Re: [Gcl-devel] Re: [Maxima] float to bfloat

[Richard Fateman]

Raymond Toy wrote:

> I don't think I would want my Lisp to be able to integer-decode-float
> a double-float positive infinity, assuming that that is what (expt
> 9.9d0 9999) has returned.  
>
>  
I disagree:
That means any almost user of integer-decode-float must guard the call 
by a check to see if
the argument is inf or nan, and in case it IS an nan, with no way to 
decode its fraction, which
may in in fact have valuable information.  e.g. the program counter when 
it was produced,
or a pointer into a hash-table of a diagnostic message, etc.  So if 
integer-decode-float does
not decompose a NaN into pieces, someone intent on using NaNs would have 
to write
a program exactly like integer-decode-float.  (I've gone this route, 
actually..)

 You can't store extra information in Inf, but
still, you should not have to check for it specially, in my view.

RJF