Nullable<T> and Nullable

This is another post in the series "Interview questions" which intends to give short answers to popular technical interview questions.

Q: What's the difference between int and int? declarations?

The int? notation is a shorthand for writing Nullable<int>. This type is a wrapper for value types that allows them to take null as a valid value.

This is useful for cases when a variable (or field, or property) is expected to have an indeterminate state, meaning it was not initialized as opposed to just being set to default value.

Q: How is Nullable implemented?

Generic type Nullable<T> is defined as a struct. There are also some helper methods exposed on a non-generic Nullable type which is a static class.

Below you can see an example implementation of Nullable<T>, the actual implementation in .NET Core/.NET Framework may be slightly different.

public struct Nullable<T> where T : struct
{
    private readonly bool _hasValue;
    private readonly T _value;

    public bool HasValue => _hasValue;

    public T Value => _hasValue ? _value : throw new InvalidOperationException();

    public Nullable(T value)
    {
        _hasValue = true;
        _value = value;
    }

    public T GetValueOrDefault() => _hasValue ? _value : default(T);

    public T GetValueOrDefailt(T defaultValue) => _hasValue ? _value : defaultValue;

    public override bool Equals(object other) => _hasValue ? value.Equals(other) : other == null;

    public override int GetHashCode() => _hasValue ? value.GetHashCode() : 0;

    public override string ToString() => _hasValue ? value.ToString() : "";

    public static implicit operator Nullable<T>(T value) => new Nullable<T>(value);

    public static explicit operator T(Nullable<T> value) => value.Value;
}

The generic constraint T : struct makes sure that this type can only be used with value types.

Members of the Nullable<T> struct include a property HasValue which indicates whether the value was initialized, and Value property which contains the actual value. Accessing the Value property when HasValue is false throws an exception.

There is also GetValueOrDefault() method that allows safely retrieving the value with a fallback.

To facilitate equality comparison, base methods Equals() and GetHashCode() have been overridden. The Equals method returns true if compared with null while HasValue is false.

Finally, this type implements implicit conversion between T and T? that lets us do int? a = 5, and also an explicit conversion between T? and T which lets us convert the other way around but will throw an exception if the value isn't set.

Q: How is it possible that we can assign null to Nullable<T> even though it's a struct?

Compiler provides syntactic sugar when working with Nullable<T>:

• When you write int? a = null it gets compiled to Nullable<int> a = new Nullable<int>().
• When you write if (a == 15) (where a is int?) it gets compiled to if (a.GetValueOrDefault() == 15 && a != null)

Q: What happens when you pass an instance of Nullable<T> to a method that expects an object?

In such cases, to avoid "double boxing", the Nullable<T> wrapper is discarded and you either get a T instance boxed as object or a null reference, depending on whether HasValue is true or false.