From: Rainer Schöpf <rainer.schoepf@gm...>  20130410 18:15:40

On Mon, 4 Feb 2013 at 09:32 0000, Arthur Norman wrote: > Another glitch, noted here so it is on the record... > > 1: load_package specfn; > > 2: zeta_function(2, a + b*x); > > 50000000000000*abs(sqrt(  a  b*x + 1)  1)  1 > *****  invalid as number > 100000000000000 > > > and actually all sorts of second arguments to zeta_function(2,...) seem to > cause trouble... for instance I seem to find zeta_function(2,1) just > falling into a CPU sink. Now I understand: zeta_function is for numeric (rounded) arguments only: JacobiZeta(~u,~m) => Num_Elliptic(ZETA_function,u,m) when lisp !*rounded and numberp u and numberp m where Num_Elliptic(fn,a,b) is basically a wrapper that sets the precision for the numerical computation of fn(a,b); I think the best solution is to change the name zeta_function to something like JacobiZeta!:numeric and add a bit of documentation. Rainer 