[MPlayer-dev-eng] [PATCH] vd_null to output black frames

[Reimar Döffinger]

On Wed, Oct 31, 2012 at 03:41:53PM +0800, Xidorn Quan wrote:
> On Tue, Oct 16, 2012 at 2:26 PM, Reimar Döffinger
> <Reimar.Doeffinger at gmx.de>wrote:
> > On 15 Oct 2012, at 21:48, Alexander Strasser <eclipse7 at gmx.net> wrote:
> >
> > > Xidorn Quan wrote:
> > >> On Mon, Oct 15, 2012 at 6:23 AM, Alexander Strasser <eclipse7 at gmx.net
> > >wrote:
> > >>> Reimar Döffinger wrote:
> > >>>> Sorry, I'm travelling, and won't be able to apply in the
> > >>>> next few days, but if someone else wants feel free.
> > >>>> I think it's ok, even though I would have slightly preferred to
> > >>>> keep vd_null as a simple trivial example and make a completely
> > >>>> separate vd_black or similar filter.
> > >>>
> > >>>  I separated Xidorn's code into a new decoder vd black (patch
> > >>> attached). Additionally made some small mostly cosmetic changes.
> > >>> I thought putting it in a new file is a good occusion to do so.
> > >>>
> > >>>  Xidorn, could you please test if this fulfills your needs?
> > >>>
> > >>
> > >> Yes, it fulfills my needs accurately.
> > >>
> > >>
> > >>>  Also I think I didn't test properly, so I would be glad to
> > >>> hear your use cases for the new functionality.
> > >>>
> > >>
> > >> I just use it for testing filters like vf_ass, doing performance
> > >> benchmark for vfs and vos. There is no other use cases for me.
> > >>
> > >> Because it output simple frames as quickly as possible, I can easily
> > >> speed up playback a lot without waiting for slow decoding.
> > >
> > >  Seems to work as expected.
> > >
> > >>>  If everything works as you like I hope you do not mind if
> > >>> I put your full name into the header comment as copyright
> > >>> statement and in the author field of mp_codec_info_t.
> > >>>
> > >>
> > >> I certainly would not mind that.
> > >
> > >  Done and committed.
> >
> > Thanks for taking care of it.
> > Btw. it might make sense to point to the special-case null and black
> > strcmp code from the codecs.conf entries, since it is related to why a
> > codec entry without any fourccs even makes sense.
> > Alternatively, having a flag in the codec structure to indicate such
> > "dummy" codecs might make it more obvious, too.
> >
> 
> The patch attached adds a flag for dummy codecs. Is it ok?

I had actually thought making it a flag in the codec struct in the vd_*.c file,
but I think I actually like your idea better :-)
So, looks fine to me.