Original commit message from CVS:
* ext/mad/gstmad.c: (gst_mad_handle_event):
Fix for when the first format in a discont event is not a
byte-based one. Should fix#137710.
Original commit message from CVS:
* ext/shout2/gstshout2.c:
use application/ogg instead of application/x-ogg (patch by Patrick
Guimond, fixes#142432)
* sys/oss/gstosselement.c: (gst_osselement_reset),
(gst_osselement_sync_parms):
don't set fragment size unless specified
Original commit message from CVS:
* autogen.sh:
* configure.ac:
* ext/mad/gstid3tag.c: (gst_id3_tag_chain):
compute offsets correctly for internal buffers so timestamps are set
correctly when we can't seek. Also handle cases where there are no
offsets. (based on a patch by David Moore, fixes#142507)
Original commit message from CVS:
* ext/alsa/gstalsasink.c: (gst_alsa_sink_loop):
use correct variable when determining amount of data to skip so we
don't skip into the void and segfault
Original commit message from CVS:
* gst/asfdemux/gstasfdemux.c:
- fix a mem leak and always propagate tags
- add WMV3 to known video codecs (but no decoder yet)
- replace "surplus data" at end of audio header for what
it is : codec specific data
- fix a typo
Original commit message from CVS:
reviewed by: Ronald Bultje <rbultje@ronald.bitfreak.net>
* gst-libs/gst/audio/audioclock.c:
Fix wrong return type (#142205).
Original commit message from CVS:
* ext/alsa/gstalsamixer.c: (gst_alsa_mixer_open),
(gst_alsa_mixer_close), (gst_alsa_mixer_supported),
(gst_alsa_mixer_build_list), (gst_alsa_mixer_free_list),
(gst_alsa_mixer_change_state), (gst_alsa_mixer_list_tracks),
(gst_alsa_mixer_get_volume), (gst_alsa_mixer_set_volume),
(gst_alsa_mixer_set_mute), (gst_alsa_mixer_set_record):
Fix for cases where we fail to attach to a mixer.
Original commit message from CVS:
* gst/asfdemux/gstasfdemux.c :
- freeing the event also frees its associated taglist,
that's unfortunate as we need taglist after that. fixed
Original commit message from CVS:
* gst/asfdemux/gstasfdemux.c: (gst_asf_demux_process_comment):
- process comments even if they don't end with \0\0
g_convert would ignore them if present and works well without them
Original commit message from CVS:
* gst/asfdemux/gstasfdemux.c: (gst_asf_demux_process_comment):
don't write to memory we might not write to - g_convert does that
for us anyway
(gst_asf_demux_audio_caps):
conmment out gst_util_dump_mem
Original commit message from CVS:
* ext/alsa/gstalsasink.c: (gst_alsa_sink_loop):
compute correct expected timestamps after seek (broken since
last commit)
* ext/gdk_pixbuf/pixbufscale.c: (pixbufscale_init):
rename element and debugging category to gdkpixbufscale
Original commit message from CVS:
* ext/alsa/gstalsasink.c: (gst_alsa_sink_loop):
add error checking to snd_pcm_delay and remove duplicate call to
snd_pcm_delay that caused issues (see inline code comments)
* ext/alsa/gstalsasink.c: (gst_alsa_sink_get_time):
make more readable and fix return value when snd_pcm_delay fails
Original commit message from CVS:
* ext/esd/esdsink.c: (gst_esdsink_chain): Fix crash when ESD
is killed while we're playing.
* gst/qtdemux/qtdemux.c: (qtdemux_parse): call
gst_element_no_more_pads().
Original commit message from CVS:
* gst-libs/gst/riff/riff-read.c :
- fix INFO tag extraction in RIFF/AVI files
because gst_event_unref (event) also freed taglist
- avoid a mem leak
Original commit message from CVS:
* ext/mad/gstid3tag.c : move from "Codec/(Dem/M)uxer" to "Codec/(Dem/M)uxer/Audio"
* gst/wavenc/gstwavenc.c : move from "Codec/Encoder/Audio" to "Codec/Muxer/Audio"
* gst/auparse/gstauparse.c :
- add code (commented for now) to support audio/x-adpcm on src pad
(we have no decoder for those layout yet)
* gst/cdxaparse/gstcdxaparse.c :
* gst/cdxaparse/gstcdxaparse.h :
- partial rewrite using RiffRead (ripped iain's wavparse code)
* gst/rtp/gstrtpL16enc.c : typo
* gst/rtp/gstrtpgsmenc.c : typo
Original commit message from CVS:
* gst-libs/gst/riff/riff-ids.h :
- add CDXA to the list of RIFF types
- add plst (playlist ?) to wav chunk list (only diff with wavparse/riff.h)
Original commit message from CVS:
* ext/audiofile/gstafsrc.c: (gst_afsrc_get):
Remove old debug output
* ext/dv/gstdvdec.c: (gst_dvdec_quality_get_type),
(gst_dvdec_class_init), (gst_dvdec_loop), (gst_dvdec_change_state),
(gst_dvdec_set_property), (gst_dvdec_get_property):
Change the quality setting to an enum, so it works from gst-launch
Don't renegotiate a non-linked pad. Allows audio only decoding.
* gst/deinterlace/gstdeinterlace.c: (gst_deinterlace_getcaps),
(gst_deinterlace_link), (gst_deinterlace_init):
* gst/videodrop/gstvideodrop.c: (gst_videodrop_getcaps),
(gst_videodrop_link):
Some caps negotiation fixes
Original commit message from CVS:
* ext/tarkin/gsttarkin.c :
- Change RANK from NONE to PRIMARY (decoder)
* ext/gdk_pixbuf/gstgdkpixbuf.c :
- Change RANK from NONE to MARGINAL (decoder)
* ext/divx/gstdivxenc.c :
- Change RANK from PRIMARY to NONE (encoder/spider issue)
Original commit message from CVS:
* configure.ac:
enable shout2 by default
* ext/shout2/gstshout2.c: (gst_shout2send_protocol_get_type),
(gst_shout2send_base_init), (gst_shout2send_init),
(gst_shout2send_connect), (gst_shout2send_change_state):
* ext/shout2/gstshout2.h:
make this work again. Based on a patch by Zaheer Merali (fixes
#142262)
* ext/theora/theora.c: (plugin_init):
don't set rank on encoders
Original commit message from CVS:
* gst/auparse/gstauparse.c :
- Document all audio encoding we can encounter from Solaris 9
headers and libsndfile information.
- Increase max. rate from 48000 to 192000 (to match other elements)
- Don't try to play junk data between header and samples
Original commit message from CVS:
* gst/cdxaparse/gstcdxaparse.c :
Add mpegversion to CAPS to make it link
Rank is as GST_RANK_SECONDARY instead of NONE
