[FFmpeg-devel] [PATCH] Add asink_anullsink - Null audio sink

[Stefano Sabatini]

On date Wednesday 2010-09-15 19:49:07 +0200, Michael Niedermayer encoded:
> On Wed, Sep 15, 2010 at 04:31:07PM +0200, Stefano Sabatini wrote:
> > On date Wednesday 2010-09-15 14:07:40 +0200, Michael Niedermayer encoded:
> > > On Wed, Sep 15, 2010 at 12:53:37AM +0200, Stefano Sabatini wrote:
> > > > On date Thursday 2010-08-19 17:06:53 +0200, Stefano Sabatini encoded:
> > > > > On date Thursday 2010-08-19 06:11:12 -0700, S.N. Hemanth Meenakshisundaram encoded:
> > > > > > Renamed, updated doc.
> > > > > > 
> > > > > > ---
> > > > > >  doc/filters.texi              |   13 +++++++++++++
> > > > > >  libavfilter/Makefile          |    2 ++
> > > > > >  libavfilter/allfilters.c      |    2 ++
> > > > > >  libavfilter/asink_anullsink.c |   40 ++++++++++++++++++++++++++++++++++++++++
> > > > > >  4 files changed, 57 insertions(+), 0 deletions(-)
> > > > > >  create mode 100644 libavfilter/asink_anullsink.c
> > > > > 
> > > > > Looks fine.
> > > > 
> > > > I'll apply in three days if I see no objections.
> > > 
> > > Does this filter work?
> > > Does it consume audio or does it never read a single sample? 
> > 
> > Consume audio, but input samples are completely ignored (but that's
> > not the point of the filter, which is to help testing/debuggging
> > filterchain setup), and is the equivalent of the video nullsink.
> 
> iam not so sure if this filter actually works
> the thing is that nothing seems to ever call request_frame and so nothing
> will ever be pushed towards this sink. its like ffmpeg will wait at frame
> 1 for ever.

That's the point of a sink. It will never be requested of frames as
it is a null sink.

Of course you can have other sinks which currently sucks data like
here:
asrc, anull, anull, asplit [A1][A2]; [A1] anullsink; [A2] asink

Patch updated.

Regards.
-- 
FFmpeg = Frenzy & Furious Muttering Porno Exuberant Geek