When the socket is congested, we should simply skip sending.
When there is a soft-error, we should resend and simply ignore.

Instead, we were dropping the socket if the failure remained on the
second try, which was always the case in the congestion scenario.
This should fix the socket dropping problem described at:
http://forum.videolan.org/viewtopic.php?f=4&t=56493

Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

size_t are 64-bit long on Win64, adapt the wrapper for it
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

Since we now have a general work around, this hack is not needed
anymore. This reverts commit 6f1a7b41Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

This patch works around unsupported %z modifiers, by using a wrapper
around *printf functions that patches the format string on the fly. The
format string is duplicated, and "%z"'s are replaced by "%l"'s.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

This fixes a regression introduced in commit: 75adef44

It could happen if the input_thread_t die before/without sending any
meta event.

It avoid locking the whole playlist_t when fetching art, which can take a lot
of time...

Until a valid icon is designed/found, I used a DRM text label.

Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

stats: Add stream quality counters to Qt4 stastics by counting corrupted blocks and discontinuities found.

Note the quality counters are mere indications and not exact reporting of how much blocks/discontinuity really occured. It will only tell you how often it is noticed.