libcompat: added dummy.c from the devlopment branch to fix compilation on setups, which don't rely on libcompat

This will do for now to meet popular demand
(cherry picked from commit 04be17ebea21153734bcc8fa04948bd4cc0af67b)
Signed-off-by: Pierre Ynard <linkfanel@yahoo.fr>

Not needed anymore
(cherry picked from commit cedbf7f3a7879caebca543ee0deab02ccdc81743)
Signed-off-by: Pierre Ynard <linkfanel@yahoo.fr>

(cherry picked from commit 01356cfbd1b3ad567aa391245f3ba218a3b13048)
Signed-off-by: Pierre Ynard <linkfanel@yahoo.fr>

Why did they got 2 CVE for the same issue?

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>

Same problem than in 1d4c5a496b8b6d21c20ccb7598216195069f346e

poll() revents are flags, and the lua module tests for equality...
Obviously this has been working merely by accident given that a remote
host resetting a connection generates only POLLIN, and neither POLLHUP
nor POLLERR.

lua doesn't support bitwise operators so this is the simplest way.
(cherry picked from commit ce00244d121cfba2bbbc5065fcd3a6eb84bbba6c)
Signed-off-by: Pierre Ynard <linkfanel@yahoo.fr>

Contrary to V4L2, dshow does not extract options from the MRL. So there
i effectively no way to set anything via playlist. This allows the
options that more-or-less correspond to MRL options in V4L2.

Parsing the MRL might make more sense than marking options as safe.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
(cherry picked from commit 77a3fbe87694a1acd762db61e22eda0424786f63)

Conflicts:

	modules/access/dshow/dshow.cpp

(cherry picked from commit 4390c31a0ade28f9938e44ef1d5c20885659bffc)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Too many complaints, notably on Reddit

requires latest contrib package

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

This seems to be the latest fashion in poland, for {y:i} (italic) and
{y:b} (bold), but I don't know how to correctly support it:
 - should we remove the tags in the demuxer like the MPL2 one (bad
   solution for MPL2 too)?
 - should we replace it by <i> in codecs?
(cherry picked from commit 195b652da5facd60397504f706a44d97bd70eb90)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

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

With just "skins" and "qt", only those specific interfaces would be
probed. Thus VLC could not be started from X-less command line.
(cherry picked from commit 727b24988dd766eac2f8a63f93cf841aba826a92)

(cherry picked from commit f271a2effd810941493500bf22b8dea92a0386a9)
Signed-off-by: Pierre Ynard <linkfanel@yahoo.fr>

Signed-off-by: Ilkka Ollakka <ileoo@videolan.org>
(cherry picked from commit 7aaf63551d1e87d3219acd5a3cec1fa25250bfe3)

This fixes problems with lost characters and failure to close
connections in some cases
(cherry picked from commit 1b76f4d5b6e247ebc1f3f1786c99eef1f0510c73)
Signed-off-by: Pierre Ynard <linkfanel@yahoo.fr>

Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>