platform/upstream/gst-plugins-base.git
10 years agotypefind: Fix identification of some MPEG files
Reynaldo H. Verdejo Pinochet [Thu, 14 Nov 2013 04:53:31 +0000 (01:53 -0300)]
typefind: Fix identification of some MPEG files

Make sure we begin by peeking at MPEG2_MAX_PROBE_LENGTH
bytes.

Fixes:

https://bugzilla.gnome.org/show_bug.cgi?id=678011

10 years agortpbuffer: Fix gst_rtp_buffer_ext_timestamp() with clang 5 on iOS/ARM
Sebastian Dröge [Wed, 13 Nov 2013 19:12:48 +0000 (20:12 +0100)]
rtpbuffer: Fix gst_rtp_buffer_ext_timestamp() with clang 5 on iOS/ARM

The bitwise NOT operator is not defined on signed integers.
Thanks to Wim Taymans for finding the cause.

https://bugzilla.gnome.org/show_bug.cgi?id=711819

10 years agotests: fix race in streamsynchronizer test
Tim-Philipp Müller [Tue, 12 Nov 2013 18:58:43 +0000 (18:58 +0000)]
tests: fix race in streamsynchronizer test

Wait for thread to exit before starting to free the
to_push list, otherwise thread might check the final
to_push->next node only after we've freed it already.

10 years agovideodecoder: try to negotiate the buffer pool even though there is no o/p format
Sreerenj Balachandran [Mon, 11 Nov 2013 12:10:53 +0000 (14:10 +0200)]
videodecoder: try to negotiate the buffer pool even though there is no o/p format

We could have allocation query before caps event and even without caps inside
the query. In such cases , the downstream can return a bufferpool object with
out actually configuring it. This feature is helpful to negotiate the bufferpool
with out knowing the output video format. For eg: some hardware accelerated
decoders can interpret the o/p video format only after it finishes the decoding
of one buffer at least.

https://bugzilla.gnome.org/show_bug.cgi?id=687183

11 years agoappsrc: Fix deadlock that may occur when multiple threads access appsrc at once
Tom Greenwood [Thu, 7 Nov 2013 15:03:34 +0000 (15:03 +0000)]
appsrc: Fix deadlock that may occur when multiple threads access appsrc at once

https://bugzilla.gnome.org/show_bug.cgi?id=711550

11 years agotagdemux: accumulate buffers in adapter
Wim Taymans [Mon, 4 Nov 2013 08:55:17 +0000 (09:55 +0100)]
tagdemux: accumulate buffers in adapter

Accumulate buffers in an adapter instead of appending them because append causes
a lot of memcpys.
Keep track of the last tagsize and accumulate enough data before attempting to
parse more data.

This patch implements a minimal amount of changes in order to not change the
behaviour. We should really rewrite the tag handling and trimming using
the adapter API instead of merging and trimming into a buffer.

11 years agoadder: Free consistency checker instance in test_live_seeking test
Sebastian Dröge [Wed, 6 Nov 2013 11:16:31 +0000 (12:16 +0100)]
adder: Free consistency checker instance in test_live_seeking test

11 years agoadder: Release some request pads properly in the unit test
Sebastian Dröge [Wed, 6 Nov 2013 11:01:14 +0000 (12:01 +0100)]
adder: Release some request pads properly in the unit test

11 years agoAutomatic update of common submodule
Tim-Philipp Müller [Tue, 5 Nov 2013 11:18:01 +0000 (11:18 +0000)]
Automatic update of common submodule

From 865aa20 to dbedaa0

11 years agodiscoverer: fix build after last commit
Alessandro Decina [Mon, 4 Nov 2013 10:34:38 +0000 (11:34 +0100)]
discoverer: fix build after last commit

Add a forward declaration for my_g_string_append_printf that specifies
G_GNUC_PRINTF. Turn off indent on it as it drives gst-indent crazy.

