Mark Nauwelaerts [Tue, 17 Jan 2012 20:36:21 +0000 (21:36 +0100)]
adpcmdec: port to audiodecoder
Mark Nauwelaerts [Tue, 17 Jan 2012 20:36:08 +0000 (21:36 +0100)]
adpcmenc: port to audioencoder
Mark Nauwelaerts [Tue, 17 Jan 2012 17:33:09 +0000 (18:33 +0100)]
gsmdec: port to audiodecoder
Mark Nauwelaerts [Tue, 17 Jan 2012 17:32:23 +0000 (18:32 +0100)]
gsmenc: port to audioencoder
Vincent Penquerc'h [Mon, 16 Jan 2012 17:30:22 +0000 (17:30 +0000)]
debugspy: fix string leak
Vincent Penquerc'h [Mon, 16 Jan 2012 17:07:29 +0000 (17:07 +0000)]
mpegdemux: fix missing finalize chaining to parent
This code does not seem to trigger on my test streams, so it's
a bit of a blind fix, but seems safe enough to push.
Vincent Penquerc'h [Mon, 16 Jan 2012 17:02:22 +0000 (17:02 +0000)]
opencv: fix missing finalize chaining to parent class
Not compiled, I do not have the necessary dependencies.
Vincent Penquerc'h [Mon, 16 Jan 2012 17:01:18 +0000 (17:01 +0000)]
gme: fix missing dispose chaining to parent class
Not compiled, I do not have the necessary dependencies.
Vincent Penquerc'h [Mon, 16 Jan 2012 16:59:57 +0000 (16:59 +0000)]
spc: fix missing dispose chaining to parent class
Not even compiled, I do not have the dependencies needed.
Reynaldo H. Verdejo Pinochet [Fri, 13 Jan 2012 18:57:33 +0000 (15:57 -0300)]
Android, Fix wrong dinamic linking attempt
gstcodecparsers is built as an static library.
This change fixes building with the Android buildsystem.
Vincent Penquerc'h [Fri, 13 Jan 2012 13:30:57 +0000 (13:30 +0000)]
cogmse: fix finalize not calling base class (and leaking)
Vincent Penquerc'h [Fri, 13 Jan 2012 12:55:39 +0000 (12:55 +0000)]
cogmse: signal chain functions to exit on flush-start
Vincent Penquerc'h [Fri, 13 Jan 2012 12:08:28 +0000 (12:08 +0000)]
sirenenc: fix base class call mixup in finalize
Vincent Penquerc'h [Fri, 13 Jan 2012 11:22:42 +0000 (11:22 +0000)]
cogmse: fix typo using the wrong pad
Vincent Penquerc'h [Fri, 13 Jan 2012 11:22:21 +0000 (11:22 +0000)]
cogmse: fix possible buffer leak
Vincent Penquerc'h [Fri, 13 Jan 2012 10:32:59 +0000 (10:32 +0000)]
linsyssdi: fix leaks
Vincent Penquerc'h [Fri, 13 Jan 2012 10:25:22 +0000 (10:25 +0000)]
rsvgoverlay: fix adapter leak
Vincent Penquerc'h [Thu, 12 Jan 2012 17:16:32 +0000 (17:16 +0000)]
cogmse: fix pad leaks
Vincent Penquerc'h [Thu, 12 Jan 2012 17:11:14 +0000 (17:11 +0000)]
patchdetect: remove (leaking) dead code
Vincent Penquerc'h [Thu, 12 Jan 2012 16:45:39 +0000 (16:45 +0000)]
inter: remove (leaking) dead code
Vincent Penquerc'h [Thu, 12 Jan 2012 16:36:19 +0000 (16:36 +0000)]
schroenc: balance encoder and video format alloc/dealloc
They would leak if just constructing and freeing an object,
and probably crash if going from PLAYING to READY and back
again (though I haven't tested that).
Vincent Penquerc'h [Thu, 12 Jan 2012 15:58:36 +0000 (15:58 +0000)]
hlsdemux: fix queue leak
Vincent Penquerc'h [Thu, 12 Jan 2012 15:43:51 +0000 (15:43 +0000)]
dccpclientsink: fix string leak
Vincent Penquerc'h [Thu, 12 Jan 2012 15:19:01 +0000 (15:19 +0000)]
y4mdec: fix event leak
Vincent Penquerc'h [Thu, 12 Jan 2012 14:34:25 +0000 (14:34 +0000)]
mpegvideoparser: fix memory leak
Vincent Penquerc'h [Wed, 11 Jan 2012 18:24:41 +0000 (18:24 +0000)]
schrodec: fix tag leak
Vincent Penquerc'h [Wed, 11 Jan 2012 18:12:22 +0000 (18:12 +0000)]
schrodec: take out unused leaking buffer
Vincent Penquerc'h [Wed, 11 Jan 2012 16:34:35 +0000 (16:34 +0000)]
opusparse: fix caps leak
Vincent Penquerc'h [Wed, 11 Jan 2012 16:32:31 +0000 (16:32 +0000)]
opusparse: fix slist leak
Vincent Penquerc'h [Wed, 11 Jan 2012 13:32:36 +0000 (13:32 +0000)]
tests: fix buffer leaks in opus tests
Vincent Penquerc'h [Tue, 10 Jan 2012 17:27:54 +0000 (17:27 +0000)]
modplug: fix caps leak
Vincent Penquerc'h [Tue, 10 Jan 2012 13:38:50 +0000 (13:38 +0000)]
opusenc: fix slist leak
Vincent Penquerc'h [Tue, 10 Jan 2012 13:38:42 +0000 (13:38 +0000)]
opusenc: fix caps leak
Vincent Penquerc'h [Tue, 10 Jan 2012 13:25:05 +0000 (13:25 +0000)]
interlace: fix caps leaks
Vincent Penquerc'h [Tue, 10 Jan 2012 13:24:22 +0000 (13:24 +0000)]
interlace: fix ref leak
Edward Hervey [Tue, 10 Jan 2012 10:41:34 +0000 (11:41 +0100)]
tsdemux: Don't provide bogus values in non-TIME push-mode
Only take the upstream segment values in GST_FORMAT_TIME
Edward Hervey [Tue, 10 Jan 2012 10:37:13 +0000 (11:37 +0100)]
tsdemux: Handle rollover in PTS and DTS
Also hands back-rollover which can happen with PTS
Oleksij Rempel (Alexey Fisher) [Wed, 4 Jan 2012 10:05:48 +0000 (11:05 +0100)]
vp8dec: use is_alt_data option to prevent timestamp collisions
altref/invisible frames usually stored in container with same timestamp as
dependet frame. This make basevideodecoder to update timestamp for dependet
frame and couse TS colision on next frame:
^- here is altref
time : 1 2 3 4 5 6 7 8 9
webm ts : 1 3 5 5 7 9
vp8dec ts: 1 3 7 7 9
Fix bug: https://bugzilla.gnome.org/show_bug.cgi?id=655245
Signed-off-by: Oleksij Rempel (Alexey Fisher) <bug-track@fisher-privat.net>
Oleksij Rempel (Alexey Fisher) [Wed, 4 Jan 2012 10:09:23 +0000 (11:09 +0100)]
basevideodecoder: add option to hande alt frame data
altref/invisible 'frame' do not describe some frame directly, so it can't be
displayed and timestamps should not be updated.
Fix bug: https://bugzilla.gnome.org/show_bug.cgi?id=655245
Signed-off-by: Oleksij Rempel (Alexey Fisher) <bug-track@fisher-privat.net>
Mark Nauwelaerts [Mon, 9 Jan 2012 13:56:29 +0000 (14:56 +0100)]
codecparsers: mpeg4: fix parsing when recursively skipping user data
Vincent Penquerc'h [Mon, 9 Jan 2012 11:03:24 +0000 (11:03 +0000)]
h264parse: use the framerate that was copied a few lines above
Vincent Penquerc'h [Mon, 9 Jan 2012 10:58:56 +0000 (10:58 +0000)]
mpegvideoparse: add a couple consts on string literal arrays
David Biomech360 [Mon, 19 Dec 2011 16:24:31 +0000 (11:24 -0500)]
coloreffects: added support for a visual magnifier color filter to be used in gnome-video-effects
https://bugzilla.gnome.org/show_bug.cgi?id=666542
Vincent Penquerc'h [Fri, 6 Jan 2012 10:52:36 +0000 (10:52 +0000)]
h264parse: fix test and set not using the same input
This fixes H.264 output not having a set framerate, and
confusing some of the elements downstream (such as interlace).
