We need a Prolog which can call C++ functions. At the present moment,
we're using BimProlog, but there's no way to do it.

We will thank any help. Please send a direct mail to the address in
the signature (don't use a reply), since I'm neither a subscriber of
these newsgroups, nor this is my usual machine.

Thanks a lot

Jaime

==========================================================================
Jaime Simao SICHMAN IMAG/LIFIA 46,av.Felix Viallet 38031 GRENOBLE Cedex
jaime@lifia.imag.fr  bureau : 131  tel: +33 76574807  fax: +33 76574602
==========================================================================
--
=========================================================================
Jaime Simao SICHMAN IMAG/LIFIA 46,av.Felix Viallet 38031 GRENOBLE Cedex
jaime@lifia.imag.fr  bureau : 131  tel: +33 76574807  fax: +33 76574602
=========================================================================

In article <1992May18.132530.20759@imag.fr>, jaime@imag.fr (Jaime Sichman) writes:
> We need a Prolog which can call C++ functions. At the present moment,
> we're using BimProlog, but there's no way to do it.
>

I built a prototype Prolog<->C++ interface for Quintus, based on the existing
Prolog<->C interface.  It was not difficult, but it was a little awkward, mainly
because of the way that C++ tends to be implemented and the limitations in
the Unix linker.  (C++ uses some "kludges" to handle name overloading --- a
Prolog<->C++ interface has to use the same kludges; unfortunately, different
C++ implementations use different kludges).

So, the bottom line is: you can connect Quints Prolog and C++ in both directions
(sorry for the advertising plug, but I can't speak for other Prologs).  It's not
difficult but it will take you some time and effort.

[In case you're wondering, my Prolog<->C++ code was for connecting Prolog to
the Carnegie Group's ROCK/IMKA knowledge base.)

--
 Peter Ludemann  +1-415-813-3800  ludemann@quintus.com