number printing problems in 5.9.3 xmaxima on windows/GCL



Greetings!

Odd, I get (2.6.8pre):

(defun foo (x y) (declare (long-float x y)) (/ x y))

FOO

>(compile *)

Compiling /tmp/gazonk_9201_0.lsp.
End of Pass 1.  
End of Pass 2.  
OPTIMIZE levels: Safety=0 (No runtime error checking), Space=0, Speed=3
Finished compiling /tmp/gazonk_9201_0.lsp.
Loading /tmp/gazonk_9201_0.o
start address -T 0x8466e50 Finished loading /tmp/gazonk_9201_0.o
#<compiled-function FOO>
NIL
NIL

>(foo 1.0 0.0)

#<inf>

>(defun bar (x y) (declare (long-float x y)) (let ((x (expt x y))) (* x x)))

BAR

>(compile *)

Compiling /tmp/gazonk_9201_0.lsp.
End of Pass 1.  
End of Pass 2.  
OPTIMIZE levels: Safety=0 (No runtime error checking), Space=0, Speed=3
Finished compiling /tmp/gazonk_9201_0.lsp.
Loading /tmp/gazonk_9201_0.o
start address -T 0x8447eb0 Finished loading /tmp/gazonk_9201_0.o
#<compiled-function BAR>
NIL
NIL

>(bar 10.0 300.0)

#<inf>

>

Take care,

"Stavros Macrakis" <macrakis at gmail.com> writes:

> <camm at enhanced.com> wrote:
> > (setq si::*print-nans* t)
> 
> Great!  Now we have to make Maxima use it intelligently...
> 
> :lisp (setq si::*print-nans* t)
> 
> 1.0e300^2 => 1..#INF00e+000E+1  (yuck!)
> Lisp:   #<1.#INF00e+000>   (somewhat less yuck, but still yuck)
> 
> Don't have a suggestion for something better....
> 
> And of course Maxima still doesn't actually handle NaNs properly:
> 0.0*(1.0e300^2) => 0.0 (should be NaN); reader doesn't know anything
> about them, etc. etc.
> 
>           -s
> 
> 
> 

-- 
Camm Maguire			     			camm at enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah