Question

If we have the following sample application:

interface ITest
{
    string Test { get; }
}

class A : ITest
{
    string ITest.Test { get { return "Test from A!"; } }
}

class B : A, ITest
{
    string ITest.Test { get { return "Test from B!"; } }
}

Given an instance of B, is it possible to access A’s implementation of ITest?
For example:

B b = new B();
ITest test = b;
string value = test.Test; // "Test from B!"
A a = b;
test = a;
value = test.Test; // Still "Test from B!"

Note, this is no real world problem but more of a general wondering.

Answer 1

No, it’s not. At least not normally – it’s possible that you could do it with reflection.

Basically, by reimplementing ITest, B is saying that it’s taking complete responsibility for the implementation of ITest.Test within any object of type B – and you can’t even call it from within B which you’d normally be able to if you were overriding in the usual way.

EDIT: I’ve just proved (in a hacky way) that you can call it with reflection:

using System;

public interface IFoo
{
    void Foo();
}

public class Base : IFoo
{
    void IFoo.Foo()
    {
        Console.WriteLine("Base");
    }
}

public class Derived : Base, IFoo
{
    void IFoo.Foo()
    {
        Console.WriteLine("Derived");
    }
}

class Test
{
    static void Main()
    {
        var map = typeof(Base).GetInterfaceMap(typeof(IFoo));            
        var method = map.TargetMethods[0]; // There's only one method :)
        method.Invoke(foo, null);
    }
}

This prints out “Base”. It’s pretty horrible though – I’d have to be desperate to do it…