We ensure that pbox->i_size is never smaller than 8 to avoid an
integer underflow in the third argument of the subsequent call to
memcpy. We also make sure no truncation occurs when passing values
derived from the 64 bit integer p_box->i_size to arguments of malloc
and memcpy that may be 32 bit integers on 32 bit platforms.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

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

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

This reverts commit 67066bf6 and
1a2780cb as there are still some
upstream issues to resolve.

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

Check the return value of malloc to avoid a null-pointer dereference.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

The variable len is a raw 32 bit value read using GetDWBE. If this
value is larger than UINT32_MAX - sizeof(eos), this will cause an
integer overflow in the subsequent call to malloc, and finally a
buffer overflow when calling memcpy. We fix this by checking len
accordingly.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

If we provide initial segment-number, most likely we want to continue
where we have left of, so leave old playlist-file there and just
overwrite it when we updated it.

default initial-segment-number is 1 so don't test against 0

If provided initial-segment-number option, signal discontinuity so
player knows to reset timestamps etc for new segments.

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

This mandates 8.1 though

On android 5.0, PrintMsg blocks on fprintf when stderr is locked by flockfile.

This commit adds a default log callback for android in order to fix this issue,
and in order to have early logs (before libvlc_log_set is called).
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

Use it only if specified by --aout arg.
Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>

This failed on Android, for example

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

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

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

Signed-off-by: Ilkka Ollakka <ileoo@videolan.org>

When input is H.264 they don't have pts/dts sometimes before decoder, but
dropping non-dated packets screw up decoder.