On Wed, Jun 01, 2011 at 08:45:05AM -0700, Ronald S. Bultje wrote:
> Hi,
> 
> On Wed, Jun 1, 2011 at 8:27 AM, Diego Biurrun <[email protected]> wrote:
> > @@ -62,6 +72,11 @@ typedef struct FlashSVContext {
> >     uint8_t        *tmpblock;
> >     int             block_size;
> >     z_stream        zstream;
> > +    int             ver;
> > +    const uint32_t *pal;
> > +    int             is_keyframe;
> > +    uint8_t        *keyframedata;
> > +    BlockInfo      *blocks;
> >  } FlashSVContext;
> 
> Make a FlashSV2Context with embedded FlashSVContext, otherwise we have
> to clean it later just like svq3 in h264...
> 
> > +                if (zlibprime_curr) {
> > +                    int col = get_bits(&gb, 8);
> > +                    int row = get_bits(&gb, 8);
> > +                    av_log(avctx, AV_LOG_DEBUG, "%dx%d zlibprime_curr 
> > %dx%d\n", i, j, col, row);
> > +                    size -= 2;
> > +                    return -1; //NIY
> > +                }
> 
> av_log_missing_feature().

I've looked at one piece of code and it seems current Adobe stuff ignores it,
so av_log_ask_for_sample()

> Rest looks reasonably good to me. Kostya, can you practice your
> "teaching how to make things bitexact" skills on Diego for 1-2 weeks?

I can try.
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to