Vincent Penquerc'h [Fri, 6 Jan 2012 10:39:07 +0000 (10:39 +0000)]
interlace: avoid dividing by zero on unkbown framerate
If the framerate is unknown, we cannot generate meaningful
buffer timestamps/durations, so set them to _NONE instead of
calculating something wrong and dividing by 0.
Oleksij Rempel (Alexey Fisher) [Mon, 2 Jan 2012 07:28:13 +0000 (08:28 +0100)]
vp8: add initial preset file
This is initial preset file, currently with only one profile
for realtime encoding.
Signed-off-by: Oleksij Rempel (Alexey Fisher) <bug-track@fisher-privat.net>
Zhao Halley [Thu, 8 Dec 2011 04:15:39 +0000 (12:15 +0800)]
codecparsers: remove h263 gob parsing, since there is no gob immediate after svh
Zhao Halley [Thu, 8 Dec 2011 04:13:38 +0000 (12:13 +0800)]
codecparsers: add missing zero_bits in h263 parsing
Zhao Halley [Thu, 8 Dec 2011 04:12:21 +0000 (12:12 +0800)]
codecparsers: includes start_marker in h263 header for parsing
Zhao Halley [Thu, 8 Dec 2011 04:07:42 +0000 (12:07 +0800)]
codecparsers: fix offset issue in h263 parse
Zhao Halley [Thu, 8 Dec 2011 03:27:56 +0000 (11:27 +0800)]
codecparsers: fix mask pattern for h263 start marker
Vincent Penquerc'h [Thu, 5 Jan 2012 11:01:01 +0000 (11:01 +0000)]
tests: test for GTK 2.24 for new symbols used in camerabin2 tests
The whole tree is likely to switch to GTK 3.0 at some point, but
we might as well continue building with slightly less new versions
if it's not hard to do.
Tim-Philipp Müller [Wed, 4 Jan 2012 18:42:57 +0000 (18:42 +0000)]
tests: remove temporary files when done in camerabin, camerabin2 and jifmux tests
Mark Nauwelaerts [Wed, 4 Jan 2012 09:56:51 +0000 (10:56 +0100)]
h264parse: fix and tweak frame timestamping
... to run with properly init'ed variables, and to only perform interpolation
in safe cases.
Mark Nauwelaerts [Tue, 3 Jan 2012 21:09:04 +0000 (22:09 +0100)]
h264parse: prefer negotiating pass-through
Mark Nauwelaerts [Tue, 3 Jan 2012 20:59:32 +0000 (21:59 +0100)]
h264parse: mark byte-stream AU frame with inserted codec NALs as keyframe
Mark Nauwelaerts [Tue, 3 Jan 2012 19:23:16 +0000 (20:23 +0100)]
h264parse: properly determine skip amount when no NAL start found yet
Mark Nauwelaerts [Tue, 3 Jan 2012 19:21:51 +0000 (20:21 +0100)]
h264parse: some more proper parse state reset
Thiago Santos [Fri, 30 Dec 2011 03:02:46 +0000 (00:02 -0300)]
basevideocodec: Reset should reset current caps
New caps will need to be negotiated when reset happens
(PAUSED to READY). Without reseting the internal
stored format, basevideoencoder/decoder wouldn't call the
configuration function when the same caps was negotiated
again as they would believe this was the same caps as before.
The issue is that _stop has been called when going to READY and
the elements would have reset their internal codec libs/state as
well. A new configuration should be done.
Thiago Santos [Fri, 30 Dec 2011 03:02:19 +0000 (00:02 -0300)]
camerabin2: clarify some message logs
Be explicit on which encodebin we're switching profiles on the
log messages
Thiago Santos [Thu, 29 Dec 2011 16:30:13 +0000 (13:30 -0300)]
examples: camerabin2: add some error handling for format changing
Check that the format could be initialized and shows a modal dialog
in case it couldn't.
Thiago Santos [Thu, 29 Dec 2011 15:47:48 +0000 (12:47 -0300)]
examples: camerabin2: add format selection for camerabin2 example
Adds a combobox for selecting the video profile for recordings.
