platform/upstream/gstreamer.git
11 years agoavi: fix debug
Wim Taymans [Mon, 22 Oct 2012 09:55:22 +0000 (11:55 +0200)]
avi: fix debug

11 years agoqtdemux: add support for 'generic' samples
Wim Taymans [Mon, 22 Oct 2012 09:39:37 +0000 (11:39 +0200)]
qtdemux: add support for 'generic' samples

Add support for stuffing a complete stream into 1 sample.

See https://bugzilla.gnome.org/show_bug.cgi?id=686550

11 years agotests: remove superfluous g_type_init() call
Tim-Philipp Müller [Sat, 20 Oct 2012 12:01:41 +0000 (13:01 +0100)]
tests: remove superfluous g_type_init() call

It's deprecated in newer GLib and not needed here.

https://bugzilla.gnome.org/show_bug.cgi?id=686456

11 years agopulsesink: fix caps leak in acceptcaps function
Tim-Philipp Müller [Sat, 20 Oct 2012 10:32:27 +0000 (11:32 +0100)]
pulsesink: fix caps leak in acceptcaps function

11 years agoqtdemux: don't leak gst_riff_strf_auds in case of MS/RIFF audio
Tim-Philipp Müller [Fri, 19 Oct 2012 18:24:23 +0000 (19:24 +0100)]
qtdemux: don't leak gst_riff_strf_auds in case of MS/RIFF audio

https://bugzilla.gnome.org/show_bug.cgi?id=681192

11 years agomatroskamux: unsigned subtitle template
Mark Nauwelaerts [Thu, 18 Oct 2012 20:20:39 +0000 (22:20 +0200)]
matroskamux: unsigned subtitle template

11 years agopulsesink: in accept_caps() check if ring buffer is NULL before de-referencing
Tim-Philipp Müller [Thu, 18 Oct 2012 10:32:10 +0000 (11:32 +0100)]
pulsesink: in accept_caps() check if ring buffer is NULL before de-referencing

And sprinkle some thread-safety (take object lock for
accessing ring buffer, and pa main loop lock for the
context).

https://bugzilla.gnome.org/show_bug.cgi?id=683782

11 years agovideomixer2: Fix race condition where a src setcaps is ignored
Youness Alaoui [Thu, 13 Sep 2012 00:10:00 +0000 (00:10 +0000)]
videomixer2: Fix race condition where a src setcaps is ignored

If both pads receive data at the same time, they will both get their
sink_setcaps called which will call the src_setcaps, but there is
a race condition where the second one might not be called.
Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=683842

11 years agomatroskamux: do not use unoffical V_MJPEG codec id
Mark Nauwelaerts [Mon, 31 Oct 2011 14:43:25 +0000 (15:43 +0100)]
matroskamux: do not use unoffical V_MJPEG codec id

Since it's not spec'ed, consider it a VfW compatibility
case. Many applications (e.g. avidemux) don't understand
the unofficial V_MJPEG id.

Fixes #659837.

Conflicts:
gst/matroska/matroska-mux.c

11 years agojpeg, png, vpx: use gst_element_class_set_static_metadata()
Tim-Philipp Müller [Wed, 17 Oct 2012 16:03:39 +0000 (17:03 +0100)]
jpeg, png, vpx: use gst_element_class_set_static_metadata()

Avoids some string copies.

11 years agojpegdepay: store quant tables in zigzag order
Wim Taymans [Wed, 17 Oct 2012 12:23:01 +0000 (14:23 +0200)]
jpegdepay: store quant tables in zigzag order

11 years agortsession: fix compiler warning
Wim Taymans [Wed, 17 Oct 2012 11:55:45 +0000 (13:55 +0200)]
rtsession: fix compiler warning

11 years agortpbin: clarify the ntp-sync option
Wim Taymans [Wed, 17 Oct 2012 11:35:07 +0000 (13:35 +0200)]
rtpbin: clarify the ntp-sync option

11 years agortpsession: update caps in the source
Wim Taymans [Wed, 17 Oct 2012 11:15:48 +0000 (13:15 +0200)]
rtpsession: update caps in the source

Inform the source when caps changed. This was removed in the port to 1.0
leaving the source unaware of the clock-rate and unable to interpollate
rtp timestamps for SR packets.

