OM Floats

[Bruce R Miller]

On Jul 8,  9:19pm, huuskone at cc.helsinki.fi wrote:
>  ....  A routine that
> inputs an OMF element and returns a double-precision floating point
> number in IEEE format is certainly useful from that point of view.  On
> the other hand, what would the return value of a bigfloat input routine
> be?

The same as it does for a bigint ? A character string, I suppose.
That's what the Inria library does, which seems to be a reasonable approach
under the circumstances.

I dont know of the current implementation checks the syntax of the string, but
it certainly could, and that would be reasonable in the case of bigfloat too.
At least in the case of a bigfloat, the implementation could allow the
application to demand a double (although it might get an inf or nan if the
exponent were too large...)


-- 
--
bruce.miller at nist.gov
http://math.nist.gov/~BMiller/