This is as already done a long time ago for PulseAudio. VLC might
deliver audio a long time ahead, but there is no warranty thereof.
So a small latency must be assumed for the worst case.
(cherry picked from commit 894b9dfb3b1d3d5852cb0d52470d73c06be27b37)

This should fix support for or work around bugs in, some drivers:
 * explicitly disable soft-resampler,
 * define access mode before sample format,
 * set buffer before period,
 * use period time rather than period count.
(cherry picked from commit d80d6da77e3074d3ecc67fb70a8305d08f137f43)

macosx: lock the scroll direction for 0.8 secs, so the user can scroll through the movie or adjust the volume without affecting the other function (close #6893)

This is especially needed for the multitouch trackpads
(cherry picked from commit d34619d34cc2e93c02f66dd04b41f74a6ec4eeed)

no functional change
(cherry picked from commit d9fb60bec03044bb7bc20e9a18d52f43f940b75b)

close #6870
(cherry picked from commit 01533a3cb5eece7f3e6638f51415df4909ccb6c7)
Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>

(cherry picked from commit d20acdeca0078e548a68491cd27e7413d5e5a4b1)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 003ba60d1911a7f6a6a97aaf24324660f50b3959)

(cherry picked from commit 3b0b4fb354db179353d4d319484f1711f979de58)

(cherry picked from commit 506ef92b9445bc3cf43dc5630c45958312e3cd84)
Signed-off-by: David Fuhrmann <david.fuhrmann@googlemail.com>

- volume and mute hotkeys are handled by core now, providing a proper osd screen
- some fixes in detection if key combo is valid hotkey
- fix for mem leak
(cherry picked from commit 1536ed21be0f545840bf8b15e89e4f81fd1908a8)
Signed-off-by: David Fuhrmann <david.fuhrmann@googlemail.com>

Currently, two objects of the class VLCMainWindow are generated. The first is alloated
from intf.m (triggered by the first log message that arrives) and is initialized by
both init and initWithContentRect. The second object is created from inside the nib file,
and is instantiated ONLY with initWithContentRect. Therefore, up to now _o_sharedInstance
had the wrong object and all calls to sharedInstance would cause problems.
But somehow, o_mainwindow in intf.m is updated to the correct second object, although it holds
the first object at start.

Furthermore, o_fspanel is also instantiated through the nib file.
(cherry picked from commit 0d847ccb8d2e58147fe351a10fc89e777f469fb8)
Signed-off-by: David Fuhrmann <david.fuhrmann@googlemail.com>

Fixes MSVR 151, MSVR 193 and MSVR 195
Finder: Jeremy Brown at Microsoft and Microsoft Vulnerability Research (MSVR)

(cherry picked from commit 59c0cf6f125a5d6ddbb839adcfa31404c0043c89)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 18faa622b2de2b94696d3de273704cb4a588bb3c)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

Signed-off-by: Christophe Mutricy <xtophe@chewa.net>

This is a temporary workaround until someone add support for f4v.

Work-around #4915
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
(cherry picked from commit f47133cbac88a1ef04e526df27eaa4e86a6f2d5a)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Some muxers are unable to read the spec...
(cherry picked from commit 164c7d31939a16674cd5c678e453085af2cbef4c)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Ref #6652
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
(cherry picked from commit 4763a6a249a3280abe346b67238e8f9861a5999f)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

The updateStatus() function should not interrupt the user and reset his position while he is dragging the volume slider.

Close #6652
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
(cherry picked from commit e91edd90eaf9cd4339b45fd90f8af4fe8a3b31b3)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

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

(cherry picked from commit 8de46adc1189fe3a3ad7c127896ec026d9da3d62)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

The bit per samples must be exact for the avcodec decoder.
It closes #6799.
(cherry picked from commit 59ae308b755d365655dca9ccccf9b738b1415f5f)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit da39be7b8045b28a744711f7a3482b7639325707)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
(cherry picked from commit 533c3fda8892754017a26d2dfe3ef9d3cbb12dbb)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit a5b0a06c3a8a6c6276e172174405b40e6678b2a1)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Strings from the lua stack stop being valid after you pop them
(cherry picked from commit 8f048e9cabebeec8b514e981993d16acfbbd7c28)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

This is usually not visible, since you need avcodec+avformat, instead of
ts/faad. Close #6847
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
(cherry picked from commit e0a4ac832b5214561944cd8212d7a94ad5d0ce87)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>