Path: utzoo!attcan!uunet!mailrus!tut.cis.ohio-state.edu!ucbvax!pasteur!jwz@teak.berkeley.edu
From: jwz@teak.berkeley.edu (Jamie Zawinski)
Newsgroups: comp.lang.lisp
Subject: Re: code
Message-ID: <19727@pasteur.Berkeley.EDU>
Date: 19 Nov 89 06:26:18 GMT
Sender: news@pasteur.Berkeley.EDU
Lines: 57
In-reply-to: <1989Nov18.152356.7962@hellgate.utah.edu>

In article <1989Nov18.152356.7962@hellgate.utah.edu> Tim Moore <moore%cdr.utah.edu@cs.utah.edu> writes:
> If you want persistant lexical bindings you probably should be using closures
> instead. Here's you fibonacci example, rewritten to use a closure:
> 
[ incorrect version replaced: ]
> (defun fib-setup (&key ((:n-2 pn-2) 0) ((:n-1 pn-1) 1))
>   #'(lambda (&key n-2 n-1)
>       (and n-2 (setq pn-2 n-2))
>       (and n-1 (setq pn-1 n-1))
>       (psetq pn-2 pn-1 pn-1 (+ pn-2 pn-1))
>       pn-1))
> FIB-SETUP
> (setf (symbol-function 'fibonacci) (fib-setup))
> #<Interpreted Closure #x1AB688>
> (fibonacci)
> 1
> (fibonacci)
> 2
> (fibonacci)
> 3
> (fibonacci)
> 5

This seems to me to be a more straightforward way:

(let ((pn-2 0)
      (pn-1 1))
  (defun fibonacci ()
    (psetq pn-2 pn-1
	   pn-1 (+ pn-2 pn-1))
    pn-1))

Since DEFUN must expand to something that stuffs a lambda into a 
function-cell, it's equivalent to yours, but is shorter and easier 
to read.

A problem with this is, some less-swift compilers might not be smart enough
to compile top-level lexical closures like this.  (But that's wrong.)

> Note: I too think it would be a good thing if people posted more code
> and pointers to code.

Ok, here's one.  The directory /usr/jwz/public/ on spice.cs.cmu.edu is
accessible by anonymous FTP.  This directory contains 4 or 5 megs of Lisp
code.  Most of it (about 80% I think) is TI Explorer specific, but a fair
amount is common lisp; there's a good chance that a lot of the
Explorer-specific code will run on Symbolics, since they share common
ancestors.  Most of the machine dependancies are in user-interface or
graphics stuff, so it might not be hard to port to other implementations.
The file "_readme.text" contains short descriptions of all of the files, 
along with whether they are CL-compatible.

The Spice's IP address is 128.2.254.139, for those of you with brain-dead
nameservers.  If you have any neat tools that you'd like to see become more
accessible, send them to me.

		-- Jamie (jwz@teak.berkeley.edu or jwz@spice.cs.cmu.edu)