Why overload true and false instead of defining bool operator?

I've been reading 开发者_如何学Pythonabout overloading true and false in C#, and I think I understand the basic difference between this and defining a bool operator. The example I see around is something like:

public static bool operator true(Foo foo) {
  return (foo.PropA > 0);
}
public static bool operator false(Foo foo) {
  return (foo.PropA <= 0);
}

To me, this is the same as saying:

public static implicit operator bool(Foo foo) {
  return (foo.PropA > 0);
}

The difference, as far as I can tell, is that by defining true and false separately, you can have an object that is both true and false, or neither true nor false:

public static bool operator true(Foo foo) { return true; }
public static bool operator false(Foo foo) { return true; }
//or
public static bool operator true(Foo foo) { return false; }
public static bool operator false(Foo foo) { return false; }

I'm sure there's a reason this is allowed, but I just can't think of what it is. To me, if you want an object to be able to be converted to true or false, a single bool operator makes the most sense.

Can anyone give me a scenario where it makes sense to do it the other way?

Thanks

---

As the docs say, overloading true and false is intended to support (nullable) database-types (Yes/No, Y/N, 0/1, etc).

And of course you can define them inconsistently, as with any operator. It is your responsibility to return something sensible. The compiler goes no further than requiring neither or both.

---

I had no idea these operators existed. That means you can implement the self-negation paradox:

public class ThisClassIsFalse
{
    public static bool operator true(ThisClassIsFalse statement)
    {
        return statement ? false : true;
    }

    public static bool operator false(ThisClassIsFalse statement)
    {
        return statement ? true : false;
    }
}

So now we know the true solution to this classic paradox... StackOverflowException.

---

I've seen people overload the true and false overloads in order to do clever things like building expressions in .NET 2.0, before Linq existed.

Ayende worked out a syntax like this to build NHibernate criteria queries, using his NHQG project:

return Repository.FindAll(
    (Where.Publisher.Name == name) &&
    (Where.Publisher.City == city));

---

Depending on the system, true can be any non-zero value. In others, it can be any positive value.

Other systems aren't truly boolean, and allow a third state null or nill for the boolean values, which is why you might overload true and false, versus overloading a single bool operator.