST_PLUGIN_ALL, AC_CONFIG_FILES): Add videobox to
the build.
Original commit message from CVS:
2005-07-04 Andy Wingo <wingo@pobox.com>
* examples/level/:
* examples/level/Makefile.am:
* examples/level/README:
* examples/level/demo.c:
* examples/level/plot.c: Examples moved out of the source dir. Not
updated tho.
* configure.ac: Add level to the build.
* gst/level/Makefile.am:
* gst/level/gstlevel.h:
* gst/level/gstlevel.c: Cleaned up, ported to 0.9.
Original commit message from CVS:
2005-07-04 Andy Wingo <wingo@pobox.com>
* ext/aalib/gstaasink.c (gst_aasink_fixate): Update for newer
fixate prototype.
Original commit message from CVS:
* gst/udp/Makefile.am:
* gst/udp/gstudp.c:
* gst/udp/gstdynudpsink.c: (new)
* gst/udp/gstdynudpsink.h: (new)
Added new element (udpdynsink) that receives GstNetBuffers and sends the
udp packets to the source given in the buffer. It's used by rtpsession
element for now.
* gst/udp/gstudpsrc.c:
Fixed memory leak.
Original commit message from CVS:
2005-07-01 Jan Schmidt <thaytan@mad.scientist.com>
* configure.ac:
* ext/mad/Makefile.am:
* sys/oss/Makefile.am:
Roll gstreamer-interfaces-0.9.pc into gstreamer-plugins-base-0.9.pc
Original commit message from CVS:
2005-07-01 Jan Schmidt <thaytan@mad.scientist.com>
* ext/libcaca/Makefile.am:
* ext/mad/Makefile.am:
* gst/effectv/Makefile.am:
* gst/udp/Makefile.am:
Replace GST_PLUGINS_LIBS_* with GST_PLUGINS_BASE_*
* ext/mad/gstid3tag.c: (gst_id3_tag_src_query),
(gst_id3_tag_src_event), (gst_id3_tag_sink_event),
(gst_id3_tag_chain), (plugin_init):
* ext/mad/gstmad.c: (gst_mad_src_query), (gst_mad_chain):
Signedness warning fix, use gst_pad_get_peer instead of GST_PAD_PEER
in querying and event handling, because we're not holding the pad
lock and the peer may disappear.
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_subindex),
(gst_avi_demux_parse_index), (gst_avi_demux_massage_index):
Signedness warning fixes.
* gst/videofilter/gstvideotemplate.c: (plugin_init):
Remove gst_library_load
Original commit message from CVS:
* gst/avi/Makefile.am: (libgstavi_la_LIBADD):
Added linking to libgstriff-0.9
* ext/mad/gstmad.c: (gst_mad_src_query):
check the format of the upstream query and return query if it's the
same format as the requested one.
Original commit message from CVS:
2005-06-29 Andy Wingo <wingo@pobox.com>
* gst/videofilter/gstvideoexample.c: Removed gst_library_load, I
think. Whatever this plugin actually does, that I don't know.
Original commit message from CVS:
* gst/rtsp/gstrtspsrc.c: (gst_rtsp_proto_get_type),
(gst_rtspsrc_class_init), (gst_rtspsrc_create_stream),
(gst_rtspsrc_add_element), (gst_rtspsrc_set_state),
(gst_rtspsrc_stream_setup_rtp),
(gst_rtspsrc_stream_configure_transport), (find_stream),
(gst_rtspsrc_loop), (gst_rtspsrc_open), (gst_rtspsrc_play),
(gst_rtspsrc_change_state):
Fix case where outpad could not be decided.
Original commit message from CVS:
* ext/mad/gstid3tag.c: (gst_id3_tag_src_link):
* ext/mad/gstmad.c: (gst_mad_chain):
RPAD_ -> PAD
Fix args in bufferalloc function call.
Makes the mad plugin compile again
Original commit message from CVS:
2005-06-02 Andy Wingo <wingo@pobox.com>
* ext/mad/gstid3tag.c: Finish porting to 0.9: no more gstdata,
check for link functions before calling them, give
gst_message_new_tag its own copy of the tag list, set the parser
state before sending the event (because in 0.9 events are
processed immediately), casting fixes.
* ext/mad/Makefile.am (libgstmad_la_LDFLAGS): Link with
gsttagedit.
Original commit message from CVS:
* ext/mad/gstmad.c: (gst_mad_chain), (gst_mad_change_state):
* ext/sidplay/gstsiddec.cc:
* gst/alpha/gstalpha.c: (gst_alpha_chain):
* gst/goom/gstgoom.c: (gst_goom_chain):
No need to take the lock anymore, core already did
that before calling us.
Original commit message from CVS:
* Makefile.am:
* ext/Makefile.am:
* sys/Makefile.am:
Make my automake version shut up about undefined variables
* gst/goom/gstgoom.c:
GstAdapter moved to base objects.
