This fixes warnings with PulseAudio and with MingW.

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

SI NON CONFECTVS, NON REFICIAT.

When building current HEAD of master for x86_64-w64-mingw32 against current
contribs, compilation fails at libmod_plugin due to undefined symbols of
the form __imp_ModPlug_*. This patch modifies libmodplug in contribs by
adding -DMODPLUG_STATIC to CFlags in the pkg-config file. A bug is also
fixed in the main build system to make sure the CFLAGS are picked up from
this pkg-config file.

This problem occurs because of the following code in
contrib/x86_64-w64-mingw32/include/libmodplug/modplug.h. This code is
present in libmodplug obtained from running "make fetch & make" in the
contribs directory, but not present in the prebuilt contribs obtained from
running "make prebuilt".

 #if defined(_WIN32) || defined(__CYGWIN__)
 # if defined(MODPLUG_BUILD) && defined(DLL_EXPORT)
 #   define MODPLUG_EXPORT __declspec(dllexport)
 # elif defined(MODPLUG_BUILD) || defined(MODPLUG_STATIC)
 #   define MODPLUG_EXPORT
 # else
 #   define MODPLUG_EXPORT __declspec(dllimport)
 # endif

When building for Windows, if MODPLUG_STATIC is not defined, MODPLUG_EXPORT
is defined as "__declspec(dllimport)". On Windows, dynamic symbols are
prefixed with "__imp_", so this causes GCC to adjust the symbol references
to the libmodplug functions accordingly.

This patch therefore adds a patch to contribs for libmodplug's pkg-config
file. It adds -DMODPLUG_STATIC to CFlags, so that when libmod_plugin is
building it will have the correct symbol names.

However, there is also a bug in the build system that needs to be fixed for
this to work. The configure script sets a CXXFLAGS_mod containing the
output of pkg-config for libmodplug, but not CFLAGS_mod. Additionally, the
Makefile.ac for libmod_plugin sets CXXFLAGS when in fact the relevant files
(mod.c) are plain C. Autotools therefore ignores these CXXFLAGS when
generating a makefile.

The solution is to add a macro to configure.ac to set CFLAGS_mod. This then
needs to be used in modules/demux/Makefile.am to populate
libmod_plugin_la_CFLAGS instead of libmod_plugin_la_CXXFLAGS (which gets
ignored).
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

The code no longer compiles with older versions anyway.

This fixes the situation where both Qt4 and Qt5 are installed but the
global moc/rcc/uic commands are provided by Qt4 and Qt5's versions are
not globally installed. VLC previously picked up Qt4's moc/.. version
but built against Qt5.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

This plugin is capable of decoding MPEG-2 and H264 video through the Multi
Media Abstraction Layer (MMAL) by Broadcom. It is supported by recent Broadcom
SoCs and found on the widely spread Raspberry Pi.
Signed-off-by: Dennis Hamester <dennis.hamester@gmail.com>
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

This plugin uses the Multimedia Abstraction Layer (mmal) by Broadcom. It is
supported on recent Broadcom SoCs as found on the well known Raspberry Pi.
The plugin features support for native blending of overlays through DispManX.
Furthermore it supports automatic adaption of refresh rate to the video being
played through the tvservice API.
Signed-off-by: Dennis Hamester <dennis.hamester@gmail.com>
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

The global context setup are proper av_ namespaced functions,
the frames are properly refcounted and the output is fixed to
UYVY.
Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>

GStreamer is a cross-platform multimedia processing
framework for audio/video postprocess/capture/
encode/decode/render.

This module is intended for Audio/Video decoding using
the available GStreamer plugins present in the system.
Currently this module supports only video decoding
and can be extended to audio decoding as well.
ZeroCopy is currently not supported and must be added.

Codecs currently supported are h264, mpeg4, vp8, mpeg2,
flashvideo, wmv1/2/3, vc1.
Signed-off-by: Vikram Fugro <vikram.fugro@gmail.com>
Signed-off-by: Jean-Baptiste <jb@videolan.org>

Close #6649

- Previous algorithm was unable to guess the first packet
pts because of ogg design (some codecs/opus/vorbis have total
sample at page granule and only granule on last packet. You
then need to compute first packet pts backward using packets
duration).
So we were stuck either assuming a start at pts/pcr zero, or
be unable to handle ogg streaming (non zero start) correctly,
or had to always drop the 0..N-1 packets for first page.
We need libvorbis to correcly compute those durations (code
is far more complex than opus's one).
- Refactorized to discard the tons of cases appended
with each new codec.
- lots of other fixes in many places

There is no support for progress dialogs so far

This allows automagic enabling and disabling of the needed and useful aout and vout modules

FFmpeg doc/APIchanges is not trustworthy.

Ref #6649

cross-pkg-config tools have a correct default search path. The contrib
path should complement it, not replace it. If the caller intends to
(ab)use the native pkg-config for cross-compilation, the caller is
responsible for selecting an adequate default path, not VLC configure,
just as with every other build tools.

When converting 1920x1080p mp4 using x264, x264 complains that it failed
to allocate memory. This flags makes it happy.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Based on the work by Martin Storsjö.

The decoder stores opaque buffers in the p_sys member of the picture
and the vout uses a callback from the decoder to render these
buffers. When the decoder flushes or closes, all the currently in
flight pictures (filled by the decoder but not displayed yet) need to
be invalidated. A mutex is required in order to prevent the vout from using
destroyed MediaCodec buffers.

In order to avoid a deadlock when exiting the decoder, a maximum number of polling
attempts was added in order to avoid blocking the decoder indefinitely if the vout
is not releasing output buffers anymore.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Comment says:
"NB, this is a temporary encoder only module until schroedinger
offers superior encoding quality than dirac-research"

http://diracvideo.org/2012/01/schroedinger-1-0-11/ says:
"Schrödinger outperforms dirac-research in most encoding situations,
both in terms of encoding speed and visual quality."

It might only contain the vp9 decoder