Re: [boost] shared_ptr and std::less (was Re: Smart pointer status, directions, thoughts)

From: John Max Skaller (skaller_at_[hidden])
Date: 2001-05-27 09:57:24

David Abrahams wrote:
> 
> ----- Original Message -----
> From: "Darin Adler" <darin_at_[hidden]>
> 
> > not standard-conforming. But I now see that 17.3.4.1p1 specifically says
> > that the specialization must "meet the standard library requirements for
> > the original template", and that in the case of std::less, that means it
> > must have the same result as using the < operator directly.

> On first sight I'd say Peter may have uncovered a defect in the standard. I
> don't think we meant to say that about std::less<>.

        The intended requirement is that, where < is defined 
by the Standard, it must agree with less<>. In particular,
if two pointers a,b point into the same array, then

        assert(a < b == less(a,b))

On the other hand, the result of a<b isn't well defined
otherwise, so the above assertion need not be the case.

        less<> was introduced _solely_ to provide
a total ordering on pointers, since < cannot be 
relied upon to provide one. It is necessary
to create STL sets (etc) of pointers. 

-- 
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