11 years agortpbin: set PTS and DTS in jitterbufffer
Wim Taymans [Wed, 17 Oct 2012 10:46:32 +0000 (12:46 +0200)]
rtpbin: set PTS and DTS in jitterbufffer

11 years agortpbin: disable check for ntp-sync
Wim Taymans [Wed, 17 Oct 2012 10:24:22 +0000 (12:24 +0200)]
rtpbin: disable check for ntp-sync

Disable the check for the ntp-sync method. It is expected that
a rather larger offset needs to be applied with this method.

11 years agortpbin: use running-time for NTP time
Wim Taymans [Wed, 17 Oct 2012 10:17:32 +0000 (12:17 +0200)]
rtpbin: use running-time for NTP time

When use-pipeline-clock is set, use the running-time of the
pipeline to calculate the NTP timestamps. This method would previously
only work when the base-time is set to 0 but with this change it can
also work with different offsets and we can also implement pause/resume
of the sender and receiver now.

11 years agovideocrop: port to videofilter
Wim Taymans [Wed, 17 Oct 2012 08:20:12 +0000 (10:20 +0200)]
videocrop: port to videofilter

11 years agovideobox: use out_info for out properties
Wim Taymans [Wed, 17 Oct 2012 07:36:50 +0000 (09:36 +0200)]
videobox: use out_info for out properties

11 years agomedian: small cleanups
Wim Taymans [Tue, 16 Oct 2012 12:40:19 +0000 (14:40 +0200)]
median: small cleanups

11 years agomedian: remove now that it is in videofilter
Wim Taymans [Tue, 16 Oct 2012 11:56:19 +0000 (13:56 +0200)]
median: remove now that it is in videofilter

11 years agoconfigure: remove median from build
Wim Taymans [Tue, 16 Oct 2012 11:49:11 +0000 (13:49 +0200)]
configure: remove median from build

11 years agovideomedian: copy media to videomedian
Wim Taymans [Tue, 16 Oct 2012 11:47:24 +0000 (13:47 +0200)]
videomedian: copy media to videomedian

Copy the median video filter to videofilters and rename to
videomedian.

11 years agomedia: port to 1.0
Wim Taymans [Tue, 16 Oct 2012 11:12:21 +0000 (13:12 +0200)]
media: port to 1.0

11 years agoavidemux: append palette data to paletted 8-bit RGB frames
Tim-Philipp Müller [Tue, 16 Oct 2012 00:02:11 +0000 (01:02 +0100)]
avidemux: append palette data to paletted 8-bit RGB frames

Fixes playback of 8-bit indexed RGB videos, with fixes in -base.

https://bugzilla.gnome.org/show_bug.cgi?id=686046

11 years agovp8enc: And this time fix the default target-bitrate value for real
Sebastian Dröge [Mon, 15 Oct 2012 13:36:04 +0000 (15:36 +0200)]
vp8enc: And this time fix the default target-bitrate value for real

11 years agovp8enc: Fix default target-bitrate value
Sebastian Dröge [Mon, 15 Oct 2012 13:30:33 +0000 (15:30 +0200)]
vp8enc: Fix default target-bitrate value

11 years agoqtdemux: don't assert if upstream size is not available when guessing bitrates
Tim-Philipp Müller [Fri, 12 Oct 2012 23:03:29 +0000 (00:03 +0100)]
qtdemux: don't assert if upstream size is not available when guessing bitrates

Fixes abort in push mode where the source is not seekable and the
size of the file is not available, as with

  cat foo.mp4 | gst-launch-1.0 playbin uri=fd://0

Less noticable with releases, since we disable all
g_assert() there.

https://bugzilla.gnome.org/show_bug.cgi?id=686008

11 years agoqtdemux: allow more streams. Bump this constant to 32, which should be
Michael Smith [Fri, 12 Oct 2012 21:38:33 +0000 (14:38 -0700)]
qtdemux: allow more streams. Bump this constant to 32, which should be
enough for real-world files.

