Those streams should never get any packets by definition.

Due to what looks like an off-by-one error 60 * 12 * 1001 / 12 * 1001
is not tested as standard frame rate in avformat_find_stream_info().

Also adjust some comment wording / spelling and reorder header #includes.

Always use the actually read size as the offset instead of making
possibly invalid assumptions.

Addresses: CVE-2012-6618

It's shorter and more consistent with the rest of the code.

AVStream.codec is always non-NULL

avcodec_decode_*() resets the frame itself.

Generate extradata with SPS/PPS based on container dimensions.

Authors of this commit are: Reimar and Thomas Mundt

Signed-off-by: Anton Khirnov <anton@khirnov.net>

This avoids setting avg_frame_rate to invalid (negative)
values.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>

Signed-off-by: Diego Biurrun <diego@biurrun.de>

This allows handling errors from av_realloc properly.

Signed-off-by: Martin Storsjö <martin@martin.st>

Signed-off-by: Diego Biurrun <diego@biurrun.de>

CC: libav-stable@libav.org

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org

This avoids breaking some slightly incorrect (dts > pts) timestamps
in sample HLS streams from Apple.

Signed-off-by: Martin Storsjö <martin@martin.st>

If either of the deltas is too large for the multiplications to
succeed, don't use this for setting the avg frame rate.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Cc: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>

It is only used by network protocols.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org

To avoid allocating ridiculous amounts of memory for corrupted files,
read the input in chunks limited to filesize or an arbitrary large
amount when that is not known (chosen to be 50M).

This will allow us to avoid copying the packets in many cases.

This breaks ABI.

Make sure the behavior does not change with the locale.

Signed-off-by: Martin Storsjö <martin@martin.st>

The handling of the environment variable no_proxy, present since
one of the initial commits (de6d9b6404), is inconsistent with
how many other applications and libraries interpret this
variable. Its bare presence does not indicate that the use of
proxies should be skipped, but it is some sort of pattern for
hosts that does not need using a proxy (e.g. for a local network).

As investigated by Rudolf Polzer, different libraries handle this
in different ways, some supporting IP address masks, some supporting
arbitrary globbing using *, some just checking that the pattern matches
the end of the hostname without regard for whether it actually is
the right domain or a domain that ends in the same string.

This simple logic should be pretty similar to the logic used by
lynx and curl.

Signed-off-by: Martin Storsjö <martin@martin.st>

The values compared here can be more than INT64_MAX apart.  Since the
difference is always positive, converting to uint64_t before subtracting
gives the correct result without overflows.

Signed-off-by: Mans Rullgard <mans@mansr.com>

Scaling the denominator instead of the numerator if it is too large
loses precision. Fixes an assert caused by a negative frame duration in
the fuzzed sample nasa-8s2.ts_s202310.

CC: libav-stable@libav.org

This is defined by RFC 3986 section 5.4.1 to be handled this way.

Signed-off-by: Martin Storsjö <martin@martin.st>

Signed-off-by: Martin Storsjö <martin@martin.st>