From: David Richardson (dgr24_at_[hidden])
Date: 2003-10-14 13:39:01
Deane Yang wrote:
> David Abrahams wrote:
>> The fractional power thing is strange, but OK. Anyway, it's
>> essentially the same approach as mine. What Hugo's explanation is
>> missing is any demonstration of how to compute the new Dimensions type
>> which results from multiplication or division.
> Darn, that's the crucial point. Using Hugo's approach, the resulting
> type should be a typelist that contains the union of the two
> input typelists, with the rational exponents equal to the sum
> of the corresponding exponents in the input typelist.
The code I've upload is capable of handling fractional exponents.
It uses ints for the numerator and denominator. I considered
using arbitray length compile time integers, but that seemed like
too much work.
I'm cleaning up the code I put in yahoo (fixed a few bugs, gotten
it to compile on another compiler) and will send an update soon.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk