opCmp

[Ali Çehreli]

On 02/03/2012 06:44 AM, Gor Gyolchanyan wrote:
 > Good day.
 >
 > There's a problem in how opCmp works.
 >
 > I have a structure, that represents an element of a range. Let's say a
 > character. That character can be invalid.
 > I need all comparison operators to return false of at least one of the
 > operands is invalid.

As an observation, you want to implement the concept of "unordered" for 
types, similar to floating point types:

   http://dlang.org/expression.html#floating_point_comparisons

I am very surprised that the following operator works with 
non-floating-point types:

class C
{
     override int opCmp(Object o)
     {
         return 0;
     }
}

void main()
{
     auto c = new C;
     auto b = (c !<>= c);      // <-- compiles!

     int i, j;
     auto b2 = (i !<>= j);     // <-- compiles!
}

Is that supported? Is it a bug? Would using those /unordered/ operator 
help in your case?

 > with opCmp, the expression a @ b is rewritten as a.opCmp(B) @ 0, which
 > doesn't allow me to define such a logic.
 > wouldn't it be better to change the rewrite of opCmp to test for exact
 > values -1, 0 and 1? In that case I could return 2 and have all
 > comparisons fail.
 >

Ali