IDirectDrawSurface2_UpdateOverlay may fails, in which case vlc should simply
retry later.
(cherry picked from commit d0606472)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

There is a little difference in Direct3DLockSurface as it now also
sets the height field, but it should have no functionnal changes.
(cherry picked from commit a0530f4a)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

(cherry picked from commit 8e6a4513)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

(cherry picked from commit dc4b98a9)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

(cherry picked from commit 532e5d01)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

get the SD video (in flv), the HD and HQ videos (in mp4) causes seeking
back and forth and the video is unplayable
(cherry picked from commit e778af8bb60ba6a67c47e8b30c175874726414ee)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

Do not try to delete a callback and/or a mutex that do not exist.
The goto error paths still look a bit suspicious, someone should
double check them.
(cherry picked from commit 1f253df8)

Signed-off-by: Laurent Aimar <fenrir@videolan.org>
(cherry picked from commit a4f30ff9)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

(cherry picked from commit 3683cc99da0a39bca72d14a17682ffeacee7652b)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

(cherry picked from commit 7b27f950b0fe4c76acbea58ea183a486bdc7da2e)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

We now test that the selected fourcc actually works by calling vaGetImage.
Initial patch from Gwenole Beauchesne.
(cherry picked from commit 84f1839b2318ac0963bc37d43b078d5992e9e8db)
Signed-off-by: Rémi Duraffort <ivoire@videolan.org>

(cherry picked from commit 47369283f9bed68db77e0f1ba1f11840ddf1bfb3)

This was fixed after 1.1.0-RC.
(cherry picked from commit 617801b6877e3aac9651eaa15e9ee0d2aea564d8)

(cherry picked from commit 216818a84e81e8332f74a93513c249d63beee444)
Signed-off-by: Ilkka Ollakka <ileoo@videolan.org>

(cherry picked from commit 1f5e1e827fffcb075d196faffced40eee2df8c08)
Signed-off-by: Ilkka Ollakka <ileoo@videolan.org>

(cherry picked from commit 8970ff6d495ee362098545b8f4a075ebda5d8099)
Signed-off-by: Konstantin Pavlov <thresh@altlinux.org>

If 0.8.8 is detected and --enable-mod is passed, it will error,
if --disable-mod is passed, nothing will happen
else it will autodetect. This is mostly like all other PKG_ENABLE_MODULES_VLC
detections
(cherry picked from commit e043658db0f93148d2a3e499cd90e6fb76ee3fd2)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

fullscreen is now managed (activation and deactivation) in skins2
 exactly as it is done in qt4.
    - activation when a VOUT_WINDOW_SETFULLSCREEN is received
    - deactivation when (and only when) a vout_window release is received.
(cherry picked from commit 2e298fb87562b7a6470c24091a336b251be64bc4)

This was annoying because if you had a modplug.pc and didn't --enable-mod, it was auto-detected and then failed.
(cherry picked from commit a9d916728b2717206a0a7fe0702b0bc6dad16747)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Should fix #3767
(cherry picked from commit 202c4d9c4fab9bf3391c887194d4f05999634d61)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Close #3887
(cherry picked from commit 71542b4b)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

The current code seems to require _all_ metadata fields to be populated
which seems a bit unreasonable.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
(cherry picked from commit 7d9488cf0055fb8baf3b1d619d30824e7f8c4a10)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Signed-off-by: Christophe Mutricy <xtophe@videolan.org>

This brings some useless overhead for native UTF-8 systems.
Linux packagers may want to define ASSUME_UTF8 to avoid this problem.

Also, this assumes that iconv_open() recognizes "" as the current
character set (POSIX says nothing about valid iconv_open() parameters).
(cherry picked from commit 9164192f0231bed4913e5cd3bb4174343163439b)

Signed-off-by: Christophe Mutricy <xtophe@videolan.org>

(cherry picked from commit 53a9e90cea0976140f2f7f8e07c4cdc20bb0fc17)

(cherry picked from commit 749860b6bf52df7418f25a820fc14f54a08ffb32)

Move googleimage search first, and enable it again but don't use eg filename in search query.
Move musicbrainz after last.fm, and don't try more than one way to find art, also do query once in 2 sec

If you get bad hits from googleimage album-art, please report them to me.

Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
(cherry picked from commit e21730c43e7c7fdb6d5d02c4b1e97888732c2ec2)

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

(cherry picked from commit 2784bc2a8e8c022398879fbe5299d9683edfafe6)
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

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

(cherry picked from commit fb92e7a0440af1dcb237479702e559e0278a8bca)

At least with X11, the mouse may move without changing coordinates.
We need to re-arm the cursor hide timeout since the cursor was shown.
(cherry picked from commit ac6b127e2eabcda70a86d94c6ebfc7df9de3dbe6)

Signed-off-by: Christophe Mutricy <xtophe@videolan.org>

Signed-off-by: Christophe Mutricy <xtophe@videolan.org>

Signed-off-by: Christophe Mutricy <xtophe@videolan.org>