Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!wasatch!defun.utah.edu!sandra
From: sandra%defun.utah.edu@wasatch.UUCP (Sandra J Loosemore)
Newsgroups: comp.lang.lisp
Subject: Re: declarations
Message-ID: <1410@wasatch.UUCP>
Date: 23 Mar 89 17:11:08 GMT
References: <7325@phoenix.Princeton.EDU>
Sender: news@wasatch.UUCP
Reply-To: sandra%defun.utah.edu.UUCP@wasatch.UUCP (Sandra J Loosemore)
Organization: PASS Research Group
Lines: 27

In article <7325@phoenix.Princeton.EDU> eliot@phoenix.Princeton.EDU (Eliot Handelman) writes:
>Question about IGNORE:
>
>(defun foo (x)
>  (declare (ignore x))
>  :ignored)
>
>Should FOO complain about too many or too few
>arguments, or can I count on this happening in all CL's? Steele doesn't
>seem to have anything to say about it.

The IGNORE declaration doesn't have anything to do with checking for the
right number of arguments.  See CltL page 61:  "there must be exactly
as many arguments as there are parameters".  In other words, "it is
an error" if there aren't.  Sometimes an error is signalled, sometimes
it's harmless, and sometimes Bad Things will happen.  Some implementations
use OPTIMIZE declarations to decide whether to check for this error
in compiled code.

If you really want FOO to take any number of arguments and ignore them all,
try using:

(defun foo (&rest x)
  (declare (ignore x))
  :ignored)

-Sandra Loosemore (sandra@cs.utah.edu)