number printing problems in 5.9.3 xmaxima on windows/GCL
Subject: number printing problems in 5.9.3 xmaxima on windows/GCL
From: Robert Dodier
Date: Sat, 26 Aug 2006 22:06:08 -0600
On 8/26/06, Richard Fateman <fateman at cs.berkeley.edu> wrote:
> %i3) 1.797693134862315d+308;
> (%o3) 1.7976931348623149E+308
> (%i4) %*2;
>
> Maxima encountered a Lisp error:
>
> Error in EVAL [or a callee]: Can't print a non-number.
That's a bug in GCL -- it computed floating point infinity
but then it didn't know how to print it. I have reported this
behavior to the GCL bug tracker.
CL is uncomfortable with floating point special values.
That's unfortunate.
CL implementations vary as to how carefully special
values are handled.
CMUCL & SBCL default to disallowing special values but
there is a way to set a feature bit to allow them, and then
the values are handled correctly. Clisp simply banishes
special values. GCL doesn't have a consistent policy --
computations can yield special values but then it
complains when displaying them.
What does Maxima + ACL show for 1.797693134862315d+308 * 2 ?
Robert