[q-lang-users] Q 7.2 RC1 Issue 3: IEEE floats

[John C. <co...@cc...>]

Currently, Q floating-point operations do not conform to IEEE floats,
even though that is the underlying implementation on all important
machines.  Specifically, the support for infinities and NaNs is deficient.
Currently, the following expressions are evaluated as follows:

==> def pinf = 1.0e307 * 1.0e307
==> def ninf = -pinf
==> pinf
inf
==> ninf
-inf
==> pinf + ninf
nan

But none of "on inf", "-inf", or "nan" has the correct meaning input.
This should be cleaned up.

Furthermore, currently an attempt to do a floating-point division by
zero is made to fail, rather than returning the IEEE-appropriate results
(negative or positive infinity, unless the numerator is also zero,
in which case NaN).

This should be cleaned up by defining a syntax for these values to
be used on input and output, and by not failing when doing a floating
division by zero (though an exact division by zero should still fail).

-- 
Overhead, without any fuss, the stars were going out.
        --Arthur C. Clarke, "The Nine Billion Names of God"
                John Cowan <co...@cc...>