Nested Functions in C#

• Download source - 5.6 KB

Introduction

We all have heard of nested classes, nested namespaces. Have we ever heard of nested functions??? Can we write a nested function in C#???

The answer is Yes, we can write a nested function in C# using delegates. This article deals with writing nested methods in C#. Knowledge of delegates is required. For creating nested methods, we write anonymous methods and associate delegates with them.

Using the Code

For writing a nested function in C#, we would make use of delegates and anonymous functions. A delegate is nothing but a "function pointer". And an anonymous function is a function/method without a name. Below, we are declaring a delegate which will point to an anonymous function.

// Declare a delegate which will point to different anonymous methods.
public delegate long FactorialDelegate(long n);        

After declaring the delegate, we will write the anonymous functions as given below:

// This method defines anonymous functions.           
public static void FunctionInFunction()
{
    // Nested function.
    FactorialDelegate FactorialMethod = delegate(long number)
    {
        if (number < 1)
            throw new ArgumentOutOfRangeException(
            "n", "Argument must be greater than 0!");

        long result = 1;

        // Calculate factorial
        for (int iterator = 1; iterator <= number; iterator++)
        result *= iterator;

        return result;
    };

    // Nested function with recursion.
    FactorialDelegate FactorialRecursive = delegate(long number)
    {
        if (number < 1) throw new
            ArgumentOutOfRangeException(
            "n", "Argument must be greater than 0");
    
    // The current method will always be at the 0th frame on the stack.
    MethodBase method = new StackTrace().GetFrame(0).GetMethod();

    return number > 1 ? number * (long)method.Invoke(null,
        new object[] { number - 1 }) : number;
     };

Console.WriteLine("Factorial for 5 = " + FactorialMethod(5));

Console.WriteLine("Factorial for 10 = " + FactorialRecursive(10));
}

In the above example, we have declared a delegate and in the method FunctionInFunction, we are associating that delegate with two anonymous methods we defined. The first one is a normal method and the second anonymous method is recursive in nature. In the recursive method, we have used reflection to get the recursive method name, i.e. invoking this method. The current method will always be on the frame 0 of the stack trace.

Some properties of anonymous methods are as follows:

• It is an error to have a jump statement, such as goto, break, or continue, inside the anonymous method block whose target is outside the block. It is also an error to have a jump statement, such as goto, break, or continue, outside the anonymous method block whose target is inside the block.
• Unlike local variables, the lifetime of the outer variable (outer variable is a variable declared local to the outside wrapper function) extends until the delegates that reference the anonymous methods are eligible for garbage collection.
• An anonymous method cannot access the ref or out parameters of an outer scope.
• No unsafe code can be accessed within the anonymous-method-block.

More information on anonymous function can be found here.

Points of Interest

This is a wonder that we can do with delegates.

History

• 23^rd April, 2008: Initial post