Re: SCM_VALIDATE_...

[Mikael Djurfeldt <mdj at mdj dot nada dot kth dot se>]

Jim Blandy <jimb@savonarola.red-bean.com> writes:

> Clearly, SCM_REAL_P should be implemented by invoking a real? generic
> function, which the module adding these new representations can
> define...

Yes, this is one possibility, and there might be others.

The purpose of the value-oriented real test is simply to check if the
class of the value is a subclass of <real>.  If, e.g., we decide that
this test needs to be efficient, we simply dedicate a class flag for
reals and write

#define SCM_REALP(x) \
  (SCM_CLASS_FLAGS (scm_class_of (x)) & SCM_CLASSF_REAL != 0)

Probably, not all value oriented tests needs to be this generic,
though, since we can "seal" some classes against subclassing.  E.g.,
we should probably not allow people to create subclasses of <integer>.