Vincent Penquerc'h [Wed, 9 Apr 2014 10:04:32 +0000 (11:04 +0100)]
mfc: fix left crop change test
It was testing a noop
Coverity 1139643
Vincent Penquerc'h [Wed, 9 Apr 2014 10:02:00 +0000 (11:02 +0100)]
opus: add missing va_end in variadic function
Coverity 1139944
Vincent Penquerc'h [Wed, 9 Apr 2014 10:00:22 +0000 (11:00 +0100)]
shm: remove dead code
Coverity 1139683
Vincent Penquerc'h [Wed, 9 Apr 2014 09:56:16 +0000 (10:56 +0100)]
mfc: reset "in use" flag on failure to initialize
Vincent Penquerc'h [Wed, 9 Apr 2014 09:55:46 +0000 (10:55 +0100)]
mfc: test for allocation failure before dereferencing
Coverity 1139849
Vincent Penquerc'h [Wed, 9 Apr 2014 09:33:10 +0000 (10:33 +0100)]
mpeg2enc: remove dead assignment
Coverity 1139829
Vincent Penquerc'h [Wed, 9 Apr 2014 09:28:46 +0000 (10:28 +0100)]
gl: pass large structure by const pointer, not value
Avoids large pointless memcpy.
Coverity 206236, 206237
Mohammed Sameer [Tue, 8 Apr 2014 18:39:20 +0000 (21:39 +0300)]
camerabin: add any feature to viewfinderbin static pad template
https://bugzilla.gnome.org/show_bug.cgi?id=727857
Mohammed Sameer [Tue, 8 Apr 2014 17:54:26 +0000 (20:54 +0300)]
camerabin: report an error if we fail to link vfsrc and viewfinder queue
https://bugzilla.gnome.org/show_bug.cgi?id=727855
Tim-Philipp Müller [Tue, 8 Apr 2014 16:45:46 +0000 (17:45 +0100)]
d3dvideosink: post proper error message when window disappears
Vincent Penquerc'h [Tue, 8 Apr 2014 16:34:13 +0000 (17:34 +0100)]
flite: fix off by one in channel mask building
The first loop would shift 1 by 64, which is either undefined
or implementation defined, instead of clearing the top bit.
Coverity 1197691
Vincent Penquerc'h [Tue, 8 Apr 2014 16:10:27 +0000 (17:10 +0100)]
hls: restore NULL test mistakenly removed
Thanks to tpm for point out I'm an idiot.
Vincent Penquerc'h [Tue, 8 Apr 2014 15:53:54 +0000 (16:53 +0100)]
hls: only set DISCONT flag on a valid buffer
Recent refactoring causes this code to be called with either a NULL
fragment, or a non NULL fragment. In the former case, we don't have
a buffer. In the latter case, the original code dealing with DISCONT
assumed the buffer was valid. Testing for a NULL buffer here thus
does not seem to change the intent, and fixes:
Coverity 1195147
Tim-Philipp Müller [Tue, 8 Apr 2014 15:53:41 +0000 (16:53 +0100)]
mxfdemux: guard against NULL material track
Just to be on the safe side.
Vincent Penquerc'h [Tue, 8 Apr 2014 15:46:56 +0000 (16:46 +0100)]
hls: bring NULL test before dereference
Coverity 1195168
Vincent Penquerc'h [Tue, 8 Apr 2014 15:23:50 +0000 (16:23 +0100)]
gl: test for frame NULLness before dereferencing it
Coverity 1195172, 1195171
Jan Schmidt [Tue, 8 Apr 2014 14:44:08 +0000 (00:44 +1000)]
dvbsuboverlay: Avoid infinite loops on short data
Tim-Philipp Müller [Tue, 8 Apr 2014 15:09:16 +0000 (16:09 +0100)]
dvb: fix compilation
Vincent Penquerc'h [Tue, 8 Apr 2014 14:54:01 +0000 (15:54 +0100)]
gl: fix leaks
As the relevant variables are initialized to 0/NULL, we can loop
over the full range and make sure we free partial allocations
when an error happens partway through initialization.
