Re: Interest in xfig

[Russ McManus <russell.mcmanus@gs.com>]

(nobody) writes:

> This is really only a starting point but the next step is to make xfig's
> drawing objects into SMOBs and that is a big job. 

Probably it is just a bid typing job.  With a little Emacs macrology,
I get it is doable.

> Also that has some intrinsic problems because xfig can delete its
> drawing objects from the drawing at any time so I will have to
> modify the deletion operation to check is guile is keeping a
> reference to the object. If it is then there will need to be a
> mechanism for tagging the object as unlinked from xfig but still
> referenced by guile. I can't copy the objects into guile either
> because about the only unique way to keep hold of an xfig object is
> to keep its pointer. I'll think about these issues some other day.

This isn't that hard, really.  I've run into this problem in the past,
and I just took the Emacs solution for buffer lifetimes.  Under this
model, the C code is in control of the object's lifetime, however the
object may live longer in the Lisp world, but as a 'dead' object.  So
the C code needs to check the first time it unboxes a Lisp buffer
object whether the buffer is live or dead.  The Lisp world can avoid
errors by checking 'buffer-live-p'.  It's not the most elegant
solution, but it works.

> Does anyone else use xfig? Would this be something worth persuing in
> a more serious manner?

Yes, I use xfig.  I would be happy to be a beta tester.

-russ

--
Your mouse has moved. Windows NT must be restarted for the change
to take effect. Reboot now?  [ OK ]