| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Original commit message from CVS:
* ext/mpeg2dec/gstmpeg2dec.c: (gst_mpeg2dec_alloc_buffer):
Fix overflows.
* gst/avi/gstavidemux.c: (gst_avi_demux_process_next_entry):
More debug.
|
|
silence if we receive a FILLER event befo...
Original commit message from CVS:
* ext/alsa/gstalsa.c: (gst_alsa_set_default_format):
* ext/alsa/gstalsa.h:
* ext/alsa/gstalsasink.c: (gst_alsa_sink_check_event):
Simplistically attempt to pick a default format for filling
with silence if we receive a FILLER event before the caps are
negotiated.
|
|
Original commit message from CVS:
* gst/playback/gstplaybasebin.c (gst_play_base_bin_class_init)
(gst_play_base_bin_finalize):
Fix two memory leaks (a gmutex and a gcond) and
implement finalize instead of dispose.
fixes #307606
|
|
Original commit message from CVS:
* ext/cdio/cddasrc.c: (cddasrc_open), (cddasrc_change_state):
Output more debug info.
|
|
column (bug #300471)
Original commit message from CVS:
* gst/videotestsrc/videotestsrc.c: (paint_hline_I420): Fix I420
painting that ends in an odd column (bug #300471)
|
|
like when called from gst_element_make_f...
Original commit message from CVS:
* ext/cdio/cddasrc.c: (cddasrc_uri_set_uri):
Default to cdda://1 when called with empty URI (cdda://)
like when called from gst_element_make_from_uri(). We need
to return TRUE in the set_uri() function if we can handle
the protocol but don't know whether the track number
is valid yet. (fixes #307804)
|
|
Original commit message from CVS:
* gst/videoscale/gstvideoscale.c: Add 24-bit and 15-bit RGB
* gst/videoscale/vs_image.c:
* gst/videoscale/vs_image.h:
* gst/videoscale/vs_scanline.c:
* gst/videoscale/vs_scanline.h:
|
|
events with the right MIME type.
Original commit message from CVS:
2005-06-15 Martin Soto <martinsoto@users.sourceforge.net>
* gst/mpegstream/gstdvddemux.c (gst_dvd_demux_handle_dvd_event):
Only process events with the right MIME type.
(gst_dvd_demux_get_audio_stream): Stream objects were being
reallocated, but the (potentially new) pointer wasn't being stored
in the array again. Make sure to always store the pointer after
reallocation. Fixes a segmentation fault when playing DVDs with
LPCM sound.
(gst_dvd_demux_plugin_init): Rename element for local use.
|
|
event on each VTS change.
Original commit message from CVS:
2005-06-15 Jan Schmidt <thaytan@mad.scientist.com>
* ext/dvdnav/Makefile.am:
* ext/dvdnav/dvdnavsrc.c: (dvdnavsrc_init), (dvdnavsrc_finalize),
(read_vts_info), (dvdnav_build_titlelang_event), (dvdnavsrc_loop),
(dvdnavsrc_change_state):
Read VTS info using libdvdread, and use to send the stream
info event on each VTS change.
Don't gst_element_wait, rely on queues for that.
|
|
Original commit message from CVS:
* gst/videorate/gstvideorate.c (gst_videorate_link):
Fix memory leak
|
|
the lock when we emit a signal.
Original commit message from CVS:
* gst/tcp/gstmultifdsink.c: (gst_multifdsink_remove_client_link):
Avoid potential deadlock because we do not release the lock
when we emit a signal.
|
|
Original commit message from CVS:
Add 'new plugin: cddasrc' line to 0.8.9 notes.
|
|
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* ext/cdparanoia/gstcdparanoia.c: (cdparanoia_event):
Off-by-one in illogical counting (part of our ABI...), #307459.
|
|
compile with gcc4.
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* ext/cairo/gsttextoverlay.c: (gst_textoverlay_render_text):
Remove guchar * cast. This makes gst-plugins compile with gcc4.
|
|
the previous version, even though we forgot...
Original commit message from CVS:
* ext/amrnb/amrnbparse.c: (gst_amrnbparse_loop):
Fix header (which apparently accidently worked in the previous
version, even though we forgot to skip one byte).
|
|
Original commit message from CVS:
* configure.ac:
Require cairo 0.5.x
* ext/cairo/gstcairo.c:
Change plugin name to 'cairo' - 'timeoverlay' doesn't make
much sense and conflicts with the pango timeoverlay plugin.
* ext/cairo/gsttextoverlay.c: (gst_cairotextoverlay_get_type),
(gst_textoverlay_render_text), (gst_textoverlay_loop),
(gst_textoverlay_font_init), (gst_textoverlay_finalize),
(gst_textoverlay_init), (gst_textoverlay_set_property):
* ext/cairo/gsttextoverlay.h:
* ext/cairo/gsttimeoverlay.c: (gst_cairotimeoverlay_get_type),
(gst_timeoverlay_class_init), (gst_timeoverlay_setup),
(gst_timeoverlay_planar411):
* ext/cairo/gsttimeoverlay.h:
Make compile and somewhat work with cairo 0.5. The textoverlay
plugin is bitrotten and/or unfinished though and needs more
work (text positioning, invalid memory access, parse font-desc
property properly).
|
|
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (gst_play_base_bin_change_state):
Crap, double free.
|
|
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (setup_source), (prepare_output),
(gst_play_base_bin_change_state):
Mark some more for i18n.
|
|
Original commit message from CVS:
* ext/alsa/gstalsa.c: (gst_alsa_open_audio):
* gst/playback/gstplaybasebin.c: (prepare_output),
(gst_play_base_bin_change_state):
* sys/xvimage/xvimagesink.c: (gst_xvimagesink_get_xv_support):
Fix some error domains; don't print the URI in playbin errors.
|
|
Original commit message from CVS:
* gconf/gstreamer.schemas.in:
* gst-libs/gst/gconf/gconf.c: (gst_gconf_get_default_video_sink):
Default to autovideosink instead of @DEFAULT_VIDEOSINK@.
Autovideosink dynamically selects between ximagesink/xvimagesink,
and is thus more userfriendly
|
|
affect our timestamps. Fixes #140237.
Original commit message from CVS:
* ext/mad/gstmad.c: (gst_mad_check_caps_reset), (gst_mad_chain):
Update total_samples after new header, since it will affect our
timestamps. Fixes #140237.
|
|
Original commit message from CVS:
Fixed regeneration of wavpack header.
|
|
Original commit message from CVS:
Fixed time query and added position query.
|
|
wavpack code. Supports more formats than ...
Original commit message from CVS:
New wavpack plugin using the wavpack library instead of importing the whole
wavpack code. Supports more formats than the old plugin.
|
|
warning (#304754).
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* ext/cdparanoia/gstcdparanoia.c: (cdparanoia_open),
(cdparanoia_change_state):
Report device when open failed. Remove warning (#304754).
|
|
returning.
Original commit message from CVS:
* gst/matroska/matroska-mux.c: (gst_matroska_mux_loop):
Some better error reporting instead of just returning.
|
|
GST_BUFFER_SIZE(), since length is the allocated le...
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_get):
Don't use bytesused/length of a buffer for GST_BUFFER_SIZE(),
since length is the allocated length in the kernel, which may
be more, and bytesused is not set by all drivers and may be
less if part of the frame was lost. Instead, use the calculated
piximage size. Fixes #301718.
|
|
(#305658).
Original commit message from CVS:
* gst/adder/gstadder.c: (gst_adder_link):
Workaround for multi-mp3 case. Seems to work fine (#305658).
Patch based on work from Akos Maroy <darkeye@tyrell.hu>.
|
|
Original commit message from CVS:
* gst/adder/gstadder.c: (gst_adder_link):
Remove double loop (probably copy/paste error).
|
|
Original commit message from CVS:
* gst/qtdemux/qtdemux.c: (gst_qtdemux_loop_header):
Fix EOS handling (#305150).
|
|
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* gst/adder/gstadder.c: (gst_adder_init):
Use getcaps function for source pad, too (#301292).
|
|
function because it has more than one incarnati...
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (au_type_find),
(plugin_init):
Make audio/x-au its own typefind function because it has more than
one incarnation. Should fix memleak in #306938.
|
|
(#306641).
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* ext/xvid/gstxvid.c: (gst_xvid_init):
Make struct static, xvid crashes internally otherwise (#306641).
|
|
some ASF files.
Original commit message from CVS:
* gst/asfdemux/gstasfdemux.c: (gst_asf_demux_process_chunk):
Cap framerate; should fix negotiation issues for some ASF files.
|
|
Original commit message from CVS:
* gst-libs/gst/riff/riff-media.c:
(gst_riff_create_video_caps_with_data),
(gst_riff_create_video_template_caps):
Add Microsoft ISO MPEG-4 1.1 (M4S2; #306659).
|
|
Original commit message from CVS:
* gst/qtdemux/qtdemux.c: (qtdemux_parse_trak),
(qtdemux_video_caps):
Fix up SVQ3 caps (to fix frame order problems; #306530).
|
|
did the faad API change or something ? al...
Original commit message from CVS:
* ext/faad/gstfaad.c: (gst_faad_sinkconnect), (gst_faad_chain):
I don't understand how this got unnoticed for so long - did the
faad API change or something ?
also, we're getting samplerate and channels in the first call,
but not actually *doing* anything with that. Sigh.
* ext/libmng/gstmngdec.c:
switch from 0.0 to 1.0 minimum framerate, even though that's
*stupid* because we want to use mngdec for things that happen
less than once a second, but it's the only way to have
somewhat acceptable negotiation.
* ext/theora/theoraenc.c: (theora_enc_chain):
add a debug message.
|
|
(larger for video than for others).
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (gen_preroll_element):
Set max. buffer size depending on media type (larger for video
than for others).
|
|
Original commit message from CVS:
support h264 video in matroska
|
|
Original commit message from CVS:
* gst/qtdemux/qtdemux.c: (qtdemux_audio_caps):
Fix distortion in 8-bit audio tracks (#306680).
|
|
(fixes #306660)
Original commit message from CVS:
* sys/glsink/glimagesink.c: (gst_glimagesink_getcaps),
(gst_glimagesink_init_display), (gst_glimagesink_chain):
Make MESA extensions dependent on having MESA. (fixes #306660)
|
|
confused arrays and lists.
Original commit message from CVS:
* gst-libs/gst/audio/multichannel.c:
(gst_audio_set_structure_channel_positions_list):
revert my patch from 2005-06-01. I confused arrays and lists.
|
|
sound-juicer and totem, all work fine after the...
Original commit message from CVS:
* ext/cdparanoia/gstcdparanoia.c: (set_track), (cdparanoia_event):
Unbreak soundjuicer (tested gnome-cd, sound-juicer and totem,
all work fine after these changes).
|
|
Original commit message from CVS:
* gst/qtdemux/qtdemux.c: (qtdemux_parse_trak):
Add codecdata on avc1 (h264).
|
|
only the Y plane.
Original commit message from CVS:
* gst/videoscale/gstvideoscale.c: (gst_videoscale_chain):
Let's not break planar YUV by allocating only the Y plane.
|
|
state change function, proper error reporti...
Original commit message from CVS:
* sys/ximage/ximagesrc.c: (gst_ximagesrc_getcaps),
(gst_ximagesrc_change_state), (gst_ximagesrc_init),
(gst_ximagesrc_get_property), (gst_ximagesrc_base_init),
(gst_ximagesrc_class_init):
Some extra checks, move display initialization to state change
function, proper error reporting if no $DISPLAY is available.
|
|
Original commit message from CVS:
* ext/audiofile/gstafparse.c: (gst_afparse_class_init),
(gst_afparse_init), (gst_afparse_finalize):
Portability fix for non-exported function (#305435).
|
|
actually set in_duration where we want to ca...
Original commit message from CVS:
* gst/audiorate/gstaudiorate.c: (gst_audiorate_chain):
Fix copy'n'past-o from previous commit and actually
set in_duration where we want to calculate in_duration.
Make "in" and "out" properties return the number of
samples (as advertised) instead of the number of
buffers.
|
|
buffer with a greater-than-zero offset can...
Original commit message from CVS:
* gst/audiorate/gstaudiorate.c: (gst_audiorate_link),
(gst_audiorate_init), (gst_audiorate_chain):
Make case of overlapping buffers work (a sub buffer
with a greater-than-zero offset cannot have the same
length as the input buffer); also, calculate offset
fields from timestamp/duration and vice versa when
those fields are not set (we later assume they are all
set and valid, so we better set them to something
sane instead of doing calculations with random input
values). (also fixes #305428)
|
|
Original commit message from CVS:
if faad wants params in exact bit-width, then deliver them as such
|