Vincent Penquerc'h [Tue, 8 Apr 2014 14:26:04 +0000 (15:26 +0100)]
dvb: reject too long socket paths
Coverity 206004
Tim-Philipp Müller [Tue, 8 Apr 2014 13:33:06 +0000 (14:33 +0100)]
mxfdemux: implement simple KEY_UNIT seeking
If a KEY_UNIT seek was requested, adjust segment
start to position of the key frame.
Vincent Penquerc'h [Tue, 8 Apr 2014 13:24:31 +0000 (14:24 +0100)]
resindvd: guard against overflow in audio subtitle streams
Turns out there was the same issue as with subtitles.
There is space for a single audio stream, but up to 255
may be used based on a uint8_t value in a struct, which may
or may not be read from the (untrusted) data.
A comment in ifo_types.h says this value is either 0 or 1, so
we can ensure this here without drawbacks.
Coverity 1139585
Vincent Penquerc'h [Tue, 8 Apr 2014 13:19:29 +0000 (14:19 +0100)]
resindvd: guard against overflow in menu subtitle streams
There is space for a single subtitle stream, but up to 255
may be used based on a uint8_t value in a struct, which may
or may not be read from the (untrusted) data.
A comment in ifo_types.h says this value is either 0 or 1, so
we can ensure this here without drawbacks.
Coverity 1139586
Tim-Philipp Müller [Tue, 8 Apr 2014 12:26:49 +0000 (13:26 +0100)]
mxfdemux: fix seeking, send data starting from a key unit
Fixes multiple seeking issues. When doing ACCURATE or normal
non-KEYUNIT seeks, mxfdemux would just send data from the
edit unit that covered the seek position, whether that's
a keyframe or not. Decoders would only output things from
the next keyframe then, which means there's a gap between
the start of the segment and the first decoded data in
some cases. In combination with gst-editing-services this
might result in a frozen picture for the duration of that
gap at the beginning (if videorate fixes up the first
buffer's start timestamp to cover the entire gap), or
a black frame (if no videorate is used and videomixer
fills the gap). Also fixes A/V sync issue when requesting
a KEYUNIT seek.
Vincent Penquerc'h [Tue, 8 Apr 2014 11:37:30 +0000 (12:37 +0100)]
dtsdec: fix buffer overflows
Channels can be up to 7. Also add a guard in case more can be
returned in the future.
Coverity 1139820, 1139821
Xavi Artigas [Tue, 8 Apr 2014 11:20:54 +0000 (13:20 +0200)]
directsoundsrc: Fix critical due to missing debug category initialization
https://bugzilla.gnome.org/show_bug.cgi?id=727812
Vincent Penquerc'h [Tue, 8 Apr 2014 11:20:40 +0000 (12:20 +0100)]
festival: fix fd leak
Coverity 1139831
Edward Hervey [Tue, 8 Apr 2014 09:26:38 +0000 (11:26 +0200)]
speed: Clarify passthrough in switch statement
We do want the caps event to be forwarded downstream
CID #1139754
Edward Hervey [Tue, 8 Apr 2014 09:23:50 +0000 (11:23 +0200)]
soundtouch: Add missing break in switch
Whoops, modifying the pitch when modifying the output rate ...
CID #1139749
Edward Hervey [Tue, 8 Apr 2014 09:21:44 +0000 (11:21 +0200)]
rsvg: Add missing break in switch statement
This wasn't fatal before (worst case it would clear an already empty
adapter).
CID #1139758
Edward Hervey [Tue, 8 Apr 2014 09:20:29 +0000 (11:20 +0200)]
mpeg2enc: Add missing break in switch
QUERY_CAPS is fully handled, we don't need to passthrough to the default
handler.
CID #1139756
Edward Hervey [Tue, 8 Apr 2014 07:51:08 +0000 (09:51 +0200)]
bz2: Initialize variables
There is a small chance that we might end up in the done step without
having any output available.
Furthermore, when going through not_ready, we need to ensure gst_buffer_unmap
has a properly initialized GstMapInfo.
