This is the mail archive of the
guile@cygnus.com
mailing list for the Guile project.
Re: global variables & data tags (a la X-Windows callbacks)
- To: robertb@continuumsi.com
- Subject: Re: global variables & data tags (a la X-Windows callbacks)
- From: Greg Harvey <Greg.Harvey@thezone.net>
- Date: 05 May 1999 15:38:20 -0230
- Cc: Marius Vollmer <mvo@zagadka.ping.de>, guile@cygnus.com
- References: <19990503161600.26837.qmail@continuumsi.com> <87r9owwn96.fsf@zagadka.ping.de> <19990505130502.8412.qmail@continuumsi.com>
robertb@continuumsi.com writes:
> > From: Marius Vollmer <mvo@zagadka.ping.de>
> > Date: Tue May 4, 11:14pm
> >
> > robertb@continuumsi.com (Robert Brown) writes:
> >
> > > Perhaps to avoid any lack-of-backward-compatibility, a new set of C
> > > functions could be devised?:
> >
> > I'm fairly certain that your proposed functions can be implemented on
> > top of the existing stuff, using compiled closures, but I've never
> > done that. Maybe I'll surprise you with some code tomorrow, but don't
> > count on it.
> >
> > - Marius
>
> Actually, I'm attempting to work around the problem via object-properties.
> If this works, then my proposal will not be necessary.
>
> (So far, though, it's become fairly messy. As I was telling someone else,
> I was hoping to "evolve" the code towards embedded-Scheme. It would be
> too much work to make everything "smob-able" given the size of the C
> program and this proposed (I guess) emulation of closures in C would ease
> the transition. ("Smob-ification" would have been a lot easier if I could
> have used SWIG, but apparently Guile support has been pulled.))
I just picked up swig to check this, and it does still have some guile
support. It appears to totally blow goats, and is based around guile
1.0, but it's in there ;).
--
Greg