Mark Nauwelaerts [Mon, 10 May 2010 16:33:03 +0000 (18:33 +0200)]
qtdemux: fix partial parsing of ctts table
Fixes #616516.
Mark Nauwelaerts [Mon, 10 May 2010 16:32:15 +0000 (18:32 +0200)]
qtdemux: cleanup a comment and add some debug and conditional compilation
Sebastian Dröge [Tue, 11 May 2010 08:01:52 +0000 (10:01 +0200)]
configure: Check for GTK+ 3.0 and if it's not available for GTK+ 2.0
Jan Urbański [Mon, 10 May 2010 20:11:10 +0000 (22:11 +0200)]
flvmux: only store the last buffer timestamp if it's valid
Fixes bug #618305
Olivier Crête [Sat, 9 Jan 2010 03:13:59 +0000 (22:13 -0500)]
rtph264pay: Re-send SPS/PPS when requested
https://bugzilla.gnome.org/show_bug.cgi?id=606689
Mark Nauwelaerts [Fri, 7 May 2010 15:09:16 +0000 (17:09 +0200)]
rtph264pay: fix typo in debug message
Mark Nauwelaerts [Fri, 7 May 2010 13:42:23 +0000 (15:42 +0200)]
rtptheorapay: add config-interval parameter to re-insert config in stream
Add a new config-interval property to instruct the payloader to insert
configuration headers at periodic intervals in the stream
(when a keyframe is countered).
Mark Nauwelaerts [Fri, 7 May 2010 13:31:03 +0000 (15:31 +0200)]
rtptheoradepay: fix in-band configuration parsing
Also make configuration header parsing a bit more relaxed with respect
to length field interpretation.
Mark Nauwelaerts [Fri, 7 May 2010 13:30:30 +0000 (15:30 +0200)]
rtpvorbisdepay: fix in-line configuration parsing
Also make configuration header parsing a bit more relaxed with respect
to length field interpretation.
Mark Nauwelaerts [Tue, 4 May 2010 14:57:35 +0000 (16:57 +0200)]
rtptheorapay: do not discard downstream flow return
Mark Nauwelaerts [Tue, 4 May 2010 14:57:11 +0000 (16:57 +0200)]
rtptheorapay: refactor buffer payloading
Sebastian Dröge [Fri, 7 May 2010 18:41:04 +0000 (20:41 +0200)]
deinterlace: Add support for UYVY
Wim Taymans [Fri, 7 May 2010 17:06:35 +0000 (19:06 +0200)]
rtpsession: fix return value
Wim Taymans [Fri, 7 May 2010 17:02:21 +0000 (19:02 +0200)]
rtspsrc: don't leak the session
Wim Taymans [Fri, 7 May 2010 16:59:42 +0000 (18:59 +0200)]
rtsp: configure bandwidth properties in the session
Wim Taymans [Fri, 7 May 2010 16:58:58 +0000 (18:58 +0200)]
rtpsession: add properties to configure the bandwidth
Add properties to proxy the bandwidth configuration to the session object.
Wim Taymans [Fri, 7 May 2010 16:57:13 +0000 (18:57 +0200)]
rtpsession: add properties to configure bandwidths
Add properties to configure the sender and receiver bandwidths.
Configure the bandwidths before calculating the RTCP timeout when we need to.
Wim Taymans [Fri, 7 May 2010 16:56:30 +0000 (18:56 +0200)]
rtpstats: add some debug info
Wim Taymans [Fri, 7 May 2010 16:55:34 +0000 (18:55 +0200)]
rtpsession: small cleanups
Wim Taymans [Fri, 7 May 2010 14:55:13 +0000 (16:55 +0200)]
rtpstats: make bandwidths more configurable
Add a method to configure the various bandwidths in the session.
Wim Taymans [Fri, 7 May 2010 11:32:30 +0000 (13:32 +0200)]
rtpsession: handle NONE RTCP intervals
Prepare for handling RTCP reporting intervals of GST_CLOCK_TIME_NONE, which
means don't send RTCP at all.
Wim Taymans [Fri, 7 May 2010 10:51:05 +0000 (12:51 +0200)]
rtspsrc: fall back to SDP ports instead of server_port
In multicast, fall back to the ports in the SDP instead of the server_port
attribute as this is more in line with the RFC.
Wim Taymans [Fri, 7 May 2010 10:24:51 +0000 (12:24 +0200)]
rtspsrc: refactor collecting the transport info
Make a method to collect the ports and destination address.
Wim Taymans [Fri, 7 May 2010 09:28:36 +0000 (11:28 +0200)]
rtspsrc: handle servers that send broken Transports
Handle servers that send their port pairs with the wrong name.
Fixes #617537
Wim Taymans [Thu, 6 May 2010 14:52:26 +0000 (16:52 +0200)]
rtspsrc: use the SDP connection info in multicast
Parse the connection info from the SDP.
When we need to configure the multicast destination, fall back to the SDP
connection info when the transport did not specify a destination and ttl.
Fixes #617537
Stefan Kost [Thu, 6 May 2010 12:42:38 +0000 (15:42 +0300)]
goom,monoscope: truncate own caps, instead of copying and using the first only
We got the caps from an intersect, it is our own, hence we can truncate it.
Stefan Kost [Thu, 6 May 2010 12:40:33 +0000 (15:40 +0300)]
pulsesrc: reflow to truncate caps just once
We get writable cpas from the intersection (unless it failed). As we truncate
those anyway, we don't need to manyaly copy the first structure.
Stefan Kost [Thu, 6 May 2010 12:39:31 +0000 (15:39 +0300)]
gdkpixbuf: don't leak template caps
Stefan Kost [Thu, 6 May 2010 12:38:35 +0000 (15:38 +0300)]
auto{audio,video}{src,sink}: use can_intersect to avoid a caps copy
Stefan Kost [Tue, 27 Apr 2010 10:36:35 +0000 (13:36 +0300)]
flvdemux: tell what we can do
Any-caps are bad. If apps scan the registry, they'd like to know what we can
output.
Stefan Kost [Tue, 27 Apr 2010 10:43:29 +0000 (13:43 +0300)]
jpegenc: also lift the arbitrary restrictions for width and height
This was already done for jpegdec.
Sebastian Dröge [Thu, 6 May 2010 12:03:11 +0000 (14:03 +0200)]
pulsesrc: Allocate/free PA mainloop during state changes
...also destroy the stream and context during state changes.
Sebastian Dröge [Thu, 6 May 2010 11:57:01 +0000 (13:57 +0200)]
pulsesink: Allocate and free the custom clock in NULL<->READY
Sebastian Dröge [Thu, 6 May 2010 11:51:59 +0000 (13:51 +0200)]
pulsesink: Create and free the PA mainloop in NULL->READY/READY->NULL
This fixes a race condition, when stopping the mainloop during finalization
is done from a mainloop callback.
Fixes bugs #614765 and #590662.
Sebastian Dröge [Wed, 5 May 2010 17:35:48 +0000 (19:35 +0200)]
videomixer: Make selection of a sinkpad number threadsafe
Sebastian Dröge [Wed, 5 May 2010 15:39:32 +0000 (17:39 +0200)]
deinterlace: Add support for all common RGB formats
Sebastian Dröge [Wed, 5 May 2010 14:06:51 +0000 (16:06 +0200)]
deinterlace: Add support for AYUV
Wim Taymans [Tue, 4 May 2010 14:34:27 +0000 (16:34 +0200)]
rtspsrc: make setup url in a smarter way
Make sure we always separate the base and control url parts with a / when
creating the setup url.
Alessandro Decina [Tue, 4 May 2010 14:04:39 +0000 (16:04 +0200)]
rtspsrc: handle SEEKING queries.
Mark Nauwelaerts [Tue, 4 May 2010 09:13:45 +0000 (11:13 +0200)]
rtpmp4vpay: add config-interval parameter to re-insert config in stream
Add a new config-interval property to instruct the payloader to insert
config (VOSH, VOS, etc) at periodic intervals in the stream
(when a GOP or VOP-I is encountered).
Based on patch by <marc.leeman at gmail.com>
Fixes #607452.
Alessandro Decina [Mon, 3 May 2010 11:26:32 +0000 (13:26 +0200)]
rtpjitterbuffer: move some initialization code from change_state to _init.
Set ->active to TRUE in _init so it can be set to FALSE after creating the
jitterbuffer and it won't be mistakenly reset to TRUE in the change_state
function.
This is needed to start the jitterbuffer as inactive when rtpbin is buffering.
Alessandro Decina [Mon, 3 May 2010 09:56:58 +0000 (11:56 +0200)]
rtpbin: fix a bug handling BUFFERING messages.
If a session exists but has no streams, set the min buffering percent to 0
since it means that we haven't received anything for that session yet.
Alessandro Decina [Mon, 3 May 2010 09:51:37 +0000 (11:51 +0200)]
rtpbin: when a stream is created, pause the jitterbuffer if rtpbin is buffering.
Alessandro Decina [Mon, 3 May 2010 09:23:59 +0000 (11:23 +0200)]
rtpbin: fix a bug calculating stream offsets.
Sebastian Dröge [Sat, 1 May 2010 12:20:59 +0000 (14:20 +0200)]
matroskamux: Write previous cluster's size
This is useful for backwards playback, which should be implemented
in matroskademux at some point.
Sebastian Dröge [Sat, 1 May 2010 12:15:49 +0000 (14:15 +0200)]
matroskademux: Set interlaced flag in the caps if the flag is set in the Matroska file
Sebastian Dröge [Sat, 1 May 2010 12:12:28 +0000 (14:12 +0200)]
matroskamux: Write interlaced flag if the input video content is interlaced
Unfortunately Matroska has no way to specify TFF and friends...
Tim-Philipp Müller [Sat, 1 May 2010 10:25:26 +0000 (11:25 +0100)]
rtp: fix printf format of some debug messages
Tim-Philipp Müller [Sat, 1 May 2010 10:06:53 +0000 (11:06 +0100)]
matroska: init variable to avoid compiler warning on OSX
Fixes (bogus) "'offset' may be used uninitialized in this function"
warning on build bot (also spotted by philn).
David Schleef [Sat, 1 May 2010 00:19:44 +0000 (17:19 -0700)]
qtdemux: UYVY is 4:2:2, not 4:2:0
Sebastian Dröge [Fri, 30 Apr 2010 20:22:25 +0000 (22:22 +0200)]
pulse: Don't compare values of two different enum types
Sebastian Dröge [Fri, 30 Apr 2010 20:13:30 +0000 (22:13 +0200)]
deinterlace: Make automatic detection of interlacing the default
Previously "force deinterlacing" was the default, which is a not very
sensible default for the normal use case where deinterlace should act
in passthrough mode unless interlaced content is present.
Mark Nauwelaerts [Thu, 29 Apr 2010 14:26:49 +0000 (16:26 +0200)]
jpegdec: optimise buffer scanning
Specifically, when needing more data, do not rescan from start next time
around, but resume from last position.
See also #583047.
Mark Nauwelaerts [Thu, 29 Apr 2010 13:38:49 +0000 (15:38 +0200)]
jpegdec: disregard superfluous lines when indirect decoding
Mark Nauwelaerts [Tue, 27 Apr 2010 13:44:39 +0000 (15:44 +0200)]
jpegdec: add support for RGB and grayscale color space
Also refactor src caps negotiation and setting.
Mark Nauwelaerts [Tue, 27 Apr 2010 10:19:22 +0000 (12:19 +0200)]
jpegenc: support more colour spaces and some cleanups
Mark Nauwelaerts [Fri, 30 Apr 2010 10:47:01 +0000 (12:47 +0200)]
jpegenc: more generic sink getcaps
Mark Nauwelaerts [Fri, 30 Apr 2010 10:42:42 +0000 (12:42 +0200)]
jpegdec: more sanity checks on input
Specifically, verify input components / colour space is as code
subsequently expects, thereby avoiding crashes or otherwise bogus output.
Presently, that means 3 components YCbCr colour space, and somewhat
limited sampling factors.
Fixes #600553.
Mark Nauwelaerts [Thu, 22 Apr 2010 10:28:22 +0000 (12:28 +0200)]
rtptheoradepay: also accept in-band configuration
Fixes #574416 (theora).
Mark Nauwelaerts [Thu, 22 Apr 2010 10:27:35 +0000 (12:27 +0200)]
rtpvorbisdepay: also accept in-line configuration
Fixes #574416 (vorbis).
Olivier Crête [Wed, 7 Apr 2010 21:21:55 +0000 (17:21 -0400)]
rtptheoradepay: Ignore packets without a known codebook
Don't produce an error if a packet is received without a valid codebook,
it's possible that the codebook will just be coming later.
See #574416.
Mark Nauwelaerts [Tue, 20 Apr 2010 10:17:26 +0000 (12:17 +0200)]
y4menc: adjust unit test to element behaviour
Benjamin M. Schwartz [Wed, 24 Feb 2010 03:16:39 +0000 (22:16 -0500)]
y4menc: add 4:2:2, 4:1:1, and 4:4:4 output support
Fixes #610902.
Mark Nauwelaerts [Thu, 15 Apr 2010 10:21:56 +0000 (12:21 +0200)]
rtph264depay: DELTA_UNIT marking of output buffers
... which evidently makes (most) sense if output buffers are
actually frames.
Partially based on a patch by
Miguel Angel Cabrera <mad_aluche at hotmail.com>
Fixes #609658.
Mark Nauwelaerts [Fri, 16 Apr 2010 15:21:50 +0000 (17:21 +0200)]
rtph263depay: extra keyframe info from PTYPE header
... as opposed to taking it from h263 payload header, which need not
be so reliable.
Fixes #610172.
Mark Nauwelaerts [Fri, 16 Apr 2010 15:08:47 +0000 (17:08 +0200)]
rtph263depay: also use Picture Start Code to detect packet loss
This ensures a whole frame is dropped if a (start) packet is lost,
rather than relying only on the DISCONT flag.
Mark Nauwelaerts [Fri, 16 Apr 2010 15:06:11 +0000 (17:06 +0200)]
rtph263depay: detect frame start using Picture Start Code
So we stop dropping fragments as soon as there is a picture start (code).
In particular, this prevents dropping the first frame following
initial DISCONT.
Mark Nauwelaerts [Fri, 16 Apr 2010 14:34:06 +0000 (16:34 +0200)]
rtph263depay: handle a few FIXMEs
Mark Nauwelaerts [Fri, 16 Apr 2010 14:27:25 +0000 (16:27 +0200)]
rtph263depay: slightly refactor payload dropping
Mark Nauwelaerts [Fri, 16 Apr 2010 09:53:17 +0000 (11:53 +0200)]
rtph263pay: use found GOBs to apply Mode A payloading
... rather than falling back to sending the whole frame in one packet
if number of GOB startcodes < maximum.
One might take this further and still perform Mode B/C payloading,
but at least this should cater for decent fragments in typical cases.
Fixes #599585.
Mark Nauwelaerts [Wed, 14 Apr 2010 09:53:46 +0000 (11:53 +0200)]
matroskademux: implement push mode seeking
Tim-Philipp Müller [Thu, 29 Apr 2010 19:08:43 +0000 (20:08 +0100)]
docs: update for videofilter plugin merge and add gtk-doc blurb for new property
Sebastian Dröge [Mon, 26 Apr 2010 16:12:46 +0000 (18:12 +0200)]
deinterlace: Improve segment handling a bit
Sebastian Dröge [Mon, 26 Apr 2010 16:05:00 +0000 (18:05 +0200)]
deinterlace: Order caps by amount of contained information
Sebastian Dröge [Mon, 26 Apr 2010 15:25:38 +0000 (17:25 +0200)]
deinterlace: Properly set interlaced field in getcaps
Sebastian Dröge [Sat, 24 Apr 2010 14:28:12 +0000 (16:28 +0200)]
deinterlace: Add planar YUV support to all other simple methods
Sebastian Dröge [Sat, 24 Apr 2010 14:10:06 +0000 (16:10 +0200)]
deinterlace: Add planar YUV support to greedyh method
Sebastian Dröge [Sat, 24 Apr 2010 13:42:07 +0000 (15:42 +0200)]
deinterlace: Add support for planar YUV formats in greedyl method
Sebastian Dröge [Sat, 24 Apr 2010 11:58:03 +0000 (13:58 +0200)]
deinterlace: Add support for Y444, Y42B, I420, YV12 and Y41B
The vfir method supports them and will be used until something else
supports it.
Sebastian Dröge [Sat, 24 Apr 2010 07:16:22 +0000 (09:16 +0200)]
deinterlace: Define deinterlace method base classes as abstract types
Sebastian Dröge [Fri, 23 Apr 2010 15:40:10 +0000 (17:40 +0200)]
deinterlace: Move deinterlacing methods to their own file
Sebastian Dröge [Fri, 23 Apr 2010 15:25:12 +0000 (17:25 +0200)]
deinterlace: Simplify passthrough mode detection
Sebastian Dröge [Fri, 23 Apr 2010 12:35:44 +0000 (14:35 +0200)]
deinterlace: Fix unit test that checks caps handling
deinterlace now always adds the interlaced field to the output caps,
if it wasn't present in the input caps the output caps will still
contain interlaced=false.
Sebastian Dröge [Wed, 21 Apr 2010 15:00:05 +0000 (17:00 +0200)]
deinterlace: Refactor deinterlacing as preparation for supporting more color formats
Sebastian Dröge [Thu, 22 Apr 2010 17:05:37 +0000 (19:05 +0200)]
videobox: Add support for Y444, Y42B and Y41B
Sebastian Dröge [Thu, 22 Apr 2010 13:54:21 +0000 (15:54 +0200)]
videobox: Add support for YVYU and reorder template caps
Sebastian Dröge [Sun, 18 Apr 2010 19:11:21 +0000 (21:11 +0200)]
videobox: Translate navigation events to make sense again upstream
Sebastian Dröge [Sun, 18 Apr 2010 18:58:14 +0000 (20:58 +0200)]
videobox: Properly handle ranges/lists of width or height when transforming caps
Code partly taken from the videocrop element.
Sebastian Dröge [Thu, 22 Apr 2010 13:45:15 +0000 (15:45 +0200)]
alpha: Fix planar YUV->RGB processing
Sebastian Dröge [Thu, 22 Apr 2010 13:42:03 +0000 (15:42 +0200)]
alpha: Correctly clamp after YUV->RGB conversion
Sebastian Dröge [Thu, 22 Apr 2010 13:20:24 +0000 (15:20 +0200)]
alpha: Add support for YUY2, YVYU and UYVY
Sebastian Dröge [Sun, 18 Apr 2010 13:02:42 +0000 (15:02 +0200)]
videobox: Sync properties to the controller in before_transform
Sebastian Dröge [Fri, 16 Apr 2010 15:00:02 +0000 (17:00 +0200)]
videobox: Add support for YUY2 and UYUV
Sebastian Dröge [Wed, 21 Apr 2010 15:41:43 +0000 (17:41 +0200)]
alpha: Refactor processing and add support for other planar YUV formats
This reduces the generated code size by a factor of 2.5.
Sebastian Dröge [Wed, 21 Apr 2010 15:15:33 +0000 (17:15 +0200)]
alpha: Add support for YV12 input
Sebastian Dröge [Thu, 22 Apr 2010 11:56:58 +0000 (13:56 +0200)]
videomixer: Add support for YUY2, YVYU, UYVY
Sebastian Dröge [Tue, 20 Apr 2010 10:18:18 +0000 (12:18 +0200)]
videomixer: Add support for Y444, Y42B, Y41B and YV12
Sebastian Dröge [Wed, 21 Apr 2010 15:07:10 +0000 (17:07 +0200)]
videofilter: Order color formats by their contained amount of information
Sebastian Dröge [Tue, 20 Apr 2010 16:22:16 +0000 (18:22 +0200)]
videoflip: Drop Y41B/Y42B support
Rotating 90°/270° with subsampled YUV where horizontal
and vertical subsampling are different doesn't really work.
Sebastian Dröge [Mon, 19 Apr 2010 12:37:54 +0000 (14:37 +0200)]
videoflip: Also flip the pixel-aspect-ratio if width/height are exchanged