CID #1139923
CID #1139924
CID #1139919
CID #1139920
Edward Hervey [Tue, 8 Apr 2014 05:48:28 +0000 (07:48 +0200)]
check: Fix mpegtsmux caps
Incoming streams need to be parsed. Doesn't fix the test completely though
Edward Hervey [Tue, 8 Apr 2014 05:47:15 +0000 (07:47 +0200)]
mpegtsmux: Initialize helper mpegts library
We now use that library, we need to initialize it so that debug
categories (amongst other things) get properly initialized
Vincent Penquerc'h [Mon, 7 Apr 2014 12:25:56 +0000 (13:25 +0100)]
mxf: fix stop date parsing using the wrong input
Coverity 1139645
Vincent Penquerc'h [Mon, 7 Apr 2014 12:22:22 +0000 (13:22 +0100)]
jp2kdecimator: fix copy/paste errors in expressions
Correct formulae found in ITU T.800, B.12.1.[345]
Coverity 1139649, 1139650, 1139651
Vincent Penquerc'h [Fri, 4 Apr 2014 14:30:20 +0000 (15:30 +0100)]
tests: fix shm test deadlock
The test was not bringing the source back to NULL, so its thread
could start creating a buffer while the pads were being shutdown.
Sebastian Dröge [Sun, 6 Apr 2014 09:57:12 +0000 (11:57 +0200)]
glimagesink: Create GL context and set up window from the streaming thread
gst_gl_context_create() might need to dispatch some operations to the
application's main thread, and calling this in the change_state function
can cause deadlocks.
Sebastian Dröge [Fri, 4 Apr 2014 07:19:39 +0000 (09:19 +0200)]
configure: Fix check for iOS for the OpenGL support
There is also an i386 version of iOS, which is for the simulator.
Better use our already existing HAVE_IOS check instead of relying
on the host triplet.
Wim Taymans [Thu, 3 Apr 2014 09:54:41 +0000 (11:54 +0200)]
wildmidi: fix possible uninitialized variables
Wim Taymans [Thu, 3 Apr 2014 09:48:07 +0000 (11:48 +0200)]
mpdparse: only add location when node existed
Actually return FALSE when the location node was non-existing
Only add the location node when it existed
Sebastian Dröge [Wed, 2 Apr 2014 21:49:37 +0000 (23:49 +0200)]
mpegtsmux: Require parsed/framed input for most of the supported formats
https://bugzilla.gnome.org/show_bug.cgi?id=719519
Sebastian Dröge [Wed, 2 Apr 2014 21:08:36 +0000 (23:08 +0200)]
d3dvideosink: Only pass a dest rectangle if set, otherwise pass NULL
Call with an uninitialized rectangle will cause errors.
https://bugzilla.gnome.org/show_bug.cgi?id=714998
Tim-Philipp Müller [Wed, 2 Apr 2014 12:48:02 +0000 (13:48 +0100)]
tests: fix gl unit tests for recent API changes
Matthew Waters [Wed, 2 Apr 2014 12:05:47 +0000 (23:05 +1100)]
gl: pass video info's by reference
Matthew Waters [Wed, 2 Apr 2014 12:05:11 +0000 (23:05 +1100)]
gl: fix array initialization
Matthew Waters [Wed, 2 Apr 2014 11:54:55 +0000 (22:54 +1100)]
glbumper: remove redundant check
Matthew Waters [Wed, 2 Apr 2014 11:53:10 +0000 (22:53 +1100)]
gl/glx: use the context's display rather than asking for the window's
Matthew Waters [Wed, 2 Apr 2014 11:43:41 +0000 (22:43 +1100)]
gl: fix assignment of temporary variables
Matthew Waters [Wed, 2 Apr 2014 11:42:50 +0000 (22:42 +1100)]
gl: avoid adding a NULL pool to propose allocation
Matthew Waters [Wed, 2 Apr 2014 01:38:28 +0000 (12:38 +1100)]
gl/win32: actually build the gl plugins on win32
Regression from
c8ff215b02ee178513e076e2dfbd14e9ce77ec49
Matthew Waters [Tue, 1 Apr 2014 22:29:56 +0000 (09:29 +1100)]
gl/win32: fix detection of GL/wglext.h
Was resulting in 'present but cannot be compiled' errors
Matthew Waters [Tue, 1 Apr 2014 10:48:26 +0000 (21:48 +1100)]
gl: fix detection of extensions with GL versions < 3
Mesa, for example returns valid pointers for glGetIntegerv and
glGetStringi even if the gl version is less than that required for
both those functions to supposedly exist.
https://bugzilla.gnome.org/show_bug.cgi?id=727324
Stefan Ringel [Sun, 30 Mar 2014 19:02:45 +0000 (21:02 +0200)]
mpegts: fix docs
Signed-off-by: Stefan Ringel <linuxtv@stefanringel.de>
https://bugzilla.gnome.org/show_bug.cgi?id=727356
Stefan Ringel [Sun, 30 Mar 2014 19:02:10 +0000 (21:02 +0200)]
mpegts: add dvb-t2 delivery system descriptor
Signed-off-by: Stefan Ringel <linuxtv@stefanringel.de>
https://bugzilla.gnome.org/show_bug.cgi?id=727356
Stefan Ringel [Mon, 24 Mar 2014 16:44:47 +0000 (17:44 +0100)]
mpegts: bugfix delivery system descriptors
Signed-off-by: Stefan Ringel <linuxtv@stefanringel.de>
https://bugzilla.gnome.org/show_bug.cgi?id=726979
Stefan Ringel [Thu, 27 Mar 2014 20:22:32 +0000 (21:22 +0100)]
mpegts: use GST_READ_UINT32_BE
Signed-off-by: Stefan Ringel <linuxtv@stefanringel.de>
https://bugzilla.gnome.org/show_bug.cgi?id=727187
Stefan Ringel [Thu, 27 Mar 2014 20:21:23 +0000 (21:21 +0100)]
mpegts: add parsing data broadcast descriptor
Signed-off-by: Stefan Ringel <linuxtv@stefanringel.de>
https://bugzilla.gnome.org/show_bug.cgi?id=727187
Stefan Ringel [Thu, 27 Mar 2014 20:18:36 +0000 (21:18 +0100)]
mpegts: add parsing stream identifier
Signed-off-by: Stefan Ringel <linuxtv@stefanringel.de>
https://bugzilla.gnome.org/show_bug.cgi?id=727187
Stefan Ringel [Thu, 27 Mar 2014 12:56:26 +0000 (13:56 +0100)]
mpegts: use glib alloc and free
Signed-off-by: Stefan Ringel <linuxtv@stefanringel.de>
https://bugzilla.gnome.org/show_bug.cgi?id=727159
Matthieu Bouron [Sat, 29 Mar 2014 17:26:03 +0000 (17:26 +0000)]
dashdemux: use correct print format for gint64
https://bugzilla.gnome.org/show_bug.cgi?id=727385
Sebastian Dröge [Sun, 30 Mar 2014 16:52:38 +0000 (18:52 +0200)]
hlsdemux: Handle errors when switching playlists properly
Sebastian Dröge [Sun, 30 Mar 2014 16:12:07 +0000 (18:12 +0200)]
hlsdemux: Go EOS if the end of the segment is reached
Sebastian Dröge [Sun, 30 Mar 2014 16:06:51 +0000 (18:06 +0200)]
hlsdemux: Store buffer end position in segment.position
Sebastian Dröge [Sat, 29 Mar 2014 09:19:52 +0000 (10:19 +0100)]
mpegtspacketizer: Fix typo in debug output
Sebastian Dröge [Fri, 28 Mar 2014 17:27:16 +0000 (18:27 +0100)]
h264parse: Copy over DISCONT flag from input buffers
Sebastian Dröge [Fri, 28 Mar 2014 17:11:21 +0000 (18:11 +0100)]
tsdemux: Push next buffer after a discont with the DISCONT flag
Sebastian Dröge [Fri, 28 Mar 2014 16:32:56 +0000 (17:32 +0100)]
tsdemux: Drain remaining data on disconts
Sebastian Dröge [Fri, 28 Mar 2014 16:25:02 +0000 (17:25 +0100)]
mpegtsbase: Flush if we receive a discont buffer
Sebastian Dröge [Fri, 28 Mar 2014 16:24:46 +0000 (17:24 +0100)]
tsdemux: Don't dereference NULL if flushed before a program was selected
Sebastian Dröge [Thu, 27 Mar 2014 18:42:52 +0000 (19:42 +0100)]
tsdemux: Ensure that all pending data is pushed even if the stream was not activated before
gst_ts_demux_push_pending_data() will check if it now can activate the
stream and add the pad, we don't have to check that ourselves.
Fixes playback of very short MPEG TS files.
Sebastian Dröge [Thu, 27 Mar 2014 18:22:03 +0000 (19:22 +0100)]
mpegtsbase: Fix pull mode scanning for PCR on small files
If a file does not contain 5 PCRs until it is EOS, or does not
contain more than 655360 bytes the PCR scanning algorithm just
aborted.
Sebastian Dröge [Wed, 26 Mar 2014 21:25:07 +0000 (22:25 +0100)]
hlsdemux: Don't set the segment offset
It's causing wrong running times after seeks or bitrate
switches.
Sebastian Dröge [Wed, 26 Mar 2014 06:32:43 +0000 (07:32 +0100)]
hlsdemux: Set DISCONT flag on all buffers in reverse playback mode
Sebastian Dröge [Wed, 26 Mar 2014 06:25:13 +0000 (07:25 +0100)]
hlsdemux: Fix forwards and backwards searching in the files list
Sebastian Dröge [Wed, 26 Mar 2014 06:24:16 +0000 (07:24 +0100)]
hlsdemux: Unset DTS of all buffers
We won't get a valid DTS from the source.
Sebastian Dröge [Wed, 26 Mar 2014 06:22:54 +0000 (07:22 +0100)]
hlsdemux: Send flush start event before waiting for the tasks to finish
Otherwise we'll wait until buffers are completely processed downstream,
which might take quite some time.
Sebastian Dröge [Sun, 9 Mar 2014 18:31:31 +0000 (19:31 +0100)]
hlsdemux: Implement trick modes via I-frame variant lists
Sebastian Dröge [Sun, 9 Mar 2014 17:24:50 +0000 (18:24 +0100)]
hlsdemux: Implement parsing of #EXT-X-I-FRAME-STREAM-INF
These are I-frame-only variant lists that can be used
for trick mode playback.
Tim-Philipp Müller [Fri, 28 Mar 2014 20:22:43 +0000 (20:22 +0000)]
gl: egl: fix distcheck and out of source build
Rico Tzschichholz [Fri, 28 Mar 2014 19:47:36 +0000 (20:47 +0100)]
build: Fix make dist
Tim-Philipp Müller [Fri, 28 Mar 2014 19:46:33 +0000 (19:46 +0000)]
tests: fix LDADD order for insertbin test
Tim-Philipp Müller [Fri, 28 Mar 2014 19:41:25 +0000 (19:41 +0000)]
tests: don't link codecparser tests twice to the lib
Tim-Philipp Müller [Fri, 28 Mar 2014 19:30:16 +0000 (19:30 +0000)]
tests: fix build of gl unit tests in uninstalled setup
Put local .la files first, and add link to libgstvideo
for tests that use the libgstvideo API directly.
Matthew Waters [Thu, 27 Mar 2014 11:41:02 +0000 (22:41 +1100)]
gl/win32: allow not building the gl plugins
Fixes build on windows if <GL/wglext.h> headers are not present.
https://bugzilla.gnome.org/show_bug.cgi?id=727125
Julien Isorce [Fri, 28 Mar 2014 11:54:45 +0000 (11:54 +0000)]
eglglessink: unref last buffer on DRAIN
Similar to
093574053fae243ec5fcfd6adae8185053b1e896
in gstbasesink
https://bugzilla.gnome.org/show_bug.cgi?id=726107
Reynaldo H. Verdejo Pinochet [Thu, 27 Mar 2014 21:51:59 +0000 (18:51 -0300)]
eglglessink: fix wrong doc block for create_window
Wrong since
1ad253b9 when prop name was changed from
can_create_window to create_window.
Matthew Waters [Thu, 27 Mar 2014 11:02:22 +0000 (22:02 +1100)]
gl: don't fail configure if we do not understand the platform
Matthew Waters [Tue, 25 Mar 2014 04:40:14 +0000 (15:40 +1100)]
gl: fix out-of-source builds pt2
21b897de2f762bee504cbc007fd108148ee5ae24 did this for x11 and wayland.
Do it for the other backends as well.
Julien Isorce [Wed, 26 Mar 2014 23:13:53 +0000 (23:13 +0000)]
gl: only set CAPS_FEATURE_MEMORY_EGL_IMAGE on egl platform
Matthieu Bouron [Wed, 26 Mar 2014 17:14:40 +0000 (17:14 +0000)]
configure.ac: fix OSX host check
Makes the check works with (i386|x86_64)-apple-darwin* hosts.
https://bugzilla.gnome.org/show_bug.cgi?id=727104
Julien Isorce [Wed, 26 Mar 2014 19:17:03 +0000 (19:17 +0000)]
gl: let the user includes itself our egl headers if needed
Forgot to address this change.
https://bugzilla.gnome.org/show_bug.cgi?id=703343
Julien Isorce [Sat, 22 Mar 2014 22:01:49 +0000 (22:01 +0000)]
gl: add EGLImage support
* picked from old libgstegl:
- GstEGLImageMemory
- GstEGLImageAllocator
- last_buffer management from removed GstEGLImageBufferPool
* add-ons:
- GstEGLImageMemory now old a reference on GstGLContext
so that it can delete the EGLImage and its gltexture source
while having the associated gl context being current.
- add EGLImage support for GstVideoGLTextureUploadMeta which
mainly call EGLImageTargetTexture2D
- GstGLBufferPool now supports GstEGLImageAllocator
- glimagesink / glfilters / etc.. now propose GstEGLImageAllocator
to upstream
https://bugzilla.gnome.org/show_bug.cgi?id=703343
Julien Isorce [Mon, 24 Mar 2014 12:12:42 +0000 (12:12 +0000)]
gl: deploy egl headers in gst/gl/egl instead of gst/gl
https://bugzilla.gnome.org/show_bug.cgi?id=703343
Julien Isorce [Mon, 24 Mar 2014 12:10:00 +0000 (12:10 +0000)]
gl: remove commented and unsued code in x11 Makefile.am
https://bugzilla.gnome.org/show_bug.cgi?id=703343
Julien Isorce [Sun, 23 Mar 2014 21:55:34 +0000 (21:55 +0000)]
pkgconfig: add gstreamer-gl
https://bugzilla.gnome.org/show_bug.cgi?id=703343
Julien Isorce [Mon, 24 Mar 2014 12:04:08 +0000 (12:04 +0000)]
gl: fix crash if _build_extension_string is not called
On GLES2 then (gl->GetIntegerv && gl->GetStringi) is false
regression introduced by
cc6df204e2f58fffda5cbe90f3450aeba95889c4
https://bugzilla.gnome.org/show_bug.cgi?id=703343
Edward Hervey [Wed, 26 Mar 2014 09:14:36 +0000 (10:14 +0100)]
gl: Only enabled cocoa backend on osx (and not ios)
ios requires using EAGL
Jan Schmidt [Tue, 25 Mar 2014 22:05:15 +0000 (09:05 +1100)]
rtmpsink: Remove URL check for valid playpath.
The playpath is an optional component of the URL - don't require it.
Wim Taymans [Tue, 25 Mar 2014 16:39:18 +0000 (17:39 +0100)]
srtpenc: unlock object lock before posting error
Posting an error with the object lock will deadlock. Rework the code
a bit to make things more simple.