Functions may leak information about names(!) of localvariables.

The problem you have encountered is that t is a synonym for the truth value
true.

If you change t to tt  inside g(s), uniformly, then the answer you get for
g(t); is 64.

Is this a bug? Perhaps.  Lisp uses the variable "t"  for true, and this is
how the idea originated. Is this more confusing than useful at this time?
Probably.

At least in wxmaxima, if you write t[n]:=n^3; it is displayed as
true[n]:=n^3.

RJF


> -----Original Message-----
> From: maxima-bounces at math.utexas.edu [mailto:maxima-
> bounces at math.utexas.edu] On Behalf Of Michel Van den Bergh
> Sent: Thursday, December 28, 2006 5:33 AM
> To: maxima at math.utexas.edu
> Subject: Functions may leak information about names(!) of
> localvariables.
> 
> I really don't see how to do robust programming in maxima.
> 
> Consider
> 
> g(s):=block([t],local(t), t[4]:subvar(s,4), t[4]);
> t[n]:=n^3;
> u[n]:=n^4;
> 
> g(u);
>       256
> g(t);
>      t_4
> 
> What is the correct way of dealing with this? Or should one just say that
> arrays are not first class objects and should not be passed around in
> this way?
> That would be a pity.
> 
> Michel
> 
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima