Original commit message from CVS:
* gst/tcp/gsttcpclientsrc.c: (gst_tcp_client_src_create):
Revert Andy's newsegment change pending a more correct
fix.
Original commit message from CVS:
:
* gst/typefind/gsttypefindfunctions.c: (q3gp_type_find),
(qt_type_find), (plugin_init):
detect more files as 3gp
group and reorder the iso file formats
Original commit message from CVS:
* gst-libs/gst/tag/gstvorbistag.c: (gst_tag_from_vorbis_tag),
(gst_tag_to_vorbis_tag):
Make sure we called gst_tag_register_musicbrainz_tags()
before possibly mapping a vorbiscomment string from/to a
musicbrainz tag.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (mp3_type_find):
In case we can't find the required number of consecutive
mpeg audio frames to positively identify an MPEG audio
stream, check if there's at least a valid mpeg audio
frame right at offset 0 and if so suggest mpeg/audio
caps with a very low probability (#153004).
Original commit message from CVS:
2006-02-07 Andy Wingo <wingo@pobox.com>
* gst/tcp/gsttcpclientsrc.c (gst_tcp_client_src_create): Switch to
a TIME segment if we get timestamped buffers. Requires recent
fixes in core to work properly.
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (prepare_output):
Don't print the URI as part of the error message, it
makes error dialogs look rather ugly, especially if
the URI is very long or has characters in it that
need escaping.
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (prepare_output):
Error out if we have only text or subtitles, but nothing
else. Also error out if we have subtitles but no video
stream.
Original commit message from CVS:
* ext/gnomevfs/gstgnomevfssrc.c: (gst_gnome_vfs_src_create):
Treat GNOME_VFS_RESULT_EOF as EOS, not as error (#329194).
Post an error message on the bus when we encounter an
error, which will hopefully be more meaningful than the
'Internal Flow Error' message users get to see if we
just return GST_FLOW_ERROR.
Original commit message from CVS:
2006-02-07 Andy Wingo <wingo@pobox.com>
* configure.ac (GST_MAJORMINOR): Update core version req to
0.10.2.2, for the collectpads API addition (#330244).
Original commit message from CVS:
* ext/gnomevfs/gstgnomevfs.c: (plugin_init):
Return FALSE from plugin_init() when GnomeVFS can't
be initialised for some reason (#328423).
Original commit message from CVS:
2006-02-06 Julien MOUTTE <julien@moutte.net>
* ext/pango/gsttextoverlay.c: (gst_text_overlay_src_event):
Stick to seeking theory until i find the bug.
* gst/subparse/gstsubparse.c: (parse_subrip): Fix debug.
Original commit message from CVS:
* ext/theora/theoraenc.c: (gst_theora_enc_class_init),
(theora_enc_finalize), (theora_enc_sink_setcaps),
(theora_set_header_on_caps), (theora_enc_chain),
(theora_enc_change_state):
* tests/check/pipelines/theoraenc.c: (GST_START_TEST):
Make theoraenc and the tests leak free. Like, really.
Original commit message from CVS:
(theora_enc_finalize), (theora_enc_sink_setcaps):
Add a finalize method to ensure we clean up state even if
someone omitted the state change back to NULL.
* ext/vorbis/vorbisenc.c: (gst_vorbisenc_metadata_set1),
(gst_vorbisenc_chain):
Free some more leaked bits.
* tests/check/pipelines/theoraenc.c: (start_pipeline),
(stop_pipeline):
Wait for state changes to happen if they're ASYNC.
This ought to teach those fancy pants buildbots a lesson.
Original commit message from CVS:
* docs/libs/compiling.sgml:
Fix typo: it's pkg-config, not pkg-gconfig
* docs/libs/gst-plugins-base-libs-docs.sgml:
* docs/libs/gst-plugins-base-libs-sections.txt:
* docs/libs/tmpl/gstgconf.sgml:
There is no libgstgconf in 0.10, remove it
from the docs.
Original commit message from CVS:
* gst-libs/gst/tag/gstvorbistag.c: (gst_vorbis_tag_add):
Add comment about LANGUAGE tag inconsistency (we want
ISO-639-1, but extract three-letter identifiers?)
* po/POTFILES.in:
Add two translatable files.
Original commit message from CVS:
* gst-libs/gst/tag/Makefile.am:
* gst-libs/gst/tag/gstvorbistag.c: (gst_vorbis_tag_add):
* gst-libs/gst/tag/tag.h:
* gst-libs/gst/tag/tags.c:
(gst_tag_register_musicbrainz_tags_internal),
(gst_tag_register_musicbrainz_tags):
Forward-port some tags stuff from the 0.8 branch. This is
mostly the addition of musicbrainz tags and their mapping
to vorbistags, and a vorbistag mapping of the language tag.
Original commit message from CVS:
* gst/playback/gstdecodebin.c: (try_to_link_1):
Don't put essential function call into
g_return_*() macro, otherwise it'll all be
replaced by NOOPs when compiling with
G_DISABLE_CHECKS defined.
Original commit message from CVS:
* ext/alsa/gstalsasink.c: (gst_alsasink_finalise),
(gst_alsasink_class_init), (gst_alsasink_init),
(gst_alsasink_write), (gst_alsasink_reset):
* ext/alsa/gstalsasink.h:
Add lock to protect alsa calls.
Implement reset to flush samples ASAP, does not work
with dmix though.
Original commit message from CVS:
* gst-libs/gst/audio/gstbaseaudiosink.c:
(gst_base_audio_sink_provide_clock),
(gst_base_audio_sink_set_property),
(gst_base_audio_sink_get_property), (gst_base_audio_sink_render):
Don't try to provide a clock when we are not negotiated since
we might not be able to make it run.
Original commit message from CVS:
* gst-libs/gst/audio/TODO:
Updated.
* gst-libs/gst/audio/gstbaseaudiosink.c:
(gst_base_audio_sink_drain), (gst_base_audio_sink_event):
On EOS, wait till the last sample is played before posting EOS.
Original commit message from CVS:
2006-02-01 Philippe Kalaf <burger at speedy dot org>
* gst-libs/gst/rtp/gstbasertpdepayload.c:
Patch by Kai Vehmanen : Adds ability to enable newsegment bypass by
setting queue_delay to zero. Also avoid thread being started if
queue_delay is zero.
Original commit message from CVS:
* gst/playback/test6.c: (new_decoded_pad_cb), (show_error), (main):
Make test work again by connecting fakesinks to each decoded pad,
which makes the pipeline wait until each fakesink has a buffer
queued before going to PAUSED state. At that point we know the
decodebin pads are negotiated.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (au_type_find),
(paris_type_find), (ilbc_type_find), (plugin_init):
Fix typefinding for audio/x-au, audio/x-paris and
audio/iLBC-sh. We cannot use the START_WITH macros
here, because there can only be one typefind factory
with the same name (caps), so the second one would
replace the first one and the first one would never
be called when doing typefinding (see #161712).
Original commit message from CVS:
* ext/vorbis/vorbisdec.c: (vorbis_dec_convert),
(vorbis_handle_header_packet), (vorbis_dec_push),
(vorbis_handle_data_packet):
Use scale_int when we can, add some more scaling.
Check packettype before parsing it.
Original commit message from CVS:
* ext/theora/theoradec.c: (_theora_granule_time),
(theora_dec_src_convert), (theora_dec_sink_convert):
Call right _scale functions.
Use parameter instead of some other random value.
Original commit message from CVS:
* ext/theora/theoradec.c: (_theora_granule_frame),
(_theora_granule_time), (_inc_granulepos),
(theora_dec_src_convert), (theora_dec_sink_convert),
(theora_handle_type_packet), (theora_handle_data_packet),
(theora_dec_chain):
Use higher precision timestamps calculation.
Convert some other conversions to _scale.
Original commit message from CVS:
* gst/audiotestsrc/gstaudiotestsrc.c:
(gst_audio_test_src_create_sine_table), (plugin_init):
* gst/volume/gstvolume.c: (plugin_init):
initialize gst_controller before using
Original commit message from CVS:
* tests/check/pipelines/theoraenc.c:
* tests/check/pipelines/vorbisenc.c:
Define constant using G_GINT64_CONSTANT to avoid errors when
passing it around - otherwise it gets truncated to 32 bits.
Fixes failing tests.
