AMDG Bruno Lalande wrote:
Yes I think Boost.Math is the right place for this too. And Boost.Unit will be able to take it there instead of having it confined in a "units::details" or something, this way everybody can benefit of this useful thing.
I can write tests and docs for you, as well as study the different performance / optimization aspects with some other compilers. As I've never done this before, I'll have to read the different guidelines describing the way in which such contributions have to be made, so it can take a few days. But I think we're not in a hurry.
Steven, could you tell me if some aspects of the implementation you made for Boost.Unit are better than mine? This would permit to be sure to have the best implementation in Boost.Math.
Your implementation seems to be faster in the few tests I ran on msvc 9.0. It would be nice if boost::units::pow could be completely replaced by this utility, so here are the issues that would need to the covered: * The result type is not necessarily going to be the same as the arguments so it would need to use Boost.Typeof or some such. * Rational exponents need to work, so I would need a way to specialize the function to work with boost::units::static_rational. I would expect the customization mechanism to look something like this: struct default_value_tag; template<class T> struct pow_value_tag { typedef default_value_tag tag; }; template<class Exp> struct pow_exponent_tag { typedef typename Exp::tag type; }; template<class ValueTag, class ExponentTag> struct pow_impl; template<class Value, class Exponent> struct pow { typedef typename pow_impl< typename pow_value_tag<Value>::type, typename pow_exponent_tag<Exponent>::type >::template apply<Value, Exponent> impl; typedef typename impl::type type; static type call(const Value& v) { return(impl::call(v)); } }; In Christ, Steven Watanabe