You can use the errormsg variable. But there is a bug:
http://sourceforge.net/tracker/index.php?func=detail&aid=1994295&group_id=4933&atid=104933
HTH,
Andrej
2008/9/16 Barton Willis <willisb at unk.edu>:
>
> Do we have a way to suppress error messages? I tried ttyoff : true,
> but that didn't work.
>
> It might be nice if errcatch automatically suppressed error
> messages. Example: 'safer_subst' substitutes into an algebraic
> expression. When the substitution yields division by zero, it
> rationalizes the expression and tries again: (Alternatively, 'limit'
> might be a better way to do this; nevertheless, I'd still like to know
> how to suppress error messages).
>
> (%i29) sol : solve(a*x^2 + x + 1,x);
> (%o29) [x=-(sqrt(1-4*a)+1)/(2*a),x=(sqrt(1-4*a)-1)/(2*a)]
>
> I'd like to suppress the division by 0 messages:
>
> (%i30) safer_subst([a=0],sol);
> Division by 0
> Division by 0
> (%o30) [x=und,x=-1]
>
> equationp(e) := block([inflag : true], not(mapatom(e)) and is(op(e) =
> "="));
>
> recip(e) := if listp(e) or equationp(e) then map('recip, e)
> else if is(equal(e,0))=true then 'und else 1/e;
>
> safer_subst(l, e) := block([algebraic : true, r, ttyoff : true],
> if listp(e) or equationp(e) then map(lambda([s], safer_subst(l,s)), e)
> else (
> r : errcatch(subst(l,e)),
> if emptyp(r) then recip(safer_subst(l, ratsimp(1/e))) else first(r)));
>
> Barton
>
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima