Ognyan Tonchev [Tue, 16 Jun 2009 13:04:15 +0000 (15:04 +0200)]
multiudpsink: add support for buffer lists
Add support for BufferList and add a unit test.
Fixes #585842
Tim-Philipp Müller [Mon, 15 Jun 2009 23:02:42 +0000 (00:02 +0100)]
souphttpsrc: reset session state when stopping
Increases the chances that the element is actually reusable.
Tim-Philipp Müller [Mon, 15 Jun 2009 22:49:48 +0000 (23:49 +0100)]
souphttpsrc: log response and request headers and fix some broken indenting
Wim Taymans [Mon, 15 Jun 2009 20:40:35 +0000 (22:40 +0200)]
mp4gdepay: guess constantDuration better
Do a better job at guessing the constantDuration parameter when it is not
present in the caps.
Fixes #585205
Sebastian Dröge [Mon, 15 Jun 2009 19:09:47 +0000 (21:09 +0200)]
warptv: Clean up warptv element and fix some minor bugs and leaks
Sebastian Dröge [Mon, 15 Jun 2009 18:53:23 +0000 (20:53 +0200)]
vertigotv: Clean up vertigotv element and fix some minor bugs and leaks
Sebastian Dröge [Mon, 15 Jun 2009 18:38:01 +0000 (20:38 +0200)]
dicetv: Use guint8 instead of char (which can be signed or unsigned)
Sebastian Dröge [Mon, 15 Jun 2009 18:36:39 +0000 (20:36 +0200)]
shagadelictv: Use guint8/gint8 instead of char (which can be signed or unsigned)
Sebastian Dröge [Mon, 15 Jun 2009 18:31:30 +0000 (20:31 +0200)]
shagadelictv: Clean up element and free all memory in finalize
Sebastian Dröge [Mon, 15 Jun 2009 18:21:58 +0000 (20:21 +0200)]
revtv: Clean up revtv element
Sebastian Dröge [Mon, 15 Jun 2009 18:07:42 +0000 (20:07 +0200)]
quarktv: Simplify some code
Sebastian Dröge [Mon, 15 Jun 2009 18:07:10 +0000 (20:07 +0200)]
quarktv: Use the input data if a NULL buffer is chosen instead of the value 0
Sebastian Dröge [Mon, 15 Jun 2009 18:00:43 +0000 (20:00 +0200)]
quarktv: Fix setting the planes property of quarktv
Setting it to a value<16 would cause crashes before because
current_plane was set to the old number of planes-1. Also
fix calculations for non-2^n planes values.
Sebastian Dröge [Mon, 15 Jun 2009 15:50:41 +0000 (17:50 +0200)]
quarktv: Clean up the quarktv element
Sebastian Dröge [Mon, 15 Jun 2009 15:39:20 +0000 (17:39 +0200)]
effectv: Make elements list constant
Sebastian Dröge [Mon, 15 Jun 2009 15:37:53 +0000 (17:37 +0200)]
edgetv: Clean up edgetv element and fix memory leak
Sebastian Dröge [Mon, 15 Jun 2009 15:21:58 +0000 (17:21 +0200)]
dicetv: Clean up dicetv element and fix some smaller issues
This fixes a memory leak (the dice map) and a crash when
setting the square-bits property before caps are set.
Sebastian Dröge [Mon, 15 Jun 2009 15:20:21 +0000 (17:20 +0200)]
agingtv: Actually use GstController for syncing the properties to timestamps
Sebastian Dröge [Mon, 15 Jun 2009 15:03:38 +0000 (17:03 +0200)]
agingtv: Export some more agingtv properties via GObject properties
Sebastian Dröge [Mon, 15 Jun 2009 13:06:56 +0000 (15:06 +0200)]
agingtv: General cleanup and updating of copyright
Also make the scratch-lines property exported via a GObject
property and initialize/reset the internal state correctly.
Sebastian Dröge [Mon, 15 Jun 2009 13:05:58 +0000 (15:05 +0200)]
agingtv: Store and update state inside the instance struct
This makes the coloraging effect and pits effect visible.
Tim-Philipp Müller [Mon, 15 Jun 2009 14:51:32 +0000 (15:51 +0100)]
pulsesink: ref custom ring buffer class and type in class_init
Hack around thread-safety issues in GObject and our racy _get_type()
functions (we could easily fix the _get_type() functions, but we still
need to hack around the GObject class races until we require a newer
GLib version, I think).
Tim-Philipp Müller [Sun, 14 Jun 2009 18:19:19 +0000 (19:19 +0100)]
Remove a few old example apps from the 0.8 days
Some have been replaced by newer ones, others are demoing elements that
don't exist any longer (not in -good anyway), and others have not been
touched in many years and it seem pointless to keep them around.
Removing these files makes sure we don't have any code in our repository
that uses Gtk+ symbols which are to be removed for GNOME3, and as such
will make some script that greps for this kind of stuff give us a clean
bill of code health. Fixes #585757.
Olivier Crête [Sun, 14 Jun 2009 01:02:45 +0000 (21:02 -0400)]
rtpsirenpay: Remove deprecated symbol
Patch by: Luis Menina
Marvin Schmidt [Sat, 13 Jun 2009 08:43:55 +0000 (10:43 +0200)]
tests: Don't run the flacdec test if the plugin isn't built. Fixes #585630
Patrick Radizi [Fri, 12 Jun 2009 14:06:28 +0000 (16:06 +0200)]
rtspsrc: Add RTP blocksize functionality
Add property to make the client suggest a blocksize to the server.
Fixes #585549
Wim Taymans [Thu, 11 Jun 2009 20:30:06 +0000 (22:30 +0200)]
rtp: update README, fix some typos, mention gstrtpbin
Wim Taymans [Thu, 11 Jun 2009 17:10:53 +0000 (19:10 +0200)]
pulsesink: handle border cases in resampler
Jan Schmidt [Thu, 11 Jun 2009 12:32:22 +0000 (13:32 +0100)]
docs: Bump common. Use upload-doc.mak instead of upload.mak
Remove the local copy of upload.mak in favour of using the shared
upload-doc.make in common/
Jan Schmidt [Thu, 11 Jun 2009 10:39:25 +0000 (11:39 +0100)]
docs: Quieten a couple more docs warnings
Jan Schmidt [Thu, 11 Jun 2009 10:27:26 +0000 (11:27 +0100)]
docs: Remove gtk-doc comment marker
These comment blocks aren't gtk-doc comments and cause annoying noise in
the docs build.
Sebastian Dröge [Thu, 11 Jun 2009 08:05:32 +0000 (10:05 +0200)]
deinterlace: Implement upstream negotation
Sebastian Dröge [Wed, 10 Jun 2009 19:47:40 +0000 (21:47 +0200)]
deinterlace: Improve debugging and clean up some code
Sebastian Dröge [Wed, 10 Jun 2009 12:55:18 +0000 (14:55 +0200)]
deinterlace: Clip buffers to the current segment if possible
Sebastian Dröge [Wed, 10 Jun 2009 12:45:06 +0000 (14:45 +0200)]
deinterlace: Clean up includes and clean up order of instance struct fields
Olivier Crête [Wed, 10 Jun 2009 20:09:56 +0000 (16:09 -0400)]
rtph263pay: Default to doing A, B and C modes, not only A
Sebastian Dröge [Wed, 10 Jun 2009 07:56:11 +0000 (09:56 +0200)]
deinterlace: Fix QoS calculations
The diff is a signed integer, not an unsigned one of course.
In modes other than GST_DEINTERLACE_ALL every frame has twice the
duration of the field duration.
Olivier Crête [Tue, 9 Jun 2009 18:13:31 +0000 (14:13 -0400)]
rtpsirenpay: Put the bitrate in the RTP caps
The MS code seems to require the bitrate to interoperate and
draft-ietf-avt-rtp-g7221-00 also has it.
Sebastian Dröge [Tue, 9 Jun 2009 17:55:36 +0000 (19:55 +0200)]
deinterlace: Implement basic QoS
This change is based on Tim's QoS implementation
for jpegdec.
Sebastian Dröge [Tue, 9 Jun 2009 17:29:51 +0000 (19:29 +0200)]
deinterlace: Directly proxy events/queries to the peer pads
This removes some overhead introduced by the default handlers
that need to iterate over the other pads.
Edward Hervey [Tue, 9 Jun 2009 08:38:52 +0000 (10:38 +0200)]
avidemux: debug_memdump() unknown tags. Refactor junk parsing code.
This makes life slightly easier when debugging avi files.
Edward Hervey [Mon, 8 Jun 2009 06:21:43 +0000 (08:21 +0200)]
rtp: Don't forget to dist the headers for the CELT (de)payloaders.
Wim Taymans [Sun, 7 Jun 2009 18:54:06 +0000 (20:54 +0200)]
Revert "Revert "qtdemux: fill timestamp table completely""
This reverts commit
9f022c8a8503c2ce0fa617fdb50e41706dd412f5.
Sorry, I was thinking about the wrong module.
Wim Taymans [Sun, 7 Jun 2009 18:49:50 +0000 (20:49 +0200)]
Revert "qtdemux: fill timestamp table completely"
This reverts commit
790b050fc5302cae89cddcd23b258093967d05a9.
I forgot we were frozen.
Wim Taymans [Sun, 7 Jun 2009 18:46:45 +0000 (20:46 +0200)]
qtdemux: fill timestamp table completely
When there are less timestamps that there are samples, fill up the sample table
with the last know timestamp. This situation can happen when the last sample
does not decode and doesn't need a timestamp. We however calculate the total
track length using the last sample timestamp so we need to have something
sensible in there.
Fixes #585056
Wim Taymans [Sun, 7 Jun 2009 11:37:04 +0000 (13:37 +0200)]
wavparse: handle LIST INFO of 0 size
Handle LIST INFO chunks of 0 size instead of causing errors.
Fixes #584981
Wim Taymans [Sun, 7 Jun 2009 11:24:45 +0000 (13:24 +0200)]
Revert "wavparse: Remove dead assignments, move variable to where it's needed."
Reverts commit
44256a78f8dd79a91f3bb2ab7c3aa623c097bb8a and use the result in
error reporting so that we can see what's going on.
Wim Taymans [Fri, 5 Jun 2009 16:55:02 +0000 (18:55 +0200)]
celtdepay: add CELT depayloader
Wim Taymans [Fri, 5 Jun 2009 13:30:51 +0000 (15:30 +0200)]
rtpceltpay: add CELT RTP payloader
Jan Schmidt [Fri, 5 Jun 2009 15:54:48 +0000 (16:54 +0100)]
sunaudio: Fix switch setting on some devices. Add debug. Fix a FIXME.
Fix the setting of toggle switches on some broken audio drivers which
report that no audio ports are settable by ignoring the mod_port field
there.
Add some debug statements.
Fix a FIXME now that Good relies on a new enough gst-plugins-base.
Jan Schmidt [Thu, 4 Jun 2009 11:27:19 +0000 (12:27 +0100)]
sunaudio: Support new flags for options and actions
Use new audio mixer flags added in Base 0.10.23 to expose flags and options
on the SunAudio devices.
Fixes: #583593
Patch By: Brian Cameron <brian.cameron@sun.com>
Patch By: Garrett D'Amore <garrett.damore@sun.com>
Sebastian Dröge [Fri, 15 May 2009 09:50:38 +0000 (11:50 +0200)]
deinterlace: First try to handle DVD still frames correctly
This helps a bit with bug #582740 but still doesn't make it work.
Stefan Kost [Thu, 4 Jun 2009 14:37:03 +0000 (17:37 +0300)]
pulsesink: only notify if all checks passed
Replace goto done: with return, as those are checks when we don't want to flag a
pending notify.
Wim Taymans [Thu, 4 Jun 2009 13:19:05 +0000 (15:19 +0200)]
rtspsrc: set the right state on rtpbin
We need to set the state of gstrtpbin to the same state as our source elements.
This fixes fallback to TCP again.
Stefan Kost [Wed, 3 Jun 2009 15:23:53 +0000 (18:23 +0300)]
pulsesink: check pointer before accessing
Move existing check a few lines up, so that we check before accessing fields.
Stefan Kost [Wed, 3 Jun 2009 15:21:12 +0000 (18:21 +0300)]
pulsesink: rename gst_pulse_sink_get_time to gst_pulsesink_get_time
Rename internal method for consistency.
Stefan Kost [Wed, 3 Jun 2009 15:19:22 +0000 (18:19 +0300)]
pulsesink: use values from pa_stream_get_buffer_attr()
We were putting the requested values back into ringbuffer spec, instead of
using the queried values.
Wim Taymans [Tue, 2 Jun 2009 17:32:21 +0000 (19:32 +0200)]
vrawpay: trim output buffers
Remove the leftover unused bytes in the output buffer.
Fixes #584613
Wim Taymans [Tue, 2 Jun 2009 17:30:30 +0000 (19:30 +0200)]
vrawdepay: fix parsing of sampling field
commit
a12d9a80f225be97b3674b1a0506ac66544dbf49 broke the parsing of the
sampling.
Jan Schmidt [Wed, 27 May 2009 16:06:34 +0000 (17:06 +0100)]
pngdec: Avoid possible overflow in calculations
A malformed (or simply huge) PNG file can lead to integer overflow in
calculating the size of the output buffer, leading to crashes or buffer
overflows later. Fixes SA35205 security advisory.
Tim-Philipp Müller [Mon, 1 Jun 2009 23:48:00 +0000 (00:48 +0100)]
flacenc: some more logging - dump header packets
Also, the final fixing up of the headers is expected and not something
we should warn about.
Tim-Philipp Müller [Mon, 1 Jun 2009 23:37:15 +0000 (00:37 +0100)]
flacenc: never ever pass values >36bits to _set_total_samples_estimate()
Let's be paranoid and make sure we never pass a number that takes up
more than 36 bits to _set_total_samples_estimate(), since libFLAC
expects all the other bits to be zero, and if this is not the case
neighbouring fields in the global stream info header may get messed
up inadvertently, so that flac -d refuses to decode the stream.
See #584455.
Thomas Vander Stichele [Mon, 1 Jun 2009 20:33:02 +0000 (22:33 +0200)]
Address bad FLAC sample length encoding of #5844455
Commit
df707c666433a78d3878af6f055698d5756226c4
introduced an obvious bug in the sample length calculation,
using the wrong macro for conversion.
Brian Cameron [Mon, 1 Jun 2009 18:58:21 +0000 (11:58 -0700)]
deinterlace: Fix spurious colons in asm code
Fixes #584174.
Signed-off-by: David Schleef <ds@schleef.org>
Tim-Philipp Müller [Sun, 31 May 2009 23:40:55 +0000 (00:40 +0100)]
avidemux: skip JUNK chunks in data section in streaming mode
Skip JUNK tags in streaming mode as well instead of EOSing
prematurely. Fixes #564100.
Sebastian Dröge [Thu, 28 May 2009 12:01:17 +0000 (14:01 +0200)]
videomixer: Don't use // comments
Sebastian Dröge [Thu, 28 May 2009 11:56:15 +0000 (13:56 +0200)]
videomixer: Fix background blitting when a color mode is selected with BGRA
Sebastian Dröge [Thu, 28 May 2009 11:54:14 +0000 (13:54 +0200)]
videomixer: Some cleanup and fix the calculation of the frame size in bytes
Sebastian Dröge [Thu, 28 May 2009 11:35:52 +0000 (13:35 +0200)]
videomixer: Fix I420 blending to actually do something
For this we a) implement the checkers filling and b)
actually blend the src/dest by using the src alpha value
from the pad.
Sebastian Dröge [Thu, 28 May 2009 11:14:13 +0000 (13:14 +0200)]
videomixer: Fix ARGB blending to actually work
Sebastian Dröge [Thu, 28 May 2009 11:04:51 +0000 (13:04 +0200)]
videomixer: Blend BGRA ourselves instead of using Cairo
Alex Ugarte [Thu, 28 May 2009 10:55:16 +0000 (12:55 +0200)]
videomixer: Add support for blending BGRA and AYUV
Fixes bug #577017.
Ghislain 'Aus' Lacroix [Thu, 28 May 2009 10:39:46 +0000 (12:39 +0200)]
equalizer: Use floating point arithmetic internally for the int16 mode
By using int32 arithmetic we will introduce distortions as the
IIR filter is very sensitive to rounding errors. Fixes bug #580214.
Christian Schaller [Thu, 28 May 2009 09:55:16 +0000 (10:55 +0100)]
Update spec file with latest plugins
Jan Schmidt [Tue, 26 May 2009 16:19:08 +0000 (17:19 +0100)]
Automatic update of common submodule
From 888e0a2 to c572721
Stefan Kost [Tue, 26 May 2009 13:20:35 +0000 (16:20 +0300)]
v4l2: cleanup and commenting
Remove newlines inserted by gst-indent once. Remove unused var from instance
struct. Add comments. Add another #define for default property value.
Stefan Kost [Wed, 6 May 2009 09:43:35 +0000 (12:43 +0300)]
makefile: idea about makeing more sources/sinks testable again
John Keeping [Mon, 25 May 2009 14:33:35 +0000 (16:33 +0200)]
pngdec: match g_malloc() with g_free()
Matching g_malloc() with a g_free() is important when a custom allocator is
installed.
Fixes #583803
Wim Taymans [Tue, 12 May 2009 16:39:28 +0000 (18:39 +0200)]
rtpmp4vpay: don't look for headers in some cases
In some streams (starting with
00000100) don't look for the headers but push
data as it is.
Fixes #582153
Patrick Radizi [Wed, 13 May 2009 09:50:22 +0000 (11:50 +0200)]
rtspsrc: fix memory leak of messages
Free messages correctly.
Fixes #577318
Wim Taymans [Sun, 24 May 2009 17:32:17 +0000 (19:32 +0200)]
rtspsrc: make fakesrc silent
Make the fakesrc that is responsible for sending dummy packets silent.
Wim Taymans [Sun, 24 May 2009 14:33:42 +0000 (16:33 +0200)]
rtspsrc: don't send teardown before setup
Don't send a TEARDOWN request when we did not manage to successfully setup a
stream.
Sebastian Dröge [Thu, 14 May 2009 12:46:14 +0000 (14:46 +0200)]
matroskademux: Populate a GstIndex that is set on matroskademux
Sebastian Dröge [Thu, 14 May 2009 08:35:22 +0000 (10:35 +0200)]
flvmux: Get the max duration from upstream if there's no duration tag
Sebastian Dröge [Thu, 14 May 2009 08:29:49 +0000 (10:29 +0200)]
flvmux: Write an index table to the end of the file
Tim-Philipp Müller [Fri, 22 May 2009 00:12:35 +0000 (01:12 +0100)]
autotools: move the -Wno-portability from autogen.sh to configure.ac
If we're lucky it'll get used on automatic rebuilds as well that way.
Tim-Philipp Müller [Fri, 22 May 2009 00:10:12 +0000 (01:10 +0100)]
m4: fix 'suspicious cache id' warnings
and update common to pull in a similar fix. Also check in configure
whether the compiler supports do while macros (GLib wants this
defined and it is needed to avoid warnings with some c++ compilers
apparently).
Zeeshan Ali (Khattak) [Thu, 21 May 2009 22:39:33 +0000 (01:39 +0300)]
souphttpsrc: Bump-up libsoup-2.24 dep to >= 2.26
The helper function soup_message_headers_get_content_type that we now use
was added in 2.26.
Zeeshan Ali (Khattak) [Wed, 20 May 2009 14:57:59 +0000 (17:57 +0300)]
souphttpsrc: Set caps for audio/L16 content-type
When "Content-Type" header is "audio/L16", we need to set the caps on the
outgoing buffers so that downstream elements can have means to detect the
stream type and handle it appropriately. Tested with HTTP stream provided
by pulse-audio's http module (git master).
Zeeshan Ali (Khattak) [Wed, 20 May 2009 12:06:25 +0000 (15:06 +0300)]
souphttpsrc: Rename icy_caps to src_caps
Philippe Normand [Thu, 21 May 2009 21:39:13 +0000 (23:39 +0200)]
jpegdec: bump max size to 65535x65535
Remove artificial jpeg image limits.
Fixes #583048.
Jan Schmidt [Thu, 21 May 2009 20:36:02 +0000 (21:36 +0100)]
win32: Update the win32 config.h
Jan Schmidt [Tue, 19 May 2009 14:12:09 +0000 (15:12 +0100)]
matroskademux: Recognise PGS subpicture streams - the bluray format.
Recognise and apply appropriate caps to PGS (Presentation Graphic Stream)
subpicture streams.
Jan Schmidt [Fri, 15 May 2009 09:42:19 +0000 (10:42 +0100)]
pulsesink: Convert an erroneous assertion
Occasionally, we get a change callback for an old stream, triggering
the assertion unnecessarily. Just ignore such callbacks.
Olivier Crête [Wed, 20 May 2009 20:14:40 +0000 (16:14 -0400)]
pulse: Print a warning on under/overflows
Wim Taymans [Wed, 20 May 2009 16:45:45 +0000 (18:45 +0200)]
qtdemux: parse in24 boxes to get endianness
in24 samples are normally big-endian but an enda box can change this to
little-endian. Recurse into the in24 box and find the enda box so that we get
the endianness right.
Fixes #582515
Wim Taymans [Wed, 20 May 2009 12:14:31 +0000 (14:14 +0200)]
multipartdemux: add proper padtemplate
Wim Taymans [Wed, 20 May 2009 12:02:43 +0000 (14:02 +0200)]
multipartdemux: add more mime types
Add mime-type for Panasonic g726 and add more required caps properties for other
G726 mime-types.
Make mime-types case insensitive.
See #582169
Wim Taymans [Wed, 20 May 2009 11:47:52 +0000 (13:47 +0200)]
multipartdemux: add flow aggregation
Arnout Vandecappelle [Wed, 20 May 2009 11:29:02 +0000 (13:29 +0200)]
multipartdemux: allow content to be empty.
gst_adapter_take_buffer doesn't allow buffer to be empty.
Simply skip any part where the content is empty. Don't
create a pad for it either.
See #582169