[Openmcl-devel] Compiler warnings

[Ron Garret]

On Oct 18, 2009, at 11:19 AM, Taoufik Dachraoui wrote:

>
> As a beginner how do I guess the behavior of SETF when I learned from
> the specs that there is three kinds of symbols: dynamic, lexical and  
> constant.

Very badly, apparently.  (Think about that.)

But if you meant to ask: How do I FIND OUT the behavior of SETF in the  
absence of a preceding DEFVAR?  Then the answer is: you can't learn  
that from the spec, because the spec simply says it's undefined.   
CCL's behavior (and every other Lisp implementation's behavior for  
that matter) is an EXTENSION to the spec.  To find out how a given  
implementation extends the spec in that case you have to go to a  
different source of information.

> But the (SETF X 1) in CCL does not create a dynamic variable as in
> CMUCL and this is the source of my confusion.

You are confused about more than just that.  Have you read the paper  
and the blog entry that I referred you to?  No, you (still) have not.   
Because if you had would no longer be confused (or at the very least  
you would be asking different questions).  You are not the first  
person to have the confusion(s) you are having.  But you are the the  
first (in my experience) to obstinately refuse to accept the answer(s)  
that you've been given.

rg