On 5/19/10, Gerd Hoffmann <kra...@redhat.com> wrote:
> >
> > >
> > > > Then there would be no need of parsing.
> > > >
> > >
> > > You would need _two_ bitmaps (e.g. mask and cursor, so that mask=1
> gives
> > > transparent, mask=0 cursor=0 gives black and mask=0 cursor=1 gives
> white).
> > >
> >
> > Yes, but it's still packed more efficiently.
> >
>
> Well. You can't have both. We can have a efficiently packed format (i.e.
> two bitmaps). Or we can do it in a way which doesn't need parsing, but that
> wouldn't be the most compact format ...
You're right, so packing or introducing a small conversion function is
not critical. I'd still prefer a standard format if possible.
> > There's yet another way:
> > #define _ 0,
> > #define X 0xff000000,
> > #define o 0xffffffff,
> > {
> > _ _ _ X o X _ _ _
> > }
> > #undef _
> > #undef X
> > #undef o
> >
>
> Neat idea ;)
>
> cheers,
> Gerd
>
>