11 years agoqtdemux: support more different fourcc values for other ProRes variants.
Michael Smith [Fri, 12 Oct 2012 21:35:24 +0000 (14:35 -0700)]
qtdemux: support more different fourcc values for other ProRes variants.

11 years agoexamples: update some element names for 1.0 in RTP examples
Tim-Philipp Müller [Thu, 11 Oct 2012 21:36:21 +0000 (22:36 +0100)]
examples: update some element names for 1.0 in RTP examples

gstrtpbin -> rtpbin
ffdec_*   -> avdec_*
ffenc_*   -> avenc_*

11 years agortspsrc: remove unused include
Wim Taymans [Wed, 10 Oct 2012 10:05:34 +0000 (12:05 +0200)]
rtspsrc: remove unused include

11 years agomultiudpsink: add multicast-iface property
Rasmus Rohde [Wed, 10 Oct 2012 08:55:28 +0000 (10:55 +0200)]
multiudpsink: add multicast-iface property

udpsrc already has support for setting the multicast interface, which
is useful for multi-homed machines. This patch adds the same code to
the multiudpsink.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=685864

11 years agomultiudpsink: don't error on send errors but only warn
Wim Taymans [Wed, 10 Oct 2012 09:32:17 +0000 (11:32 +0200)]
multiudpsink: don't error on send errors but only warn

Don't error on send errors but simply post a warning, it's possible
that the next packet will be fine.

11 years agomultiudpsink: add force-ipv4 option
Rasmus Rohde [Wed, 10 Oct 2012 08:28:24 +0000 (10:28 +0200)]
multiudpsink: add force-ipv4 option

Add an option to the multiudpsink that makes it possible to force
the use of an IPv4 socket.

This can e.g. be used to handle the issue described in
https://bugzilla.gnome.org/show_bug.cgi?id=682481

11 years agomultiudpsink: remove unused field
Wim Taymans [Wed, 10 Oct 2012 08:18:52 +0000 (10:18 +0200)]
multiudpsink: remove unused field

11 years agoudpsrc: use negotiated allocator or pool
Wim Taymans [Wed, 10 Oct 2012 08:10:26 +0000 (10:10 +0200)]
udpsrc: use negotiated allocator or pool

Use the base class to allocate a buffer for us because it knows how
to use the negotiated allocator or bufferpool.

11 years agomultiudpsink: post error when something goes wrong
Wim Taymans [Wed, 10 Oct 2012 08:09:37 +0000 (10:09 +0200)]
multiudpsink: post error when something goes wrong

11 years agospectrum: elements post element messages
Wim Taymans [Wed, 10 Oct 2012 08:09:10 +0000 (10:09 +0200)]
spectrum: elements post element messages

11 years agoBack to development (bug fixing)
Tim-Philipp Müller [Sun, 7 Oct 2012 15:56:38 +0000 (16:56 +0100)]
Back to development (bug fixing)

11 years agoRelease 1.0.1
Tim-Philipp Müller [Sun, 7 Oct 2012 14:31:12 +0000 (15:31 +0100)]
Release 1.0.1

11 years agoAutomatic update of common submodule
Tim-Philipp Müller [Sat, 6 Oct 2012 13:57:10 +0000 (14:57 +0100)]
Automatic update of common submodule

From 6c0b52c to 6bb6951

11 years agodeinterleave: output channels should be marked as MONO, not FRONT_LEFT, if
Michael Smith [Fri, 5 Oct 2012 22:12:27 +0000 (15:12 -0700)]
deinterleave: output channels should be marked as MONO, not FRONT_LEFT, if
we're not preserving input channel positions.

11 years agointerleave: use gst_audio_channel_positions_to_mask instead of a local copy
Michael Smith [Thu, 4 Oct 2012 22:13:20 +0000 (15:13 -0700)]
interleave: use gst_audio_channel_positions_to_mask instead of a local copy
of half of it. Handles some values more correctly.

11 years agogstrtpdepay: don't leak input buffer
Rasmus Rohde [Thu, 4 Oct 2012 18:32:45 +0000 (20:32 +0200)]
gstrtpdepay: don't leak input buffer

The rtp buffer is never unmapped in the normal code exit path
of gst_rtp_gst_depay_process(..) resulting in a memory leak.

https://bugzilla.gnome.org/show_bug.cgi?id=685512

11 years agovideobalance: Add support for NV12 and NV21
Sebastian Dröge [Thu, 4 Oct 2012 16:37:18 +0000 (18:37 +0200)]
videobalance: Add support for NV12 and NV21

11 years agortph264pay: do not push unmapped data
Patricia Muscalu [Mon, 1 Oct 2012 13:11:05 +0000 (15:11 +0200)]
rtph264pay: do not push unmapped data

Also do not use a GstBuffer after it has been pushed into the adapter.

https://bugzilla.gnome.org/show_bug.cgi?id=685213

11 years agometa info: threadsafe registration using g_once
Michael Smith [Wed, 3 Oct 2012 17:51:45 +0000 (10:51 -0700)]
meta info: threadsafe registration using g_once

11 years agoavidemux: push mode; handle some initial junk before hdrl list
Mark Nauwelaerts [Mon, 1 Oct 2012 13:44:01 +0000 (15:44 +0200)]
avidemux: push mode; handle some initial junk before hdrl list

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=685059

11 years agotests: port gdkpixbufsink test
Tim-Philipp Müller [Mon, 1 Oct 2012 13:03:19 +0000 (14:03 +0100)]
tests: port gdkpixbufsink test

11 years agoPurge references to liboil
Tim-Philipp Müller [Sat, 29 Sep 2012 10:59:31 +0000 (11:59 +0100)]
Purge references to liboil

https://bugzilla.gnome.org/show_bug.cgi?id=673285

11 years agoavidemux: recognize all xsub frames as keyframes
Mark Nauwelaerts [Fri, 28 Sep 2012 14:51:01 +0000 (16:51 +0200)]
avidemux: recognize all xsub frames as keyframes

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=684977

11 years agoavidemux: push mode: find the correct chunk for segment following seek
Mark Nauwelaerts [Fri, 28 Sep 2012 14:50:25 +0000 (16:50 +0200)]
avidemux: push mode: find the correct chunk for segment following seek

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=684977

11 years agoqtdemux: fix parsing in push mode when moov atom is at the end
Arnaud Vrac [Thu, 27 Sep 2012 21:17:49 +0000 (22:17 +0100)]
qtdemux: fix parsing in push mode when moov atom is at the end

When playing an mp4 file with the MOOV atom at the end of the file, playback
fails with the error message "no 'moov' atom within the first 10 MB". This is
due to a mistake in the upstream_size typing, making the seek to the end of
file never happening.

https://bugzilla.gnome.org/show_bug.cgi?id=684972

11 years agogamma: remove duplicate entries at format at caps
Andre Moreira Magalhaes (andrunko) [Thu, 27 Sep 2012 18:50:49 +0000 (15:50 -0300)]
gamma: remove duplicate entries at format at caps

Avoids extra caps/structures processing

11 years agortpvrawdepay: negotiate pool with srcpad caps
Wim Taymans [Thu, 27 Sep 2012 12:13:42 +0000 (14:13 +0200)]
rtpvrawdepay: negotiate pool with srcpad caps

11 years agodvdemux: The convert and duration queries are not supposed to change the format
Sebastian Dröge [Thu, 27 Sep 2012 09:02:51 +0000 (11:02 +0200)]
dvdemux: The convert and duration queries are not supposed to change the format

11 years agovideomixer: clear video frame more correctly
Tim-Philipp Müller [Wed, 26 Sep 2012 08:28:59 +0000 (09:28 +0100)]
videomixer: clear video frame more correctly

Make sure not to touch memory that doesn't belong to
our frame, we might be one part of a side-by-side 3D
frame, or in a picture-in-picture scenario.

11 years agoflvdemux: minor clean-up
Tim-Philipp Müller [Tue, 25 Sep 2012 23:44:59 +0000 (00:44 +0100)]
flvdemux: minor clean-up

Use GstByteWriter, because we can, and g_value_take_boxed.

11 years agoflvdemux: fix speex audio decoding by creating fake stream header
Dmitriy Samonenko [Mon, 10 Sep 2012 06:27:28 +0000 (10:27 +0400)]
flvdemux: fix speex audio decoding by creating fake stream header

https://bugzilla.gnome.org/show_bug.cgi?id=683622

11 years agovideomixer: fix warnings when using transparent background
Tim-Philipp Müller [Tue, 25 Sep 2012 20:21:15 +0000 (21:21 +0100)]
videomixer: fix warnings when using transparent background

gst_video_frame_map() increases the refcount, which makes
the buffer not writable any more technically, so calling
gst_buffer_memset() on it will cause nasty warnings.

Unit test disabled because it very rarely (for me)
fails, possibly negotiation-related.

https://bugzilla.gnome.org/show_bug.cgi?id=684398

11 years agodeinterlace: Add some useful debug logging
Robert Swain [Tue, 25 Sep 2012 08:43:28 +0000 (10:43 +0200)]
deinterlace: Add some useful debug logging

11 years agodeinterlace: Fix telecine
Robert Swain [Tue, 25 Sep 2012 08:41:44 +0000 (10:41 +0200)]
deinterlace: Fix telecine

This only affects behaviour in telecine cases with pattern locking
enabled. The default case should be untouched.

This works with the output from fieldanalysis at least, but the field
order looks swapped for telecine mixed buffers with the
David_slides_Schleef clip.

11 years agovp8enc: Disable GLIB deprecation warnings
Edward Hervey [Tue, 25 Sep 2012 12:43:15 +0000 (14:43 +0200)]
vp8enc: Disable GLIB deprecation warnings

GValueArray has been deprecated since 2.32 ... but there's no usable
replacement for it.

See https://bugzilla.gnome.org/show_bug.cgi?id=667228

11 years agovideomixer: Fix leak
Edward Hervey [Tue, 25 Sep 2012 12:18:35 +0000 (14:18 +0200)]
videomixer: Fix leak

11 years agoBack to development (bug fixing)
Tim-Philipp Müller [Mon, 24 Sep 2012 15:46:18 +0000 (16:46 +0100)]
Back to development (bug fixing)

11 years agoRelease 1.0.0
Tim-Philipp Müller [Mon, 24 Sep 2012 13:06:42 +0000 (14:06 +0100)]
Release 1.0.0

11 years agotests: remove g_printerr() that's not needed any longer
Tim-Philipp Müller [Mon, 24 Sep 2012 10:56:56 +0000 (11:56 +0100)]
tests: remove g_printerr() that's not needed any longer

now that tcase_skip_broken_test() prints it as well.

11 years agotests: disable failing replaygain tests
Tim-Philipp Müller [Sun, 23 Sep 2012 18:50:42 +0000 (19:50 +0100)]
tests: disable failing replaygain tests

11 years agosmpte: send stream-start event
Tim-Philipp Müller [Sun, 23 Sep 2012 15:31:37 +0000 (16:31 +0100)]
smpte: send stream-start event

11 years agomultipartmux: send stream-start event
Tim-Philipp Müller [Sun, 23 Sep 2012 15:10:36 +0000 (16:10 +0100)]
multipartmux: send stream-start event

11 years agomatroskamux: send stream-start
Tim-Philipp Müller [Sun, 23 Sep 2012 15:02:19 +0000 (16:02 +0100)]
matroskamux: send stream-start

11 years agoqtmux: send stream-start event
Tim-Philipp Müller [Sun, 23 Sep 2012 14:57:35 +0000 (15:57 +0100)]
qtmux: send stream-start event

11 years agointerleave: add a bunch of FIXMEs
Tim-Philipp Müller [Sun, 23 Sep 2012 14:48:54 +0000 (15:48 +0100)]
interleave: add a bunch of FIXMEs

Needs some more work, so stream-start, caps and tags are
sent in the right order.

