GitLab

The configuration system cannot store both a static list and a dynamic
one (and AFAIK, it never could). It only makes to use either a static
list or a callback.

The frame rate is retrieved from the capture parameters nowadays. This
option has no effects at the moment.

It might be nice to reintroduce it so that it can select one rate when
a device supports more than one. But that was never implemented so far.

Anyway V4L1 is dead, so there is no ambiguity.

Especially fix memory leaks on error.

The streaming parameters depend on the format, so the format must be
set explicitly, then the streaming parameters queried every time.

This essentially reverts 1bb938e4.

This brings --v4l2-width and --v4l2-height back into use.
A value range is now added as well.

Cropping depends on the image resolution, which is selected while
setting the format.

This disables driver-specific modes that VLC would not know about.
This also disables high quality (i.e. still camera) mode since we want
video, as well as other any other future abnormal modes.

 * Enumerate frame sizes once rather than twice.
 * Do not enumerate frame rates if not supported.
 * Get actual frame rate from the device driverr.
 * Get exact fractional frame rate rather than round to single precision
   floating point.

Notes:
 * --v4l2-fps becomes totally redumdant. It should probably be redefined
   to select a maximum capture frame rate.
 * --v4l2-width and --v4l2-height are ignored. This is a regression.
   Maybe they should be redefined as maxima as well as --v4l2-fps.

There is not that much common code left for the access case, except for
debug messages. The resulting code is a lot easier to read IMHO.
Also I have no way to test the access plugin, so lets split it where it
will not get more broken.

There is really no point in enumerating all inputs here. Eventually
the service discovery plugin should take of that properly.

Do not bother if the input does not support/need it.

This should be set for all NTSC standard variants, not only for the
whole set of NTSC standards. More importantly, this should used the
detected standard (VIDIOC_QUERYSTD) rather than the configured one
(VIDIOC_S_STD) if at all possible.

The tuner index is determined by the input. The separate tuner index
setting was thus wrong and is removed.

The audio mode now defaults to primary language rather than being
unspecified. If this is not supported, it falls back to stereo, then
mono.

The audio capture capability flag is irrelevant. Instead the selected
video input specifies its allowed set of audio inputs.

This is not to say that devices of some other types should not be
supported. But none of them currently are really supported.
(Also I think a different MRL scheme would be preferable for some.)

This removes the build-time and install-time dependency.

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