Problem with $EV and $SOLVERADCAN [ECL port]

I don't have such a working system (only have ECL installed). Could anybody
tell me what should happen in $EV when the top-level DO-loop finds $SOLVERADCAN?

In my case:
	- the IF statement doesn't execute: (ATOM (CAR L)) is true
	- the COND statement goes to second clause
	- the OR statement goes to second alternative (LET)
	- (MEVAL (CAR L)) returns NIL

Michael.

> NIL should not have a property list.  There is probably
> a bug somewhere else. Try tracing in a working system.
> 
> RJF
> Goffioul Michael wrote:
> 
> > Hi,
> > 
> > I'm trying to port maxima to ECL, it's almost fully 
> functional but there's
> > still a problem with a limitation of ECL that is triggered 
> by maxima. The
> > (temporary) limitation is that you can't change property 
> list of the symbol
> > NIL. However, maxima tries to do so, but I don't know if 
> this is normal or
> > not.
> > 
> > The problem is triggered by "ev(solve(5^f(x) = 
> 125,f(x)),solveradcan);" (coming
> > from rtest8.mac). The backtrace is more or less:
> > 	$ev [mlisp.lisp: 1244]
> > 	$nounify [spurv1.lisp: 918]
> > 	putprop [clmacs.lisp: 218]
> > $NOUNIFY is called with FL=NIL, resulting from the (MEVAL 
> (CAR L)), where
> > (CAR L) == $SOLVERADCAN.
> > 
> > Note that I'm using CVS from about a week.
> > 
> > Is this what is expected from maxima execution, or is there 
> a problem somewhere?
> > Thanks.
> > 
> > Michael.
> > 
> > _______________________________________________
> > Maxima mailing list
> > Maxima@www.math.utexas.edu
> > http://www.math.utexas.edu/mailman/listinfo/maxima
> 
>