11 years agoflvmux: send stream-start event
Tim-Philipp Müller [Sun, 23 Sep 2012 14:18:54 +0000 (15:18 +0100)]
flvmux: send stream-start event

11 years agoavimux: send stream-start event
Tim-Philipp Müller [Sun, 23 Sep 2012 14:16:14 +0000 (15:16 +0100)]
avimux: send stream-start event

11 years agoAutomatic update of common submodule
Tim-Philipp Müller [Sat, 22 Sep 2012 15:08:05 +0000 (16:08 +0100)]
Automatic update of common submodule

From 4f962f7 to 6c0b52c

11 years agortspsrc: answer URI query
Tim-Philipp Müller [Fri, 21 Sep 2012 20:54:36 +0000 (21:54 +0100)]
rtspsrc: answer URI query

Without this, something also answered the query
with TRUE but without setting a uri, not sure
what that was..

11 years agortph264pay: Make sure the caps don't have duplicated sps/pps
Olivier Crête [Thu, 20 Sep 2012 21:28:47 +0000 (17:28 -0400)]
rtph264pay: Make sure the caps don't have duplicated sps/pps

11 years agopulsesrc: Mute stream post-connection if required
Arun Raghavan [Thu, 20 Sep 2012 17:58:12 +0000 (19:58 +0200)]
pulsesrc: Mute stream post-connection if required

A bug in PulseAudio causes PA_STREAM_START_MUTED to be rejected on
record streams. Until this is fixed upstream, we mute the stream
manually at startup. Based on a patch by Alban Browaeys
<prahal@yahoo.com>.

https://bugzilla.gnome.org/show_bug.cgi?id=684469

11 years agoqtdemux: 24 bit audio here is S24LE, not S24_3LE.
Michael Smith [Fri, 21 Sep 2012 01:00:59 +0000 (18:00 -0700)]
qtdemux: 24 bit audio here is S24LE, not S24_3LE.

11 years agov4l2src: handle latency query before setting up the bufferpool
Sjoerd Simons [Thu, 20 Sep 2012 08:07:24 +0000 (10:07 +0200)]
v4l2src: handle latency query before setting up the bufferpool

Fixes crash if no bufferpool is set up yet.

https://bugzilla.gnome.org/show_bug.cgi?id=684430

11 years agoosxaudiosink: Specify endianness in IEC 61937 payloading
Arun Raghavan [Wed, 19 Sep 2012 03:47:03 +0000 (09:17 +0530)]
osxaudiosink: Specify endianness in IEC 61937 payloading

Corresponds to an API change in gst-plugins-base. This needs to be fixed
to query the expected byte order using appropriate API.

https://bugzilla.gnome.org/show_bug.cgi?id=678021

11 years agodirectsoundsink: Specify endianness in IEC 61937 payloading
Arun Raghavan [Wed, 19 Sep 2012 03:45:53 +0000 (09:15 +0530)]
directsoundsink: Specify endianness in IEC 61937 payloading

DirectSound expects native endian byte order.

https://bugzilla.gnome.org/show_bug.cgi?id=678021

11 years agopulsesink: Specify endianness in IEC 61937 payloading
Arun Raghavan [Wed, 19 Sep 2012 03:43:11 +0000 (09:13 +0530)]
pulsesink: Specify endianness in IEC 61937 payloading

Corresponds to an API change in gst-plugins-base.

https://bugzilla.gnome.org/show_bug.cgi?id=678021

11 years agodeinterlace: Remove incorrect logic
Robert Swain [Tue, 18 Sep 2012 22:39:01 +0000 (00:39 +0200)]
deinterlace: Remove incorrect logic

I don't understand why these lines were added, they don't make sense to
me now and both David and I agree that removing them moves closer to
related logic being correct, therefore, they're being removed.

I've tested a few progressive, interlaced and telecine clips and they
all behave properly timestamp-wise and visually after these changes.

11 years agodeinterlace: Fix field duration
Robert Swain [Tue, 18 Sep 2012 22:17:49 +0000 (00:17 +0200)]
deinterlace: Fix field duration

The frame rate fraction is correctly adjusted in the cases preceding the
field duration calculation and so the factor of 2 is incorrect.

11 years agovideobox: Fix U/V strides for a number of cases.
Michael Smith [Tue, 18 Sep 2012 17:34:03 +0000 (10:34 -0700)]
videobox: Fix U/V strides for a number of cases.

11 years agovideomixer: init videoinfo
Mark Nauwelaerts [Tue, 18 Sep 2012 10:13:21 +0000 (12:13 +0200)]
videomixer: init videoinfo

... to prevent random bogus caps fields.

11 years agovideomixer: chain up to collectpads query function
Mark Nauwelaerts [Tue, 18 Sep 2012 10:12:39 +0000 (12:12 +0200)]
videomixer: chain up to collectpads query function

11 years agovideomixer: Don't let GstCollectPad shadow custom sink pad query func
Nicolas Dufresne [Mon, 17 Sep 2012 17:17:00 +0000 (13:17 -0400)]
videomixer: Don't let GstCollectPad shadow custom sink pad query func

In the current implementation, the custom pad query function is not called.
This patch, set that query function on the GstCollectPads to avoid this
shadowing.

See https://bugzilla.gnome.org/show_bug.cgi?id=684237

11 years agotests: dist image.jpg for jpeg test
Tim-Philipp Müller [Mon, 17 Sep 2012 17:23:11 +0000 (18:23 +0100)]
tests: dist image.jpg for jpeg test

11 years agoRelease 0.11.99
Tim-Philipp Müller [Mon, 17 Sep 2012 16:57:58 +0000 (17:57 +0100)]
Release 0.11.99

11 years agodocs: update
Tim-Philipp Müller [Mon, 17 Sep 2012 15:53:04 +0000 (16:53 +0100)]
docs: update

11 years agoFix spec file for vp8 move
Christian Fredrik Kalager Schaller [Mon, 17 Sep 2012 11:30:15 +0000 (13:30 +0200)]
Fix spec file for vp8 move

11 years agoannodex: Add to the CRUFT_DIRS
Sebastian Dröge [Mon, 17 Sep 2012 11:23:36 +0000 (13:23 +0200)]
annodex: Add to the CRUFT_DIRS

11 years agodocs: update
Tim-Philipp Müller [Mon, 17 Sep 2012 11:14:07 +0000 (12:14 +0100)]
docs: update

11 years agovp8enc: Correctly finish frames
Sebastian Dröge [Mon, 17 Sep 2012 07:48:56 +0000 (09:48 +0200)]
vp8enc: Correctly finish frames

Previously we would always get the same frame if multiple frames are pending,
leaking memory of the previous frames and breaking timestamps.

11 years agovp8enc: Allow changing bitrate and other parameters during playback
Sebastian Dröge [Mon, 17 Sep 2012 07:40:41 +0000 (09:40 +0200)]
vp8enc: Allow changing bitrate and other parameters during playback

Fixes bug #648276.

11 years agovp8enc: Store configuration in the vpx_codec_enc_cfg_t struct instead of duplicating...
Sebastian Dröge [Mon, 17 Sep 2012 07:16:39 +0000 (09:16 +0200)]
vp8enc: Store configuration in the vpx_codec_enc_cfg_t struct instead of duplicating all variables

Also protect encoder with a mutex.