11 years agodiscoverer: fix -Wformat-nonliteral warning
Alessandro Decina [Mon, 4 Nov 2013 10:17:30 +0000 (11:17 +0100)]
discoverer: fix -Wformat-nonliteral warning

11 years agoaudio: Add unit test for filling memory with silence samples
Sebastian Dröge [Sun, 3 Nov 2013 14:57:54 +0000 (15:57 +0100)]
audio: Add unit test for filling memory with silence samples

11 years agoaudio: Update ORC dist files
Sebastian Dröge [Sun, 3 Nov 2013 11:23:12 +0000 (12:23 +0100)]
audio: Update ORC dist files

11 years agoaudio-format: Use ORC for filling memory with silence samples
Sebastian Dröge [Sun, 3 Nov 2013 11:22:33 +0000 (12:22 +0100)]
audio-format: Use ORC for filling memory with silence samples

11 years agortspconnection: Add new API to the docs and .def file
Sebastian Dröge [Fri, 1 Nov 2013 16:02:22 +0000 (17:02 +0100)]
rtspconnection: Add new API to the docs and .def file

11 years agortspconnection: Fix indention in header
Sebastian Dröge [Fri, 1 Nov 2013 15:43:56 +0000 (16:43 +0100)]
rtspconnection: Fix indention in header

11 years agortspconnection: allow setting tls certificate validation
Aleix Conchillo Flaque [Fri, 1 Nov 2013 14:25:01 +0000 (07:25 -0700)]
rtspconnection: allow setting tls certificate validation

Added new functions gst_rtsp_connection_set_tls_validation_flags() to
allow setting the TLS certificate validation flags when establishing a
TLS connection.
A getter is also available, gst_rtsp_connection_get_tls_validation_flags().

https://bugzilla.gnome.org/show_bug.cgi?id=711231

11 years agosdp: fix duplicate 'const' declaration warnings
Matthieu Bouron [Fri, 1 Nov 2013 14:22:13 +0000 (14:22 +0000)]
sdp: fix duplicate 'const' declaration warnings

https://bugzilla.gnome.org/show_bug.cgi?id=711258

11 years agoplayback: Add subpicture/x-dvb as raw caps
Thibault Saunier [Wed, 16 Oct 2013 19:46:05 +0000 (16:46 -0300)]
playback: Add subpicture/x-dvb as raw caps

https://bugzilla.gnome.org/show_bug.cgi?id=710325

11 years agovideoscale: fix adding borders when NV12 is used
Antonio Ospite [Mon, 28 Oct 2013 11:36:04 +0000 (12:36 +0100)]
videoscale: fix adding borders when NV12 is used

When the frame buffer is NV12 the borders are not added at all, fix that
and fill them to black.

https://bugzilla.gnome.org/show_bug.cgi?id=711003

11 years agovideoconvert: remove unneeded guint comparaison
Matthieu Bouron [Wed, 23 Oct 2013 15:43:32 +0000 (16:43 +0100)]
videoconvert: remove unneeded guint comparaison

https://bugzilla.gnome.org/show_bug.cgi?id=710760

11 years agodiscoverer: also filter 'framed' field when looking for same streams
Stefan Sauer [Mon, 14 Oct 2013 16:45:16 +0000 (18:45 +0200)]
discoverer: also filter 'framed' field when looking for same streams

Fixes extra streams for some mp4 files containing aac audio.

11 years agooggdemux: fix copy'n'paste in comment
Stefan Sauer [Tue, 8 Oct 2013 19:57:11 +0000 (21:57 +0200)]
oggdemux: fix copy'n'paste in comment

11 years agotheoraenc: Do nothing when flushing the encoder when no caps were set
Thibault Saunier [Thu, 10 Oct 2013 18:56:32 +0000 (15:56 -0300)]
theoraenc: Do nothing when flushing the encoder when no caps were set

In case we receive a flush event before having our caps set, we will
end up trying to create a theora encoder even though we are not ready.
Avoid that situation making sure we are initialized before accepting to
be flushed.

https://bugzilla.gnome.org/show_bug.cgi?id=709858

11 years agonavigation: Add missing out parameter annotations to GstNavigation
Stephan Sundermann [Fri, 11 Oct 2013 19:51:00 +0000 (21:51 +0200)]
navigation: Add missing out parameter annotations to GstNavigation

https://bugzilla.gnome.org/show_bug.cgi?id=709938

11 years agoexamples/overlay: handle the case when xvimagesink is not found
Julien Isorce [Thu, 10 Oct 2013 13:09:19 +0000 (14:09 +0100)]
examples/overlay: handle the case when xvimagesink is not found

So that ximagesink can have a chance to be found.
In qtgv-videooverlay.

11 years agoexamples/overlay: unref sink only when found
Julien Isorce [Thu, 10 Oct 2013 13:01:44 +0000 (14:01 +0100)]
examples/overlay: unref sink only when found

In gtk-videooverlay and qt-videooverlay examples.

11 years agoencodebin: Handle changes in encoding_profile::restriction during playback
Thibault Saunier [Mon, 7 Oct 2013 17:52:00 +0000 (14:52 -0300)]
encodebin: Handle changes in encoding_profile::restriction during playback

There are cases where we want to change the restrictions caps during
playback, handle that in encodebin.

https://bugzilla.gnome.org/show_bug.cgi?id=709588

11 years agoalsa: Add channel map API support
Takashi Iwai [Tue, 8 Oct 2013 15:07:02 +0000 (17:07 +0200)]
alsa: Add channel map API support

The initial support for the new ALSA chmap API.
Just translate the current chmap to GstAudioChannelPosition during the
setup.  No function to specify the channel map manually yet, so still
impossible to assign any non-standard positions or to configure in a
different order even if the hardware allows.

https://bugzilla.gnome.org/show_bug.cgi?id=709755

11 years agoaudioringbuffer: Don't clear need_reorder flag too early
Takashi Iwai [Tue, 8 Oct 2013 14:02:46 +0000 (16:02 +0200)]
audioringbuffer: Don't clear need_reorder flag too early

gst_audio_ring_buffer_set_channel_positions() checks whether the given
positions are identical with the current setup and returns
immediately if so.  But it also clears need_reorder flag before this
comparison, thus this flag might be wrongly cleared if the function is
called twice with the same channel positions.

Move the flag clearance after the check.

https://bugzilla.gnome.org/show_bug.cgi?id=709754

11 years agovideotestsrc: improve test for backwards playback
Thiago Santos [Tue, 8 Oct 2013 19:13:58 +0000 (16:13 -0300)]
videotestsrc: improve test for backwards playback

Improve test by checking that timestamps are decreasing

11 years agovideotestsrc: implement duration query
Thiago Santos [Tue, 8 Oct 2013 19:10:54 +0000 (16:10 -0300)]
videotestsrc: implement duration query

Add duration query to videotestsrc, it can answer this query when
the num-buffers property is set.

https://bugzilla.gnome.org/show_bug.cgi?id=709646

11 years agotests: test videotestsrc in reverse playback
Thibault Saunier [Fri, 7 Jun 2013 20:32:23 +0000 (16:32 -0400)]
tests: test videotestsrc in reverse playback

https://bugzilla.gnome.org/show_bug.cgi?id=701813

11 years agovideotestsrc: implement reverse playback
Thiago Santos [Tue, 8 Oct 2013 03:08:34 +0000 (00:08 -0300)]
videotestsrc: implement reverse playback

Decrement the n_frames counter when doing reverse playback to
have timestamps and offsets reducing instead of increasing

https://bugzilla.gnome.org/show_bug.cgi?id=701813

11 years agovideodecoder: don't overflow in bytes<->time conversion
Stefan Sauer [Tue, 8 Oct 2013 07:13:50 +0000 (09:13 +0200)]
videodecoder: don't overflow in bytes<->time conversion

fps_n and _d values can be large and this can overflow a uint. Also fix
copy'n'paste mistake in comments.

11 years agodiscoverer: filter 'parsed' field when checking for same caps
Stefan Sauer [Mon, 7 Oct 2013 20:52:27 +0000 (22:52 +0200)]
discoverer: filter 'parsed' field when checking for same caps

We're checking the caps to see if we got more caps details after a parser got
plugged. This will also have a flipped 'parsed' field. If the field was already
present before the parse the match will fail. Add a function that will do the
check while excluding this field.

11 years agodiscoverer: don't shadow local variables
Stefan Sauer [Mon, 7 Oct 2013 20:51:46 +0000 (22:51 +0200)]
discoverer: don't shadow local variables

11 years agodiscoverer: early return when we have no streams
Stefan Sauer [Mon, 7 Oct 2013 20:51:04 +0000 (22:51 +0200)]
discoverer: early return when we have no streams

11 years agodiscoverer: also log stream-id
Stefan Sauer [Mon, 7 Oct 2013 20:49:52 +0000 (22:49 +0200)]
discoverer: also log stream-id

11 years agodiscoverer: fix quark-mismatch for toc and stream-id
Stefan Sauer [Mon, 7 Oct 2013 16:53:18 +0000 (18:53 +0200)]
discoverer: fix quark-mismatch for toc and stream-id

Seems like a copy'n'paste from 15ee41df.