Useful for testing multiple formats a/v sync when recording.
Thiago Santos [Wed, 28 Dec 2011 19:59:15 +0000 (16:59 -0300)]
tests: remove comment for already fixed issues on makefile
Some issues on camerabin2 and jifmux tests were fixed some
time ago, remove the comments about them
Thiago Santos [Wed, 28 Dec 2011 19:57:25 +0000 (16:57 -0300)]
tests: camerabin2: Fix message leak in tests
Thiago Santos [Tue, 27 Dec 2011 14:59:16 +0000 (11:59 -0300)]
camerabin2: fix gvalue leak
Thiago Santos [Tue, 27 Dec 2011 13:23:18 +0000 (10:23 -0300)]
tests: camerabin2: fix mainloop leaks
Thiago Santos [Tue, 27 Dec 2011 13:20:48 +0000 (10:20 -0300)]
camerabin2: secure event ref when sending event to parent class
camerabin2 needs to hold an extra ref on send_event before sending
it to the parent class as it will need to use it after the parent
handles it.
Mark Nauwelaerts [Fri, 23 Dec 2011 08:30:31 +0000 (09:30 +0100)]
faac: standardize some field data types
Fixes #666750.
Thiago Santos [Thu, 22 Dec 2011 18:19:05 +0000 (15:19 -0300)]
basecamerabinsrc: remove redundant functions
gst_bin_get_by_interface should be enough for applications.
Haven't seen anyone using those other than wrappercamerabinsrc.
Thiago Santos [Thu, 22 Dec 2011 18:01:47 +0000 (15:01 -0300)]
camerabin2: remove unused helper function
This drop_eos_probe function is not used anymore, remove it
Mark Nauwelaerts [Thu, 22 Dec 2011 16:15:28 +0000 (17:15 +0100)]
faac: recreate encoder after flushing
... since it appears the existing one can't be bothered anymore to produce
some output.
Mark Nauwelaerts [Thu, 22 Dec 2011 16:14:09 +0000 (17:14 +0100)]
faac: refactor encoder library setup
Mark Nauwelaerts [Thu, 22 Dec 2011 16:10:18 +0000 (17:10 +0100)]
voaacenc: no need to duplicate baseclass sanity checks
Mark Nauwelaerts [Thu, 22 Dec 2011 16:09:30 +0000 (17:09 +0100)]
voamrwbenc: no need to duplicate baseclass sanity checks
Thiago Santos [Mon, 19 Dec 2011 19:49:08 +0000 (16:49 -0300)]
camerabin2: Fix A/V sync for captures
Audio src was pushing newsegments starting from 0, even when its buffers
were not starting with 0 as their timestamps. Add a buffer probe that
checks the ts of the first buffer and uses that to push a newsegment,
just like we do for video on wrappercamerabinsrc.
Thiago Santos [Mon, 19 Dec 2011 18:17:22 +0000 (15:17 -0300)]
tests: camerabin2: enable tests
Tests are passing now, enable them
Thiago Santos [Mon, 19 Dec 2011 18:16:42 +0000 (15:16 -0300)]
wrappercamerabinsrc: Push newsegments on video capture start
Pushing newsegments to inform muxers about the start time of the
video buffer timestamps.
Mark Nauwelaerts [Tue, 20 Dec 2011 13:04:52 +0000 (14:04 +0100)]
codecparsers: h264: clear sei message prior to parsing
Fixes #666584.
Vincent Penquerc'h [Mon, 19 Dec 2011 13:18:51 +0000 (13:18 +0000)]
mpegtsdemux: only warn when we see a PID with unknown type
This is quite frequent, and we will just ignore that stream,
so an error is not warranted.
Alessandro Decina [Mon, 19 Dec 2011 11:32:26 +0000 (12:32 +0100)]
h264parse: avoid pushing SPS/PPS at start only if both are found in stream
Avoid injecting SPS/PPS nals only if both are already present in the stream.
Fixes some AVC clips that happen to have only PPS nals in stream.
Alessandro Decina [Sun, 18 Dec 2011 17:37:08 +0000 (18:37 +0100)]
mpeg4videoparse: handle force key unit events
Alessandro Decina [Sun, 18 Dec 2011 16:40:57 +0000 (17:40 +0100)]
h264parse: do wait for a keyframe when handling force key unit events
Although some decoders can start decoding with just PPS and SPS and waiting for
a keyframe is bad when the keyframe interval is large, gst-ffmpeg does need a
keyframe to start decoding.
Alessandro Decina [Sun, 18 Dec 2011 16:32:11 +0000 (17:32 +0100)]
mpegtsmux: unset the current PCR stream when starting a new unit
Unset the current PCR stream when starting a new unit so that afterwards a new
stream is selected as PCR and PAT and PMT tables are regenerated.
Alessandro Decina [Sun, 18 Dec 2011 16:31:18 +0000 (17:31 +0100)]
mpegtsmux: fix key unit event handling when muxing both audio and video
Take only video buffers into account when deciding when to cut so that new units
start with a video keyframe.
Tim-Philipp Müller [Thu, 15 Dec 2011 18:14:24 +0000 (18:14 +0000)]
katetiger: don't unref NULL buffer
Thiago Santos [Thu, 15 Dec 2011 14:11:41 +0000 (11:11 -0300)]
tests: jifmux: re-enable jifmux's tests
Fix for its test is now available on -base module.
Enabling again.
Vincent Penquerc'h [Thu, 15 Dec 2011 13:47:55 +0000 (13:47 +0000)]
videomeasure: port to GstCollectPads2
Vincent Penquerc'h [Thu, 15 Dec 2011 13:39:41 +0000 (13:39 +0000)]
mxfmux: port to GstCollectPads2
Only tested with audio, I can't get it to accept video (before
the port).
Vincent Penquerc'h [Thu, 15 Dec 2011 13:22:24 +0000 (13:22 +0000)]
mpegtsmux: port to GstCollectPads2
I *think* the stream lock use is correct, not 100% sure.
Vincent Penquerc'h [Thu, 15 Dec 2011 13:08:42 +0000 (13:08 +0000)]
mpegpsmux: port to GstCollectPads2
Vincent Penquerc'h [Thu, 15 Dec 2011 13:04:51 +0000 (13:04 +0000)]
frei0r: port to GstCollectPads2
Untested, no sample pipeline mentioned, and I've no idea what
this plugin does, but since all the previous ports work out of
the box, this is deemed to work until evidence it doesn't.
Vincent Penquerc'h [Thu, 15 Dec 2011 13:02:16 +0000 (13:02 +0000)]
compare: port to GstCollectPads2
Vincent Penquerc'h [Thu, 15 Dec 2011 12:59:57 +0000 (12:59 +0000)]
asfmux: port to GstCollectPads2
Mark Nauwelaerts [Wed, 14 Dec 2011 16:36:55 +0000 (17:36 +0100)]
dtsdec: avoid crashing upon draining
... and clean up some variable init and debug statement.
Thiago Santos [Wed, 14 Dec 2011 11:54:32 +0000 (08:54 -0300)]
tests: mpegtsmux: remove unused variable
Tommi Myöhänen [Wed, 14 Dec 2011 11:32:50 +0000 (08:32 -0300)]
camerabin2: fix leak on error handling
Camerabin2 parses warning messages with gst_message_parse_warning(message,
&err, &debug) but doesn't free given GError and debug strings.
Documentation shows that the ownership of those fields is transferred
to caller (they are marked "[transfer full]" in the API docs).
Julien Isorce [Mon, 12 Dec 2011 14:52:20 +0000 (15:52 +0100)]
mpegtsparse: check offset when retrieving table_id on malformed packets
Fix bug #665988
Sebastian Dröge [Tue, 13 Dec 2011 09:44:42 +0000 (10:44 +0100)]
Revert "mpegtsparse: check offset when retrieving table_id on malformed packets"
This reverts commit
e62978d045d57ee9fed7598c02b9017eb974ea36.
Sebastian Dröge [Tue, 13 Dec 2011 09:17:41 +0000 (10:17 +0100)]
basevideocodec: Don't use G_TYPE_REGISTER_BOXED() yet
This would require GLib 2.26.
Vincent Penquerc'h [Mon, 12 Dec 2011 20:01:17 +0000 (20:01 +0000)]
tiger: advertise all the new YUV formats we can now use