Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!mcsun!inria!ilog!davis
From: davis@ilog.UUCP (Harley Davis)
Newsgroups: comp.lang.lisp
Subject: Re: FUNCALL question
Message-ID: <6904@ilog.UUCP>
Date: 2 Feb 90 13:24:33 GMT
Organization: ILOG, Paris, France
Lines: 41
In-reply-to: gat@robotics.Jpl.Nasa.Gov's message of 1 Feb 90 00:15:46 GMT


 > > interpreted as a variable.  Erann Gat made the mistake of calling
 > > these two interpretations the symbol-function and symbol-value
 > > when those terms are properly applied only to symbols (acting
 > > as global variables and function names).  Local variables and
 > > function names don't have symbol-values or symbol-functions in
 > > that sense.  
 > 
 > I want to thank Jeff Dalton for coming to my defense.  However, the
 > above is totally incorrect (said with a smile).  Temporary variables
 > do have value and function bindings.  That is why there are LET and
 > FLET forms in CL.  For example:
 > 
 > (flet ((x (x) (+ x x)))
 >   (let ((x 1))
 >     (x x)))

The above is totally incorrect (:^)).  Jeff correctly pointed out that
lexically-scoped variables do not, properly speaking, have
symbol-values or symbol-functions, because these variables, despite
all appearances, are not symbols.  Special variables, on the other
hand, are symbols, to which symbol-function and symbol-value are
applicable.   The non-symbolhood of lexical variables permits nifty
compiler optimizations.

The following code is erroneous unless x has been proclaimed special:

(flet ((x (x) (+ x x)))
  (let ((x 1))
    (list (symbol-function x) (symbol-value x))))

[ref. CLtL edition 1 p. 90]

-- Harley

------------------------------------------------------------------------------
Harley Davis			internet: davis@ilog.fr
ILOG S.A.			uucp:  ..!mcvax!inria!davis
2 Avenue Gallie'ni, BP 85	tel:  (33 1) 46 63 66 66	
94253 Gentilly Cedex		
France