[MPlayer-dev-eng] [PATCH] [TEST AFILTER] Encode ac3 at runtime using lavc

[Ulion]

2007/12/11, Ulion <ulion2002 at gmail.com>:
> 2007/12/11, Reimar Döffinger <Reimar.Doeffinger at stud.uni-karlsruhe.de>:
> > Hello,
> > On Tue, Dec 11, 2007 at 01:18:28AM +0800, Ulion wrote:
> > [...]
> > > +.IPs <srate>
> > > +Selects the output sample rate to be used, supports 48000, 44100, or 32000.
> >
> > Can't the sample be achieved by adding e.g. lavcresample=... in front of
> > that? If yes I am against having yet another way to do the same thing.
>
> Oh, then removed with man page update.
>
> >
> > > +    while (left > 0/* || c->eof && s->pending_len > 0*/) {
> > > +        if (left + s->pending_len < s->expect_len) {
> > > +//            if (left > 0) {
> > > +                memcpy(s->pending_data + s->pending_len, src, left);
> > > +                src += left;
> > > +                s->pending_len += left;
> > > +                left = 0;
> > > +//            }
> > > +//            if (!c->eof)
> > > +                break;
> > > +//            memset(s->pending_data + s->pending_len, 0, s->expect_len - s->pending_len);
> > > +//            s->pending_len = s->expect_len;
> > > +        }
> >
> > I do not like loads of having commented-out code in new code.
> > The same applies to all those af_msg, either they are useful then
> > they should be enable (possibly at DBG or DBG2 level) or they are not
> > and should go completely.
>
> OK, fixed.
>
> >
> > > +        if (s->add_iec61937_header) {
> > > +            int16_t *out = (int16_t *)buf;
> > > +            out[0] = 0xF872;   // iec 61937 syncword 1
> > > +            out[1] = 0x4E1F;   // iec 61937 syncword 2
> > > +            out[2] = 0x0001;   // data-type ac3
> > > +            out[2] |= (buf[8+5] & 0x7) << 8; // bsmod
> > > +            out[3] = len << 3; // number of bits in payload
> > > +
> > > +#ifndef WORDS_BIGENDIAN
> > > +            swab(buf + 8, buf + 8, len);
> >
> > As said before this is not a valid way to use swab, it does not have to
> > work for from == to.
> > And this code wouldn't be needed at all if we can just get the ac3
> > endianness stuff done first.
>
> Fixed. As the ac3 endianness stuff, it indeed should be spdif or
> iec958 endianness, and wiith a inner format flag, that's another
> unfinished discussion.
>
> Updated patch is here. If there's no objection, I will commit it in 2 days.

Commited.

-- 
Ulion