On Thu, Apr 28, 2016 at 03:24:10PM +0200, Nicolas George wrote:
> Le decadi 10 floréal, an CCXXIV, Hendrik Leppkes a écrit :
> > It probably does generally no harm to include it since its tiny,
>
> That was my reasoning.
>
> > however it would still show up in configure and be disable-able in
> > configure (since the list is generated from the extern const bitstream
> > filter entries in bitstream_filters.c) - it would just not actually
> > disable it.
> > I think this is rather confusing.
>
> Indeed. Still, some users may think they do not need bitstream filters and
> configure with --disable-bsfs, and end up with non-working applications
> because they require this one to handle the no-filtering case. That is
> confusing too.
>
> I changed the declaration to hide it from configure.
>
> > Also, something I forgot, there is a plan to generate the
> > bitstream_filters array from configure instead of hardcoding it, which
> > would make special cases not governed by the usual rules a bit more
> > annoying.
>
> I am rather more optimistic: the array is generated by configure, but the
> declaration of the array,
> "static const AVBitStreamFilter *bitstream_filters[] = {",
> needs to be hardcoded somewhere. Adding the hardcoded filters there should
> be easy enough.
>
> IMHO, the cleanest design would be to have configure generate just the list,
> and have:
>
> static const AVBitStreamFilter *bitstream_filters[] = {
> #include "bistream_filters_list.h"
> };i think a system that allows user apps to add their own bitstream filters would be good [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB If you think the mosad wants you dead since a long time then you are either wrong or dead since a long time.
signature.asc
Description: Digital signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
