In article <·············@acm.org> Sin-Yaw Wang <······@acm.org> writes:
>It is possible to define a Scheme->C interface, that a Scheme function
>can call a C function. I am not sure it is possible to define a
>C->Scheme interface at the same time. The stumbling block is the
>continuation, a concept really does not exist in C.
>
>Imagine a C function calling a user-level Scheme funciton that calls
>call/cc. Later, another Scheme function invokes that continuation.
Are you saying in the case where the C function returned and then
later a scheme function invokes the continuation?
I think as long as you specify that the continuation should not depend
on any state within the C code, and as long as you don't care that the
continuation eventually returns back to the C entry point, even though
you jumped all over the place within the Scheme, it's ok. Seems like a
reasonable restriction.
--
---------------------------------------------------------------
| Chris Bitmead.....................................9690 5727 |
| ·············@Alcatel.com.au............................... |
---------------------------------------------------------------
The sum of the intelligence of the world is constant.
The population is, of course, growing.