Delegates and Events in C#

Delegates and Events

A delegate in C# is similar to a function pointer in C or C++. Using a delegate allows the programmer to encapsulate a reference to a method inside a delegate object. The delegate object can then be passed to code which can call the referenced method, without having to know at compile time which method will be invoked. Unlike function pointers in C or C++, delegates are object-oriented, type-safe, and secure.

A delegate declaration defines a type that encapsulates a method with a particular set of arguments and return type. For static methods, a delegate object encapsulates the method to be called. For instance methods, a delegate object encapsulates both an instance and a method on the instance. If you have a delegate object and an appropriate set of arguments, you can invoke the delegate with the arguments.

An interesting and useful property of a delegate is that it does not know or care about the class of the object that it references. Any object will do; all that matters is that the method's argument types and return type match the delegate's. This makes delegates perfectly suited for "anonymous" invocation.

Note
Delegates run under the caller's security permissions, not the declarer's permissions.

There are two examples which demonstrate the following,

Example 1 shows how to declare, instantiate, and call a delegate.
Example 2 shows how to combine two delegates.

Example 1 [shows how to declare, instantiate, and call a delegate]

The following example illustrates declaring, instantiating, and using a delegate. The BookDB class encapsulates a bookstore database that maintains a database of books. It exposes a method ProcessPaperbackBooks, which finds all paperback books in the database and calls a delegate for each one. The delegate type used is called ProcessBookDelegate. The Test class uses this class to print out the titles and average price of the paperback books.

The use of delegates promotes good separation of functionality between the bookstore database and the client code. The client code has no knowledge of how the books are stored or how the bookstore code finds paperback books. The bookstore code has no knowledge of what processing is done on the paperback books after it finds them.

// bookstore.cs  
using System;  
// A set of classes for handling a bookstore:  
namespace Bookstore {  
    using System.Collections;  
    // Describes a book in the book list:  
    public struct Book {  
        public string Title; // Title of the book.  
        public string Author; // Author of the book.  
        public decimal Price; // Price of the book.  
        public bool Paperback; // Is it paperback?  
        public Book(string title, string author, decimal price, bool paperBack) {  
            Title = title;  
            Author = author;  
            Price = price;  
            Paperback = paperBack;  
        }  
    }  
    // Declare a delegate type for processing a book:  
    public delegate void ProcessBookDelegate(Book book);  
    // Maintains a book database.  
    public class BookDB {  
        // List of all books in the database:  
        ArrayList list = new ArrayList();  
        // Add a book to the database:  
        public void AddBook(string title, string author, decimal price, bool paperBack) {  
            list.Add(new Book(title, author, price, paperBack));  
        }  
        // Call a passed-in delegate on each paperback book to process it:   
        public void ProcessPaperbackBooks(ProcessBookDelegate processBook) {  
            foreach(Book b in list) {  
                if (b.Paperback)  
                    // Calling the delegate:  
                    processBook(b);  
            }  
        }  
    }  
}  
// Using the Bookstore classes:  
namespace BookTestClient {  
    using Bookstore;  
    // Class to total and average prices of books:  
    class PriceTotaller {  
        int countBooks = 0;  
        decimal priceBooks = 0.0 m;  
        internal void AddBookToTotal(Book book) {  
            countBooks += 1;  
            priceBooks += book.Price;  
        }  
        internal decimal AveragePrice() {  
            return priceBooks / countBooks;  
        }  
    }  
    // Class to test the book database:  
    class Test {  
        // Print the title of the book.  
        static void PrintTitle(Book b) {  
            Console.WriteLine(" {0}", b.Title);  
        }  
        // Execution starts here.  
        static void Main() {  
            BookDB bookDB = new BookDB();  
            // Initialize the database with some books:  
            AddBooks(bookDB);  
            // Print all the titles of paperbacks:  
            Console.WriteLine("Paperback Book Titles:");  
            // Create a new delegate object associated with the static   
            // method Test.PrintTitle:  
            bookDB.ProcessPaperbackBooks(new ProcessBookDelegate(PrintTitle));  
            // Get the average price of a paperback by using  
            // a PriceTotaller object:  
            PriceTotaller totaller = new PriceTotaller();  
            // Create a new delegate object associated with the nonstatic   
            // method AddBookToTotal on the object totaller:  
            bookDB.ProcessPaperbackBooks(new ProcessBookDelegate(totaller.AddBookToTotal));  
            Console.WriteLine("Average Paperback Book Price: ${0:#.##}", totaller.AveragePrice());  
        }  
        // Initialize the book database with some test books:  
        static void AddBooks(BookDB bookDB) {  
            bookDB.AddBook("The C Programming Language", "Brian W. Kernighan and Dennis M. Ritchie", 19.95 m, true);  
            bookDB.AddBook("The Unicode Standard 2.0", "The Unicode Consortium", 39.95 m, true);  
            bookDB.AddBook("The MS-DOS Encyclopedia", "Ray Duncan", 129.95 m, false);  
            bookDB.AddBook("Dogbert's Clues for the Clueless", "Scott Adams", 12.00 m, true);  
        }  
    }  
}

Output

Paperback Book Titles:
The C Programming Language
The Unicode Standard 2.0

Dogbert's Clues for the Clueless
Average Paperback Book Price: $23.97

Code Discussion

Declaring a delegate

The following statement,

public delegate void ProcessBookDelegate(Book book);

declares a new delegate type. Each delegate type describes the number and types of the arguments, and the type of the return value of methods that it can encapsulate. Whenever a new set of argument types or return value type is needed, a new delegate type must be declared.

Instantiating a delegate

Once a delegate type has been declared, a delegate object must be created and associated with a particular method. Like all other objects, a new delegate object is created with a new expression. When creating a delegate, however, the argument passed to the new expression is special - it is written like a method call, but without the arguments to the method.

The following statement,

bookDB.ProcessPaperbackBooks(new ProcessBookDelegate(PrintTitle));

creates a new delegate object associated with the static method Test.PrintTitle. The following statement,

bookDB.ProcessPaperbackBooks(new ProcessBookDelegate(totaller.AddBookToTotal));

creates a new delegate object associated with the nonstatic method AddBookToTotal on the object totaller. In both cases, this new delegate object is immediately passed to the ProcessPaperbackBooks method.

Note that once a delegate is created, the method it is associated with never changes - delegate objects are immutable.

Calling a delegate

Once a delegate object is created, the delegate object is typically passed to other code that will call the delegate. A delegate object is called by using the name of the delegate object, followed by the parenthesized arguments to be passed to the delegate. An example of a delegate call is,

processBook(b);

A delegate can either be called synchronously, as in this example, or asynchronously by using BeginInvoke and EndInvoke methods.

Example 2 [shows how to combine two delegates]

This example demonstrates composing delegates. A useful property of delegate objects is that they can be composed using the "+" operator. A composed delegate calls the two delegates it was composed from. Only delegates of the same type can be composed.

The "-" operator can be used to remove a component delegate from a composed delegate.

// compose.cs  
using System;  
delegate void MyDelegate(string s);  
class MyClass {  
    public static void Hello(string s) {  
        Console.WriteLine(" Hello, {0}!", s);  
    }  
    public static void Goodbye(string s) {  
        Console.WriteLine(" Goodbye, {0}!", s);  
    }  
    public static void Main() {  
        MyDelegate a, b, c, d;  
        // Create the delegate object a that references   
        // the method Hello:  
        a = new MyDelegate(Hello);  
        // Create the delegate object b that references   
        // the method Goodbye:  
        b = new MyDelegate(Goodbye);  
        // The two delegates, a and b, are composed to form c:   
        c = a + b;  
        // Remove a from the composed delegate, leaving d,   
        // which calls only the method Goodbye:  
        d = c - a;  
        Console.WriteLine("Invoking delegate a:");  
        a("A");  
        Console.WriteLine("Invoking delegate b:");  
        b("B");  
        Console.WriteLine("Invoking delegate c:");  
        c("C");  
        Console.WriteLine("Invoking delegate d:");  
        d("D");  
    }  
}

Output

Invoking delegate a:
Hello, A!
Invoking delegate b:
Goodbye, B!
Invoking delegate c:
Hello, C!
Goodbye, C!
Invoking delegate d:
Goodbye, D!

Delegates and Events

Delegates are ideally suited for use as events - notifications from one component to "listeners" about changes in that component. For more information on the use of delegates for events,

Delegates vs. Interfaces

Delegates and interfaces are similar in that they enable the separation of specification and implementation. Multiple independent authors can produce implementations that are compatible with an interface specification. Similarly, a delegate specifies the signature of a method, and authors can write methods that are compatible with the delegate specification. When should you use interfaces, and when should you use delegates?

Delegates are useful when,

A single method is being called.
A class may want to have multiple implementations of the method specification.
It is desirable to allow using a static method to implement the specification.
An event-like design pattern is desired.
The caller has no need to know or obtain the object that the method is defined on.
The provider of the implementation wants to "hand out" the implementation of the specification to only a few select components.
Easy composition is desired.

Interfaces are useful when,

The specification defines a set of related methods that will be called.
A class typically implements the specification only once.

The caller of the interface wants to cast to or from the interface type to obtain other interfaces or classes.