Original commit message from CVS:
2006-01-31 Andy Wingo <wingo@pobox.com>
* sys/v4l/gstv4lsrc.c (gst_v4lsrc_set_caps): Don't segfault if the
caps being set doesn't have a framerate value. Basically a stopgap
measure.
* ext/ogg/gstoggmux.c (GST_BUFFER_END_TIME): New macro. Not
technically correct enough to put into core though.
(gst_ogg_mux_dequeue_page): Use END_TIME instead of TIMESTAMP +
DURATION. Fixes theoraenc ! oggmux.
* sys/v4l/gstv4lsrc.c (gst_v4lsrc_fixate): Fixate to the nearest
fraction, not double.
Original commit message from CVS:
2006-01-30 Andy Wingo <wingo@pobox.com>
* ext/ogg/gstoggmux.c (gst_ogg_mux_dequeue_page): Compare
timestamp + duration, not just timestamp -- ogg pages should be
ordered by stop time. Necessary fix given the change in vorbis
timestamps.
Original commit message from CVS:
2006-01-30 Andy Wingo <wingo@pobox.com>
* ext/theora/theoraenc.c (theora_enc_sink_setcaps)
(gst_theora_enc_init): Pull the granule shift out of the encoder.
(granulepos_add): New function, handles the messiness of adjusting
granulepos values.
(theora_buffer_from_packet):
(theora_enc_chain):
(theora_enc_sink_event): Use granulepos_add, not +.
* tests/check/pipelines/theoraenc.c
(check_buffer_granulepos_from_starttime): Just check the frame
count, not the actual granulepos -- we can't dictate to the
encoder when it should be placing keyframes.
Original commit message from CVS:
* ext/gnomevfs/gstgnomevfssrc.c: (gst_gnome_vfs_src_start):
SERVICE_NOT_AVAILABLE happens for example when you're trying to
play an http:// stream from a server that's not serving
Original commit message from CVS:
2006-01-30 Andy Wingo <wingo@pobox.com>
* tests/check/pipelines/vorbisenc.c (TIMESTAMP_OFFSET):
* tests/check/pipelines/theoraenc.c (TIMESTAMP_OFFSET): Totally
remove the UINT64_CONSTANT macro, doesn't appear to be needed or
available.
Original commit message from CVS:
2006-01-30 Andy Wingo <wingo@pobox.com>
* ext/theora/gsttheoraenc.h:
* ext/theora/theoraenc.c: Same changes as were done to vorbisenc,
although theoraenc was timestamping correctly. Added handling of
streams that start with nonzero timestamps.
* tests/check/Makefile.am:
* tests/check/pipelines/theoraenc.c: New file, basically does same
tests as vorbisenc.
* tests/check/pipelines/vorbisenc.c: I claim these bugs.
Original commit message from CVS:
* gst-libs/gst/audio/gstaudiosink.c:
(gst_audioringbuffer_class_init), (gst_audioringbuffer_release),
(gst_audioringbuffer_pause):
Implement pause that does not wait for completion.
* gst-libs/gst/audio/gstbaseaudiosink.c:
(gst_base_audio_sink_render), (gst_base_audio_sink_change_state):
Don't drop buffers when going to PAUSED but perform preroll on
remaining samples now that core base class supports this.
* gst-libs/gst/audio/gstringbuffer.c: (gst_ring_buffer_release),
(gst_ring_buffer_pause_unlocked), (gst_ring_buffer_stop),
(gst_ring_buffer_commit):
Pause should not signal waiters.
Implement return value of _commit correctly.
Original commit message from CVS:
2006-01-30 Andy Wingo <wingo@pobox.com>
* tests/check/Makefile.am (check_vorbis): Add pipelines/vorbisenc.
* ext/vorbis/vorbisenc.c (gst_vorbisenc_buffer_from_packet): Logic
updated to timestamp from the first sample, not the last.
(gst_vorbisenc_buffer_from_header_packet): New function, takes
special care of granulepos and timestamp for header packets.
(gst_vorbisenc_chain): Reflow, fix some leaks, and handle the case
when the first buffer has a nonzero timestamp.
* ext/vorbis/vorbisenc.h (GstVorbisEnc.granulepos_offset)
(GstVorbisEnc.subgranule_offset): New members. Take care of the
case when the first audio buffer we get has a nonzero timestamp.
(GstVorbisEnc.next_ts): Renamed from prev_ts, because now we
properly timestamp vorbis buffers with the time of the first
sample, not the last.
* ext/vorbis/vorbisenc.c (granulepos_to_clocktime): Renamed from
vorbis_granule_time_copy -- now it takes the granule/subgranule
offset into account.
* tests/check/pipelines/vorbisenc.c: New test for correctness of
timestamps, durations, and granulepos on buffers produced by
vorbisenc.
Original commit message from CVS:
* gst/ffmpegcolorspace/gstffmpegcodecmap.c:
(gst_ffmpeg_pixfmt_to_caps), (gst_ffmpeg_caps_to_pixfmt):
Patch from Eric Jonas to support conversions to/from UYVY
(Fixes: #324626)
Original commit message from CVS:
* gst-libs/gst/audio/audio.h: (GST_CLOCK_TIME_TO_FRAMES)
* gst-libs/gst/audio/gstbaseaudiosink.c: (gst_base_audio_sink_render)
use of gst_guint64_to_gdouble to be compliant with vs6
* gst/playback/gstdecodebin.c: (try_to_link_1)
* gst/videorate/videorate.c: (gst_video_rate_blank_data)
use of G_GINT64_CONSTANT for int64 constants
* win32/common/libgstinterfaces.def:
export some symbols (gst_mixer_get_type,gst_mixer_track_get_type)
* win32/vs6:
update and add new project files
Original commit message from CVS:
* Makefile.am:
* win32/MANIFEST:
* win32/common/interfaces-enumtypes.c:
(gst_color_balance_type_get_type), (gst_mixer_type_get_type),
(gst_mixer_track_flags_get_type),
(gst_tuner_channel_flags_get_type):
* win32/common/interfaces-enumtypes.h:
* win32/common/multichannel-enumtypes.c:
(gst_audio_channel_position_get_type):
* win32/common/multichannel-enumtypes.h:
add a win32-update rule like in core, and copy over enumtypes files
Original commit message from CVS:
* ext/alsa/gstalsasink.c: (gst_alsasink_init), (set_hwparams),
(set_swparams), (gst_alsasink_prepare), (gst_alsasink_unprepare),
(gst_alsasink_close), (gst_alsasink_write), (gst_alsasink_reset):
* ext/alsa/gstalsasrc.c: (gst_alsasrc_init), (set_hwparams),
(set_swparams), (gst_alsasrc_open), (gst_alsasrc_prepare),
(gst_alsasrc_unprepare), (gst_alsasrc_read):
Update all error messages. All of them should either use
the default translated message, or actually provide a
translatable string.
Make the string for channel count problems meaningful.
Original commit message from CVS:
* gst/videoscale/vs_scanline.c: Oops, *that's* why I never
checked in this change -- it requires liboil features not
in 0.3.6. Revert parts.
Original commit message from CVS:
* ext/libvisual/visual.c: (get_buffer):
When pad_alloc returns a GstFlowReturn other
than GST_FLOW_OK, make sure it is passed upstream.
Original commit message from CVS:
* ext/alsa/gstalsasink.c: (gst_alsasink_finalise),
(gst_alsasink_class_init):
Free the device name string.
* ext/ogg/gstoggmux.c: (gst_ogg_mux_class_init),
(gst_ogg_mux_request_new_pad), (gst_ogg_mux_release_pad),
(gst_ogg_mux_handle_src_event), (gst_ogg_mux_clear_collectpads):
Don't remove a pad from the collectpads structure until it
is released - it's a request pad, and may receive data again
if the element gets moved back to PLAYING state.
* sys/xvimage/xvimagesink.c: (gst_xvimagesink_get_xv_support):
Ensure we turn on double buffering on the Xv port, and
set the colour key to something dark and mysterious that
isn't black.
Original commit message from CVS:
* ext/alsa/gstalsaplugin.c: (plugin_init):
* ext/cdparanoia/gstcdparanoiasrc.c:
(gst_cd_paranoia_src_base_init), (plugin_init):
* ext/gnomevfs/gstgnomevfs.c: (plugin_init):
* ext/ogg/gstoggdemux.c: (gst_ogg_demux_plugin_init):
- a library should not call setlocale. see Libraries node in
gettext manual
- make sure all plugins that use translation do bindtextdomain
to point to the localedir
* gst/playback/gstplaybin.c: (gen_vis_element), (add_sink),
(setup_sinks), (plugin_init):
all this, and check for NULL when creating sinks
Original commit message from CVS:
2006-01-27 Julien MOUTTE <julien@moutte.net>
* gst/subparse/gstsubparse.c: (gst_subparse_type_find),
(plugin_init): Make typefinding of subtitles work again.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (aac_type_find),
(mp3_type_frame_length_from_header), (mp3_type_find),
(wavpack_type_find), (m4a_type_find), (ircam_type_find),
(plugin_init):
Backport a bunch of typefinding fixes from the 0.8 branch.
Also, improve wavpack typefinding: if we can't peek the
entire wavpack block, try to parse the bits we can get and
see if we find what we're looking for in those.
Original commit message from CVS:
2006-01-26 Julien MOUTTE <julien@moutte.net>
* sys/ximage/ximagesink.c:
(gst_ximagesink_calculate_pixel_aspect_ratio):
* sys/xvimage/xvimagesink.c:
(gst_xvimagesink_calculate_pixel_aspect_ratio): Handle some
more cases of pixel aspect ratio.
Original commit message from CVS:
* gst/playback/gstdecodebin.c: (pad_probe):
Also consider the flush-start and tag events as unblockers
for the pad probes.
Original commit message from CVS:
2006-01-26 Julien MOUTTE <julien@moutte.net>
* gst/playback/gstplaybin.c: (gst_play_bin_init),
(gst_play_bin_dispose), (gst_play_bin_vis_unblocked),
(gst_play_bin_vis_blocked), (gst_play_bin_set_property):
On the fly visualisation switch, works disabling, enabling as
well but it won't be able to enable vis in a playbin that was
created with no visualisation.
Original commit message from CVS:
* gst-libs/gst/audio/gstbaseaudiosink.c:
(gst_base_audio_sink_render), (gst_base_audio_sink_change_state):
Undo previous commit, it breaks resume after pause.
Original commit message from CVS:
* gst-libs/gst/audio/gstbaseaudiosink.c:
(gst_base_audio_sink_setcaps), (gst_base_audio_sink_event),
(gst_base_audio_sink_preroll), (gst_base_audio_sink_render):
Improve debugging.
Post error when caps cannot be parsed.
Resync on discontinuity in the stream.
Clip samples to segment boundaries.
return WRONG_STATE sooner when we are flushing.
* gst-libs/gst/audio/gstbaseaudiosrc.c: (gst_base_audio_src_init),
(gst_base_audio_src_get_time), (gst_base_audio_src_create):
Make audiosrc operate in TIME.
Set TIMESTAMP and DURATION on buffers.
Original commit message from CVS:
* gst/playback/gstdecodebin.c: (gst_decode_bin_init),
(free_pad_probes), (remove_fakesink), (pad_probe),
(close_pad_link), (gst_decode_bin_change_state):
Replace GstPadBlockCallback with pad probes that detect
first buffer AND eos before removing fakesink.
Fixes hang with demuxers doing EOS while pre-rolling.
Solves #328279
Original commit message from CVS:
2006-01-22 Julien MOUTTE <julien@moutte.net>
* sys/ximage/ximagesink.c: (gst_ximagesink_ximage_destroy):
* sys/xvimage/xvimagesink.c: (gst_xvimage_buffer_destroy),
(gst_xvimagesink_buffer_alloc): Playbin keeps some ref to some
frames. We might get a frame destroyed after changing state to
NULL, adding a safety check on xcontext.
Original commit message from CVS:
* gst-libs/gst/interfaces/xoverlay.c:
Fix prepare-xwindow-id code example in the docs - we need to
ignore all messages that aren't element messages as well.
Original commit message from CVS:
2006-01-21 Julien MOUTTE <julien@moutte.net>
* sys/xvimage/xvimagesink.c: (gst_xvimagesink_buffer_alloc):
I think one day i'll completely undestand how caps negotiation
is supposed to work. This refactoring handles buffer_alloc
called with caps we can't handle. We definitely don't want a
set_caps with those caps, so we define and allocate a buffer
we would like to receive.
Original commit message from CVS:
* gst-libs/gst/audio/gstbaseaudiosink.c:
(gst_base_audio_sink_render):
Fix playback of non-synchronised streams by assuming a rate
of 1.0 instead of a random one.
Makes this work again:
gst-launch filesrc location=raw_audio.file ! 'audio/x-raw-int,
endianness=(int)4321, signed=(boolean)true, width=(int)16,
depth=(int)16, rate=(int)44100, channels=(int)2' ! audioconvert !
audioresample ! alsasink
Original commit message from CVS:
* gst/playback/gststreaminfo.c: (gst_stream_info_set_mute):
* gst/playback/gststreamselector.c:
(gst_stream_selector_set_property):
Comment out broken code that connects to the state-changed signal.
At this point, changing current stream selection is broken, but
stuff like gst-launch playbin current-audio=1 works and filters
to the chosen stream.
Original commit message from CVS:
* ext/theora/theoradec.c: (theora_handle_comment_packet):
Post taglist actually on bus instead of just freeing it
(fixes#327114 and totem bug #327080).
* ext/vorbis/vorbisdec.c: (vorbis_handle_comment_packet):
Use gst_element_found_tags_for_pad(), so that the tags
are sent downstream as an event as well.
Original commit message from CVS:
* sys/ximage/ximagesink.c: (gst_ximage_buffer_finalize),
(gst_ximagesink_ximage_new), (gst_ximagesink_ximage_put),
(gst_ximagesink_buffer_alloc):
* sys/xvimage/xvimagesink.c: (gst_xvimage_buffer_finalize),
(gst_xvimagesink_xvimage_put), (gst_xvimagesink_show_frame),
(gst_xvimagesink_buffer_alloc):
move all regularly occurring messages to GST_LOG level
add some more object logs
Original commit message from CVS:
2006-01-14 Thomas Vander Stichele <thomas at apestaart dot org>
* ext/ogg/gstoggmux.c: (gst_ogg_mux_collected):
fix a silly segfault
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (setup_source):
Fix playback for sources that emit raw audio or
raw video streams (e.g.: cd audio sources) (#325984).
Original commit message from CVS:
* gst-libs/gst/cdda/gstcddabasesrc.c:
(gst_cdda_base_src_handle_track_seek):
No need to post a tag message on the bus when seeking
within the same track, only post it when the current
track changes.
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (group_destroy),
(probe_triggered), (new_decoded_pad), (mute_group_type),
(set_active_source):
* gst/playback/gststreaminfo.c: (gst_stream_info_set_mute):
* gst/playback/gststreamselector.c:
(gst_stream_selector_base_init),
(gst_stream_selector_set_property),
(gst_stream_selector_request_new_pad):
Reenable stream selection. These mechanisms need a complete overhaul
in the face of 0.8->0.10 changes though.
Original commit message from CVS:
* ext/ogg/gstoggdemux.c:
Change the pad template to src_%d to match the pads that
are created from it. decodebin needs this information in order
to decide that oggdemux is capable of producing multiple pads
(and hence needs queues inserted).
* ext/ogg/gstoggmux.c: (gst_ogg_mux_queue_pads),
(gst_ogg_mux_collected):
Make debug output more useful by using GST_PTR_FORMAT.
Original commit message from CVS:
Reviewed by: Tim-Philipp Müller <tim at centricular dot net>
* gst-libs/gst/audio/gstringbuffer.c: (gst_ring_buffer_parse_caps):
Set depth and width for alaw/mulaw (fixes#326601).
Original commit message from CVS:
* ext/cdparanoia/gstcdparanoiasrc.c: (gst_cd_paranoia_src_close),
(gst_cd_paranoia_paranoia_callback):
Don't try to free NULL pointers.
Original commit message from CVS:
* gst/audiorate/gstaudiorate.c: (gst_audio_rate_chain),
(gst_audio_rate_change_state), (plugin_init):
Add debugging category.
Fix type issues.
Add case for incoming buffers without valid offset/offset_end.
Original commit message from CVS:
* gst/playback/gstplaybin.c: (gen_audio_element):
Don't leak an autoaudiosink/alsasink when we generate
a new audio element. (old code, I guess)
Original commit message from CVS:
* gst/audiorate/gstaudiorate.c: (gst_audio_rate_setcaps):
Support float audio in audiorate.
Use width rather than depth for selecting sample width.