Richard Maine wrote:
>
> William F Mitchell writes:
> > I am about to start a project that uses the proposed C Interoperability standard.
> > The NASoftware compiler has an implementation of the original ISO/IEC 1539-1:1997
> > proposal,
>
> I'm confused by that. ISO/IEC 1539-1:1997 would be the f95 standard.
> It doesn't have anything on C interop. Perhaps you mean the proposed
> interopability TR draft N1277. You might be confused by its first
> page, which does mention ISO/IEC 1539-1:1997. But N1277 isn't actually
> that ISO/IEC document - it is proposed changes to it.
Yes, my mistake. I read the wrong the number off the first page.
Henry Zongaro wrote:
>
> It might not be a good idea to use the proposed interoperability
> Technical Report (N1277) as the model for C Interoperability.
>
> For various reasons (mainly schedule related), the work item for that
> TR was dropped, and J3 was asked to complete the work as part of Fortran
> 200x instead. The syntax and some of the semantics for interoperability
> that has been adopted by J3 are sufficiently different that you would
> probably be facing significant re-work once Fortran 200x is adopted.
Right. I have Section 16 from J3/99-007R2 and would accept what it says over
what is in the other 3 reports. I expected it would be basically the same as
N1277 + N1298 + N1299, but haven't compared them carefully yet. Anyway, the
point is to take a careful look at the current proposal to see if I can do a
portable implementation of f90gl (without all those #if's) with it _before_
it is too late to suggest improvements. I was hoping I might be able to test
it out, too.
Bill
--
William F. Mitchell
Mathematical and Computational Sciences Division
National Institute of Standards and Technology
[log in to unmask] http://math.nist.gov/~mitchell
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|