Re: [boost] Fixed-Point Decimal Type

From: John Max Skaller (skaller_at_[hidden])
Date: 2001-05-20 01:49:48

Beman Dawes wrote:
> 
> Daniel Frey wrote:
> 
>  >In finance there is another, sometimes important way of rounding: On a
>  >tie, round so that the last digit is even. To illustrate that:
> 
> Gabriel Dos Reis wrote:
> 
>  > I think the ability of choosing rounding mode should not be
>  > overlooked.  It is an essential part of LIA-1 and later standards.
> 
> I'm convinced.
> 
> So a question is do you support that by having a whole bunch of different
> functions, or by the class template taking some kind of a parameter
> (perhaps a traits or policy class) performing the rounding desired?

        Rounding is a property of the operation,
NOT the operands. For example, with args scale n,m,
and result fixed at scake n+m, there's no rounding
to consider!

        If you want scale n * scale n -> scale n,
we could have

        template<typename rounding_mode, int n>
        decimal<n> mul(decimal<n>,decimal<n>)

and partially specialise per rounding mode, so that
you call by, for example,

        mul<round_up>(a,b)

This is the same as

        mul_round_up(a,b) ...

except the use of a rounding mode type makes naming
various other functions (such as div, etc) more consistent.

-- 
John (Max) Skaller, mailto:skaller_at_[hidden]
10/1 Toxteth Rd Glebe NSW 2037 Australia voice: 61-2-9660-0850
checkout Vyper http://Vyper.sourceforge.net
download Interscript http://Interscript.sourceforge.net