Re: [boost] strict_weak_ordering

From: David Abrahams (dave_at_[hidden])
Date: 2002-11-25 11:02:42

Ben Young <ben.young_at_[hidden]> writes:

> On Mon, 25 Nov 2002, David Abrahams wrote:
>
>> Ben Young <ben.young_at_[hidden]> writes:
>>
>> > Recently at our company we had a couple of bugs relating to people
>> > incorrectly writing comparison operators, so I came up with this little
>> > helper. It is very trivial, but useful none the less as bogus comparison
>> > operators can be the cause of many bugs.
>> >
>> > Obviously it could be improved to take a predictate of some kind and
>> > checged to use the boost pre-processor library, but, otherwise, is anyone
>> > interested
>>
>> Did you ever consider using
>>
>>     make_tuple(x1, y1, z1) < make_tuple(x2, y2, z2)
>>
>> ?
>
> Yes, but actually I am of the feeling that perhaps the tuple library
> should be using the strict_... form rather than the other way around.
>
> Won't the make_tuple method make a copy of all its arguments?

OK:

        make_tuple(ref(x1), ref(y1), ref(z1)) < ...

Anyway, in the past I have needed/wanted the strict_... function
family you propose, but:

       1. I would call it lexicographic_ordering

       2. I would like to be able to specify the comparison used for
          each of the parameters
-- 
                       David Abrahams
   dave_at_[hidden] * http://www.boost-consulting.com
Boost support, enhancements, training, and commercial distribution