From: Douglas Gregor (gregod_at_[hidden])
Date: 2002-09-05 19:41:03


On Thursday 05 September 2002 06:13 pm, David Abrahams wrote:
> It's not unreasonable to think of using a rel_ops approach for these
> things, though the operators won't propagate into generic algorithms:
>
> namespace mine {
> boost::interval f(boost::interval* start, boost::interval* finish)
> {
> using boost::intervals::my_preferred_comparison_ops;
>
> ... start[0] < start[1] ... ; // ok
>
> std::sort(start, finish); // error
> };
> }

Thanks Dave! This is the only solution I've seen so far that doesn't increase
the size of my bump of trouble. It allows coexistence of different comparison
semantics without choosing a default (apparently, no good default exists),
and without unnaturally binding the type of comparison into the interval.

        Doug