Original commit message from CVS:
* configure.ac:
* gst/alpha/Makefile.am:
* gst/alpha/gstalpha.c: (gst_alpha_class_init), (gst_alpha_init),
(gst_alpha_sink_setcaps), (gst_alpha_chain):
Ported alpha, remove alphacolor as functionality is in
ffmpegcolorspace.
Original commit message from CVS:
Move core plugins out of core. I don't mind fdsrc/fdsink
going back into the core; they were just disabled there, so
I moved them. Some of this stuff could (should) be deleted.
* gst/oldcore/Makefile.am:
* gst/oldcore/gstaggregator.c:
* gst/oldcore/gstaggregator.h:
* gst/oldcore/gstelements.c:
* gst/oldcore/gstfdsink.c:
* gst/oldcore/gstfdsink.h:
* gst/oldcore/gstfdsrc.c:
* gst/oldcore/gstfdsrc.h:
* gst/oldcore/gstmd5sink.c:
* gst/oldcore/gstmd5sink.h:
* gst/oldcore/gstmultifilesrc.c:
* gst/oldcore/gstmultifilesrc.h:
* gst/oldcore/gstpipefilter.c:
* gst/oldcore/gstpipefilter.h:
* gst/oldcore/gstshaper.c:
* gst/oldcore/gstshaper.h:
* gst/oldcore/gststatistics.c:
* gst/oldcore/gststatistics.h:
Original commit message from CVS:
* ext/mad/gstmad.c: (gst_mad_init), (gst_mad_src_query),
(index_seek), (normal_seek), (gst_mad_sink_event):
Port to new query API and replace gst_pad_convert()
and gst_pad_get_formats() usage. gstid3tag looks like
it needs some more love before it will work again, if
not a rewrite.
Original commit message from CVS:
Fixed a few things to enable the mad and effectv to be able to find the headers in the gst-plugins-base/gst-libs and to link against the libs in there.
Original commit message from CVS:
* sys/oss/gstosselement.c: (gst_osselement_class_probe_devices):
Kick the hell out of gcc for not warning me about a symbol conflict.
Original commit message from CVS:
Since dirac 0.5.0 the framerate in dirac is expressed as a rational number. Fix build and up requirement to 0.5.0, and also pass parameters to gst_diracdec_link in the right order. (fixes#167959)
Original commit message from CVS:
* gst-libs/gst/riff/riff-media.c:
(gst_riff_create_audio_caps_with_data):
Do actually fix invalid RIFF fmt header values for alaw
and mulaw audio instead of just saying so.
* gst/wavparse/gstwavparse.c: (gst_wavparse_fmt):
Give gst_riff_create_audio_caps_with_data() a chance to
fix up broken format header fields before extracting any
parameters from the header. (fixes#167633)
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_invert):
Declare variables at beginning of block and make gcc-2.95 happy
(fixes # 167482, patch by Gergely Nagy).
* gst/tcp/gsttcpclientsrc.c:
* gst/tcp/gsttcpclientsrc.h:
Move some includes into the header, so that struct sockaddr_in is
defined when it should be defined on FreeBSD as well (fixes
#167483).
* gst/tcp/gsttcpserversrc.c: (gst_tcpserversrc_init_receive):
Don't pass uninitialised values to setsockopt() here either.
Original commit message from CVS:
* sys/ximage/ximagesink.c: (gst_ximagesink_chain),
(gst_ximagesink_send_pending_navigation),
(gst_ximagesink_navigation_send_event), (gst_ximagesink_finalize),
(gst_ximagesink_init):
* sys/ximage/ximagesink.h:
* sys/xvimage/xvimagesink.c: (gst_xvimagesink_chain),
(gst_xvimagesink_send_pending_navigation),
(gst_xvimagesink_navigation_send_event),
(gst_xvimagesink_finalize), (gst_xvimagesink_init):
* sys/xvimage/xvimagesink.h:
Use a mutex protected list to marshal navigation
events into the stream thread from whichever thread
sends them.
Original commit message from CVS:
Add query function to GstSpeed, so that the stream length and current position get adjusted when queried (note that current position queries may still be wrong if the audio sink returns values based on buffer timestamps instead of passing on the query
Original commit message from CVS:
* gst/audioconvert/gstchannelmix.h:
include missing header file
* gst/audioconvert/gstchannelmix.c:
(gst_audio_convert_fill_compatible):
use same sign for both channels when converting to/from compatible
channel. Previously used different signs made the signals cancel
each other out and appear like silence. (fixes#167269)
Original commit message from CVS:
Don't send 'Hey! You gave me a NULL pointer yo naughty person' as error message when we can't open the DVD device; send something more useful instead (fixes#167117)
Original commit message from CVS:
2005-02-11 Julien MOUTTE <julien@moutte.net>
* sys/xvimage/xvimagesink.c: (gst_xvimagesink_xvimage_put),
(gst_xvimagesink_sink_link), (gst_xvimagesink_change_state),
(gst_xvimagesink_chain), (gst_xvimagesink_buffer_free),
(gst_xvimagesink_buffer_alloc), (gst_xvimagesink_set_xwindow_id),
(gst_xvimagesink_expose), (gst_xvimagesink_set_property),
(gst_xvimagesink_finalize), (gst_xvimagesink_init): Protect interface
methods from chain and negotiation and vice versa (Fixes#166142).
Fix a possible bug of images in the buffer pool being discarded because
we are looking at the wrong geometry.
* sys/xvimage/xvimagesink.h: Add stream_lock.
Original commit message from CVS:
* gst/librfb/Makefile.am: Testing stuff before committing is
for wimps... and people with fast machines. Fix stupid
mistake.
Original commit message from CVS:
* configure.ac: Pull in librfb from my CVS tree, because it is
too small and annoying to be separate. Move rfbsrc plugin
to gst/.
* ext/Makefile.am:
* ext/librfb/Makefile.am:
* ext/librfb/gstrfbsrc.c:
* gst/librfb/Makefile.am:
* gst/librfb/gstrfbsrc.c:
* gst/librfb/rfb.c:
* gst/librfb/rfb.h:
* gst/librfb/rfbbuffer.c:
* gst/librfb/rfbbuffer.h:
* gst/librfb/rfbbytestream.c:
* gst/librfb/rfbbytestream.h:
* gst/librfb/rfbcontext.h:
* gst/librfb/rfbdecoder.c:
* gst/librfb/rfbdecoder.h:
* gst/librfb/rfbutil.h:
Original commit message from CVS:
* ext/dts/gstdtsdec.c: (gst_dtsdec_init), (gst_dtsdec_channels),
(gst_dtsdec_handle_event), (gst_dtsdec_handle_frame),
(gst_dtsdec_chain), (gst_dtsdec_change_state):
* ext/dts/gstdtsdec.h:
Don't clobber the stack constructing the channels array.
Make the element chain-based. DTS tracks can now be played.
Original commit message from CVS:
* testsuite/gst-lint: Check for non-statically scoped
parent_class variables. This won't be a problem once
plugins are loaded with RTLD_LOCAL.
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* ext/libpng/gstpngdec.c: (gst_pngdec_init), (gst_pngdec_chain):
* ext/libpng/gstpngenc.c:
Fix byte-order, use proper fixed caps. Fixes#164197.
Original commit message from CVS:
Include "_stdint.h" instead of <stdint.h>. Fixes build on systems that do not have stdint.h, like Solaris 9 (fixes#166631).
Original commit message from CVS:
* sys/ximage/ximagesink.c: (gst_ximagesink_change_state):
* sys/xvimage/xvimagesink.c: (gst_xvimagesink_xwindow_clear),
(gst_xvimagesink_change_state):
Clear window on PAUSED->READY instead of READY->PAUSED. Stop
Xv video (and thereby regenerate Xv colourkey) in clear() so
that PLAY -> READY -> PLAY works (fixes#162504).
Original commit message from CVS:
* sys/v4l/gstv4lmjpegsrc.c: (gst_v4lmjpegsrc_getcaps):
Switch to list instead of range, since MJPEG-devices really just
support decimations, not any size.
Original commit message from CVS:
* ext/mpeg2dec/gstmpeg2dec.c: (gst_mpeg2dec_open_decoder),
(gst_mpeg2dec_reset), (free_all_buffers),
(gst_mpeg2dec_alloc_buffer), (handle_sequence):
* ext/mpeg2dec/gstmpeg2dec.h:
The libmpeg2 user-allocated buffer management is awkward,
to say the least. Hopefully this fixes things.
Original commit message from CVS:
2005-02-04 Andy Wingo <wingo@pobox.com>
* gst/audioconvert/bufferframesconvert.c
(buffer_frames_convert_fixate): New function, fixates to 256
frames per buffer by default. (Much better than 1.)
(buffer_frames_convert_init): Set the fixate function for both src
and sink pad.
(buffer_frames_convert_link): After success setting nonfixed caps,
get the negotiated caps so we can know how many buffer-frames it
will be. No idea how this worked at all before.
Original commit message from CVS:
* ext/mpeg2dec/gstmpeg2dec.c: (gst_mpeg2dec_init),
(gst_mpeg2dec_close_decoder), (put_buffer), (check_buffer),
(free_buffer), (free_all_buffers), (gst_mpeg2dec_alloc_buffer),
(handle_sequence), (handle_picture):
* ext/mpeg2dec/gstmpeg2dec.h:
Rearrange buffer tracking and refcounting and refactor
a little for readability.
Original commit message from CVS:
* sys/qcam/gstqcamsrc.c: (gst_qcamsrc_change_state),
(gst_qcamsrc_open):
Use GST_ELEMENT_ERROR, not g_warning, if open failed.
