[vlc-devel] [PATCH v2] input/input: fix `input-slave` option for spus

[Alaric Senat]

If accepted, theses changes can be backported to 3.0 as well.

Regards

On Mon, Mar 29, 2021 at 17:51, Alaric Senat <dev.asenat at posteo.net> 
wrote:
> Since c34d719f, all files passed by the option `input-slave` were set 
> as
> audio tracks no matter what.
> This commit attempts to deduce the `input-slave` type with its file
> extension and falls back to spu in case of failure.
> 
> Fixes #25549
> ---
>  src/input/input.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/src/input/input.c b/src/input/input.c
> index ea64473a0b..aa8280e7b9 100644
> --- a/src/input/input.c
> +++ b/src/input/input.c
> @@ -984,8 +984,17 @@ static void GetVarSlaves( input_thread_t 
> *p_input,
>          if( uri == NULL )
>              continue;
> 
> +        enum slave_type i_type;
> +        if ( !input_item_slave_GetType(uri, &i_type) )
> +        {
> +            msg_Warn( p_input,
> +                     "Can't deduce slave type of `%s\": falling back 
> to spu.",
> +                     uri );
> +            i_type = SLAVE_TYPE_SPU;
> +        }
>          input_item_slave_t *p_slave =
> -            input_item_slave_New( uri, SLAVE_TYPE_AUDIO, 
> SLAVE_PRIORITY_USER );
> +            input_item_slave_New( uri, i_type, SLAVE_PRIORITY_USER );
> +
>          free( uri );
> 
>          if( unlikely( p_slave == NULL ) )
> --
> 2.29.2
>