Newsgroups: comp.emacs.xemacs,comp.emacs,comp.lang.clos
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech2!news.mathworks.com!tank.news.pipex.net!pipex!dish.news.pipex.net!pipex!exnet2!nri-e!usenet
From: gosnellj@uw0391.nomura.co.uk (Julian Gosnell)
Subject: CLOS, ELISP, XEMACS
Message-ID: <GOSNELLJ.95Nov28103757@uw0391.nomura.co.uk>
Lines: 50
Sender: usenet@nomura.co.uk (/var/spool/news)
Nntp-Posting-Host: 147.201.65.185
Organization: Nomura Research Institute Europe Ltd.
Date: Tue, 28 Nov 1995 10:37:57 GMT
Xref: glinda.oz.cs.cmu.edu comp.emacs.xemacs:7705 comp.emacs:30068 comp.lang.clos:3587


I have recently been reading 'The Art of the MetaObject Protocol', and
having no common Lisp implementations to hand copied the Closette
implementation from the book into my XEmacs and after a bit of hacking
got it up and running.

ELisp Question :

I can interpret it no problem - but it's slooooowwwww - I have been
trying to byte-compile it but since, in the boot procedure each step
assumes the environment resulting from the last I have had problems
and am getting increasingly frustrated.

e.g.

	(defclass foo () ())
	(defclass bar (foo) ())	 ;; this line relies on the last one

I have looked through the Info pages for LispRef and CL (which I have
written it on top of) - but to no avail. I have found
e.g. eval-[when/and]-compile which seem to further complicate the
situation.

I am beginning to think that my byte-compiler might be broken - I have
had problems with it before - as the reference says that it should
simply expand macros and then compile the result - mine seems to be
evaluating at compile time - hence the dependancy problem.

Could anyone shine some light on this for me ?

General Question :

Does anyone know of any other CLOSes or object systems available in
ELisp (or in C++) ? - I would be interested to learn from them.



Thanks for your time,



Jules


P.S.

My mail address is not currently operational - but I read the above
groups regularly.


