How is null represented in .NET

Question

I was having a conversation with a co-worker and the subject of null came up. He was telling me that in .NET behind the scenes it is just a really small number. I always tho

Answer 1

From MSDN:

The null keyword is a literal that represents a null reference, one that does not refer to any object

To try and make it a little clearer in the older versions of c#, a value type couldn't be null, ie, it HAD to have a value, if you didn't assign one, you got a potentially random value so something like:

int i;
i++;
console.writeline(i);

in the older versions of c# objects had to be initialized otherwise they were null, which meant they had no reference to any object.

Now with nullable value types in c# 2.0+ you can have a nullable int which means that if you have this code:

int? i;
i++;
console.writeline(i);

you will actually get an exception at i++ because i has never been initialized to anything other than null. If null was 0, this code would execute fine because it'd just evaluate to 0+1, however this is incorrect behaviour.

If null was always 0 and you had a a nullable int, and you wrote some code like:

int? i;
if (int == 0)
{
//do something
}

there is a very real possibility you could get some unexpected behaviour IF null was the same as 0 because there is no way the compiler could differentiate between the int being null and the int being explicitly set to 0.

Another example that clarifies things in my mind:

public int? AddNumbers(int? x, int? y)
{
    if (x == null || y == null)
        return null;
    if (x == 0)
        return y;
    if (y == 0)
        return x;

    return x + y;
}

in this example, it's clear that null and 0 are very different because if you were to pass in 0 for x or y, and null was equal to 0, then the code above would never get to the checks for x == 0 or y == 0, however if you run the code and pass in 0 for x or y, the checks do get executed.