This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

This is shorter and consistent across filters.

Currently it would incorrectly trigger on a string that contains a '='
but does not contain a ':', e.g. flags=<flags>.

There is now no need to explicitly pass 0:0 as width/height to scale,
those are the defaults.

The C code treats the data as arrays of uint16_t so strides must not
be in bytes but in pixels.

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

Always use the special filter for the first and last 3 columns (only).

Changes made in 64ed3976 slowed the filter to just under 3/4 of what it
was.  This commit restores the speed while maintaining identical output.

For reference, on my Athlon64:
1733222 decicycles in old
2358563 decicycles in new
1727558 decicycles in this

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

delta is in samples, not bytes. Also the sample format is not guaranteed
to be planar.

CC:libav-stable@libav.org

The functions which actually drive the filter graph by pushing
frames through it need to ensure an aligned stack for SIMD functions.

This fixes a crash in YADIF filter when using a mingw build in a MSVC
application.

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

Options are freed from the generic code now, there is no need to call
av_opt_free() from the filter.

It now allows an arbitrary number of inputs, not just two.

The current code can fail to return the last frame if it contains
exactly the requested number of samples.

Fixes the join filter test, which previously did not include the last
408 samples in most cases.

CC:libav-stable@libav.org

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