11 years agodiscoverer: report depth for video
Stefan Sauer [Sat, 5 Oct 2013 19:01:53 +0000 (21:01 +0200)]
discoverer: report depth for video

This was returning 0 in all cases. Use the data from GstVideoFormatInfo instead.

11 years agoaudioconvert: Map buffer as READWRITE if the buffer and memory is writable
Matej Knopp [Fri, 4 Oct 2013 11:57:51 +0000 (13:57 +0200)]
audioconvert: Map buffer as READWRITE if the buffer and memory is writable

and only use the input buffer as temporary buffer in that case.

https://bugzilla.gnome.org/show_bug.cgi?id=709408

11 years agortspconnection: Connect to proxy if specified
Hans Månsson [Mon, 30 Sep 2013 19:46:10 +0000 (21:46 +0200)]
rtspconnection: Connect to proxy if specified

Reference: https://bugzilla.gnome.org/show_bug.cgi?id=708880

11 years agodiscoverer: extract helper to print common stream info
Stefan Sauer [Thu, 3 Oct 2013 17:52:58 +0000 (19:52 +0200)]
discoverer: extract helper to print common stream info

Save some lnes of code by using a helper for common stream info.

11 years agodiscoverer: extract some common code
Stefan Sauer [Wed, 2 Oct 2013 09:27:41 +0000 (11:27 +0200)]
discoverer: extract some common code

Extract code to make a GstDiscovererInfo. Extracts code that sets StreamInfo.

11 years agoplaysink: If the visualisation is changing and reconfiguration is pending, do it...
Sebastian Dröge [Wed, 2 Oct 2013 13:02:44 +0000 (15:02 +0200)]
playsink: If the visualisation is changing and reconfiguration is pending, do it all during reconfiguration

Otherwise we will have two pad blocks that want to use the same mutex
and block each other via the streamlock.

https://bugzilla.gnome.org/show_bug.cgi?id=709210

11 years agowin32: Update defs file
Edward Hervey [Wed, 2 Oct 2013 11:06:03 +0000 (13:06 +0200)]
win32: Update defs file

11 years agopbutils: Add codec-utility funtions to support H265
Sreerenj Balachandran [Wed, 2 Oct 2013 09:26:59 +0000 (12:26 +0300)]
pbutils: Add codec-utility funtions to support H265

https://bugzilla.gnome.org/show_bug.cgi?id=708921

11 years agodescriptions: Add description for H.265
Sebastian Dröge [Tue, 1 Oct 2013 21:17:06 +0000 (23:17 +0200)]
descriptions: Add description for H.265

11 years agotypefind: Add typefind function for H265
Sreerenj Balachandran [Tue, 24 Sep 2013 12:51:46 +0000 (15:51 +0300)]
typefind: Add typefind function for H265

https://bugzilla.gnome.org/show_bug.cgi?id=708680

11 years agoplaybin: make sure elements are in null before disposing
Thiago Santos [Tue, 24 Sep 2013 23:47:52 +0000 (16:47 -0700)]
playbin: make sure elements are in null before disposing

If a pipeline fails to preroll, it might happen that the sinks are
put into READY state from playbin's sink activation, but they are never
set to playsink, so they aren't being managed by a GstBin and will keep
their READY state until they are unreffed, leading to a warning.

Prevent this by always forcing them to NULL when deactivating a group

https://bugzilla.gnome.org/show_bug.cgi?id=708789

11 years agoaudiocdsrc: Don't consider trailing data tracks for MusicBrainz disc id calculation
Johannes Dewender [Sat, 28 Sep 2013 11:19:02 +0000 (13:19 +0200)]
audiocdsrc: Don't consider trailing data tracks for MusicBrainz disc id calculation

MusicBrainz removes trailing data tracks from releases on the server
and also for the calculation of the MusicBrainz Disc ID.

https://bugzilla.gnome.org/show_bug.cgi?id=708991

11 years agoaudioringbuffer: check if acquired in set_timestamp
David Svensson Fors [Mon, 23 Sep 2013 09:35:43 +0000 (11:35 +0200)]
audioringbuffer: check if acquired in set_timestamp

Also use GST_OBJECT_LOCK when accessing object data in set_timestamp.

https://bugzilla.gnome.org/show_bug.cgi?id=702230

11 years agoadder: Don't take channel mask in consideration in mono or stereo
MathieuDuponchelle [Sun, 15 Sep 2013 19:48:43 +0000 (21:48 +0200)]
adder: Don't take channel mask in consideration in mono or stereo

This could cause negotiation to fail.

https://bugzilla.gnome.org/show_bug.cgi?id=708633

11 years agoaudiorate: clip buffer before pushing it
Matej Knopp [Fri, 27 Sep 2013 20:41:28 +0000 (22:41 +0200)]
audiorate: clip buffer before pushing it

https://bugzilla.gnome.org/show_bug.cgi?id=708953

11 years agoaudio: change buffer timestamp when clipping even if data hasn't been trimmed
Matej Knopp [Fri, 27 Sep 2013 20:40:28 +0000 (22:40 +0200)]
audio: change buffer timestamp when clipping even if data hasn't been trimmed

https://bugzilla.gnome.org/show_bug.cgi?id=708952

11 years agopbutils: Add entry for text/x-raw
Matej Knopp [Fri, 27 Sep 2013 20:53:43 +0000 (22:53 +0200)]
pbutils: Add entry for text/x-raw

https://bugzilla.gnome.org/show_bug.cgi?id=708954

11 years agopbutils: add MPEG 2 AAC description
Matej Knopp [Wed, 25 Sep 2013 17:29:24 +0000 (19:29 +0200)]
pbutils: add MPEG 2 AAC description

https://bugzilla.gnome.org/show_bug.cgi?id=708773

11 years agoaudiobasesink: do big correction for large drift
Wim Taymans [Wed, 25 Sep 2013 13:17:32 +0000 (15:17 +0200)]
audiobasesink: do big correction for large drift

If we are using skew slaving and we drift more than twice the allowed amount, do
a big correction to get back on track more quickly.

11 years agoAutomatic update of common submodule
Tim-Philipp Müller [Tue, 24 Sep 2013 17:28:57 +0000 (18:28 +0100)]
Automatic update of common submodule

From 6b03ba7 to 865aa20

11 years agortspconnection: Unset input/output_stream after freeing the GIOStream
Ognyan Tonchev [Tue, 24 Sep 2013 14:26:37 +0000 (16:26 +0200)]
rtspconnection: Unset input/output_stream after freeing the GIOStream

watch->input_stream and watch->output_stream are owned by the GIOStream
and should be unset after freeing the stream.

https://bugzilla.gnome.org/show_bug.cgi?id=708689

11 years agoconfigure: Actually use 1.3.0.1 as version to make configure happy
Sebastian Dröge [Tue, 24 Sep 2013 13:05:21 +0000 (15:05 +0200)]
configure: Actually use 1.3.0.1 as version to make configure happy

11 years agoBack to development
Sebastian Dröge [Tue, 24 Sep 2013 13:00:20 +0000 (15:00 +0200)]
Back to development

11 years agoRelease 1.2.0 1.2.0
Sebastian Dröge [Tue, 24 Sep 2013 12:16:22 +0000 (14:16 +0200)]
Release 1.2.0

11 years agoUpdate .po files
Sebastian Dröge [Tue, 24 Sep 2013 12:14:18 +0000 (14:14 +0200)]
Update .po files

11 years agoplaybin: Make sure to cache context types we did not store yet
Sebastian Dröge [Tue, 24 Sep 2013 10:47:26 +0000 (12:47 +0200)]
playbin: Make sure to cache context types we did not store yet

https://bugzilla.gnome.org/show_bug.cgi?id=708668

11 years agortspconnection: Only create writesrc when it is actually needed
Ognyan Tonchev [Tue, 24 Sep 2013 10:10:00 +0000 (12:10 +0200)]
rtspconnection: Only create writesrc when it is actually needed

Creating a GSource and not attaching it to a context will cause
a leak of it's child sources. That is why we create writesrc right
before attaching it to a context.

https://bugzilla.gnome.org/show_bug.cgi?id=708667

11 years agoadder: send pending segment out before checking for EOS
Mathieu Duponchelle [Sun, 22 Sep 2013 20:55:33 +0000 (22:55 +0200)]
adder: send pending segment out before checking for EOS

Otherwise there would be cases where it would not send its segment
out when the first collected after getting it would already yield EOS.

https://bugzilla.gnome.org/show_bug.cgi?id=708590

11 years agovideo-frame: copy offsets from metadata
Wim Taymans [Thu, 19 Sep 2013 15:25:05 +0000 (17:25 +0200)]
video-frame: copy offsets from metadata

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=708606

11 years agosdp: fix docs
Wim Taymans [Sat, 21 Sep 2013 13:17:54 +0000 (15:17 +0200)]
sdp: fix docs

11 years agoAutomatic update of common submodule
Edward Hervey [Fri, 20 Sep 2013 14:16:42 +0000 (16:16 +0200)]
Automatic update of common submodule

From b613661 to 6b03ba7

11 years agoAutomatic update of common submodule
Tim-Philipp Müller [Thu, 19 Sep 2013 17:42:49 +0000 (18:42 +0100)]
Automatic update of common submodule

From 74a6857 to b613661

11 years agoAutomatic update of common submodule
Tim-Philipp Müller [Thu, 19 Sep 2013 16:34:46 +0000 (17:34 +0100)]
Automatic update of common submodule

From 098c0d7 to 74a6857

11 years agodmabuf: Fix compilation if no mmap is available
Sebastian Dröge [Thu, 19 Sep 2013 14:33:29 +0000 (16:33 +0200)]
dmabuf: Fix compilation if no mmap is available

Also #ifdef some more code paths that don't make sense without mmap.

https://bugzilla.gnome.org/show_bug.cgi?id=708372

11 years agodiscoverer: Switch to playing to handle live URI
Edward Hervey [Thu, 19 Sep 2013 10:58:53 +0000 (12:58 +0200)]
discoverer: Switch to playing to handle live URI

Fixes discovery on dvb://

11 years agoBack to development
Sebastian Dröge [Thu, 19 Sep 2013 09:34:54 +0000 (11:34 +0200)]
Back to development

11 years agoRelease 1.1.90 1.1.90
Sebastian Dröge [Thu, 19 Sep 2013 08:49:58 +0000 (10:49 +0200)]
Release 1.1.90

11 years agoUpdate .po files
Sebastian Dröge [Thu, 19 Sep 2013 08:13:32 +0000 (10:13 +0200)]
Update .po files

11 years agovideodecoder: If there is no PTS at all, assume it starts from the segment start
Olivier Crête [Thu, 19 Sep 2013 00:42:55 +0000 (20:42 -0400)]
videodecoder: If there is no PTS at all, assume it starts from the segment start

This is to make the multifilesrc ! pngdec case work

https://bugzilla.gnome.org/show_bug.cgi?id=688043

11 years agopo: Update translations
Sebastian Dröge [Thu, 19 Sep 2013 07:44:47 +0000 (09:44 +0200)]
po: Update translations

11 years agoplaybin: Implement context caching for sinks that are not in playsink yet
Sebastian Dröge [Wed, 18 Sep 2013 20:05:36 +0000 (22:05 +0200)]
playbin: Implement context caching for sinks that are not in playsink yet

11 years agoplaybin: No need to set the GstContext on the sink before activating it
Sebastian Dröge [Wed, 18 Sep 2013 16:21:54 +0000 (18:21 +0200)]
playbin: No need to set the GstContext on the sink before activating it

This is all handled by the GstBin code now.

11 years agodmabuf: Make it not a singleton
Olivier Crête [Thu, 5 Sep 2013 00:21:54 +0000 (20:21 -0400)]
dmabuf: Make it not a singleton

Makes it easier to track how many users there are

Also make it possible to create a dmabuf struct on systems without mmap,
it just won't be possible to map it.

https://bugzilla.gnome.org/show_bug.cgi?id=707793

11 years agortpbuffer: check for valid payload type
Wim Taymans [Fri, 13 Sep 2013 14:01:42 +0000 (16:01 +0200)]
rtpbuffer: check for valid payload type

The payload type can't be between 72 and 76 because with the marker bit set,
this could be mistaken for an RTCP packet then. We do a relaxed check and
only refuse 72-76 when the marker bit is set. The effect is that when
we try to map an RTCP packet as an RTP packet, we will certainly fail.

11 years agoconfigure: rely solely on pkg-config to find libogg and libvorbis
Tim-Philipp Müller [Fri, 13 Sep 2013 08:17:38 +0000 (09:17 +0100)]
configure: rely solely on pkg-config to find libogg and libvorbis

And get rid of AS_SCRUB_INCLUDES

https://bugzilla.gnome.org/show_bug.cgi?id=707658

11 years agovideoscale: fix 4tap for RGB15 and RGB16
Wim Taymans [Thu, 12 Sep 2013 10:23:35 +0000 (12:23 +0200)]
videoscale: fix 4tap for RGB15 and RGB16

Fix component ordering, it's wrong in both the scanline and merge
function so it cancels eachother out and isn't really a except for
loss of precision of the green component.
Fix calculation of the filter weight

11 years agovideoscale: optimize merge for RGB15 and RGB16
Wim Taymans [Tue, 10 Sep 2013 15:02:44 +0000 (17:02 +0200)]
videoscale: optimize merge for RGB15 and RGB16

11 years agovideoscale: remove redundant MAX
Wim Taymans [Tue, 10 Sep 2013 14:55:25 +0000 (16:55 +0200)]
videoscale: remove redundant MAX

The checks above make it inpossible for the value to be smaller than
what we check against with the MAX call.

11 years agoaudioencoder/decoder: Mark pads as requiring reconfiguration again if negotiation...
Sebastian Dröge [Thu, 12 Sep 2013 07:42:36 +0000 (09:42 +0200)]
audioencoder/decoder: Mark pads as requiring reconfiguration again if negotiation fails

Otherwise we might end up in non-optimal configuration, especially
when a flush happened during reconfiguration.

11 years agovideodecoder/videoencoder: Mark pads as requiring reconfiguration again if negotiatio...
Sebastian Dröge [Thu, 12 Sep 2013 07:35:00 +0000 (09:35 +0200)]
videodecoder/videoencoder: Mark pads as requiring reconfiguration again if negotiation fails

Otherwise we might end up in non-optimal configuration, especially
when a flush happened during reconfiguration.

11 years agopbutils: Add description for TechSmith Screen Capture 2
Matej Knopp [Tue, 10 Sep 2013 19:44:33 +0000 (21:44 +0200)]
pbutils: Add description for TechSmith Screen Capture 2

https://bugzilla.gnome.org/show_bug.cgi?id=707878

11 years agoriff: Add support for TechSmith Screen Capture 2
Matej Knopp [Tue, 10 Sep 2013 19:44:21 +0000 (21:44 +0200)]
riff: Add support for TechSmith Screen Capture 2

https://bugzilla.gnome.org/show_bug.cgi?id=707878

11 years agooggdemux: check for full eos after a pad goes eos in push mode
Thiago Santos [Fri, 6 Sep 2013 18:36:12 +0000 (15:36 -0300)]
oggdemux: check for full eos after a pad goes eos in push mode

After a pad is on EOS, verify if all pads are EOS and return
upstream, avoiding keeping the buffer flow without having more
data to push

11 years agooggdemux: properly handle stop position at seeks in push mode
Thiago Santos [Fri, 6 Sep 2013 18:56:39 +0000 (15:56 -0300)]
oggdemux: properly handle stop position at seeks in push mode

Store the seek stop and seqnum and properly restore them when
receiving the corresponding Segment from upstream. Also fixes
seqnum for converted seek events.

11 years agovideoscale: fix RGB15 masks
Wim Taymans [Tue, 10 Sep 2013 14:16:51 +0000 (16:16 +0200)]
videoscale: fix RGB15 masks

11 years agovideoscale: simplify YUYV and UYVY linear scaling
Wim Taymans [Tue, 10 Sep 2013 14:06:54 +0000 (16:06 +0200)]
videoscale: simplify YUYV and UYVY linear scaling

Simplify the code and make it handle odd width

11 years agovideoscale: small cleanups
Wim Taymans [Tue, 10 Sep 2013 14:05:14 +0000 (16:05 +0200)]
videoscale: small cleanups

Use BLEND macro
Fix NV12 corner case

11 years agovideoscale: fix RGB15 masks
Wim Taymans [Tue, 10 Sep 2013 14:03:08 +0000 (16:03 +0200)]
videoscale: fix RGB15 masks

11 years agovideoscale: simplify nearest scaling
Wim Taymans [Tue, 10 Sep 2013 10:18:16 +0000 (12:18 +0200)]
videoscale: simplify nearest scaling

Round the accumulator to avoid later checks
Remove some bound checks that would never trigger
Fix odd width scaling

11 years agovideoscale: pick nearest line in scaling
Wim Taymans [Tue, 10 Sep 2013 09:31:57 +0000 (11:31 +0200)]
videoscale: pick nearest line in scaling

Use rounding to pick the nearest line instead of rounding down.