🚀 go-pugleaf

Andrue Cope [TeamB] wrote:

> The main objection I have to this is that if you overload '=='
> shouldn't you also overload '>' etc. (and what would '>' mean in this
> context come to that)?

By the same argument, would you say that complex number classes should
not implement operator== as you cannot (mathematically) order two such
numbers?

What about vector classes?

If operator== is clear in the context, then it is valuable.  If it is
at all open to questions and vagueness, then I say use a named function
call instead.  You might like to read Thomas reply to your question
before deciding how clear the equivalence is <g>

Always think about how someone else could mis-interpret your operator
overload before deciding to go with it.  They are an incredibly
valuable shorthand for communicating ideas, cross language boundaries
well in international teams, beat easily lead to mis-understanding if
not 100% closed to mis-interpretation.

You also might like to search the web for comments by Dave Abrahams on
DSELs. Domain Specific Embedded Languages.  This is something that C++
does very well, and I find it hard to think of operator overloading in
any other context now.

AlidairM