Re: Guile Binary Format 0.1

[Keisuke Nishida]

At 06 Feb 2001 02:42:57 +0200,
Michael Livshin wrote:
> 
> >       SCM_NEWCELL (obj);
> >       scm_restore_object (&SCM_CDR (obj), dstate);
> >       scm_restore_object (&SCM_CAR (obj), dstate);
> > 
> > but actually, it works...  Isn't this dangerous?
> 
> why would it be dangerous?

I thought SCM_SETCAR/CDR are doing something special, but looking
at their definitions again, I realized they aren't.  Okay, I'll
remove the redundant functions.

> > Yes, I'll create a table that maps stored typecodes to real tc
> > values.
> 
> by name?

Yes.

   Name   Stored tag      Real tc

  "foo"   127 + 0 * 256   (dynamically determined by name)
  "bar"   127 + 1 * 256   (dynamically determined by name)
  ...

Do we need some verification mechanism?  How?

Kei