Question

In this part of the C# tutorial, we will talk about some of the new features of the C# version 3.0. Microsoft introduced this version of the C# in August 2007. Mono C# compiler fully supported the new version by July 2008.

The C# 3.0 version brought the following new features to the language:

• Implicitly typed variables
• Implicitly typed arrays
• Object & collection initializers
• Automatic properties
• Anonymous types
• Extension methods
• Query expressions
• Lambda expressions
• Expression trees

Query expressions, lambda expressions and expression trees are beyond the scope of this tutorial. They are closely connected to the LINQ. LINQ (Language Integrated Query) is a Microsoft .NET Framework component that adds native data querying capabilities to .NET languages.

Implicitly typed local variables & arrays

Both implicitly typed local variables & arrays are connected with the var keyword. It is an implicit data type. In some cases, we do not have to specify the type for a variable. When the usage of the var keyword is allowed, the compiler will find and use the type for us. The process is called type inference.

In some cases, the var keyword is not allowed. It can be used only on a local variable. It cannot be applied on a field in a class scope. It must be declared and initialized in one statement. The variable cannot be initialized to null .

using System;

public class CSharpApp
{
  static void Main()
  {
    int x = 34;
    var y = 32.3f;
    
    var name = "Jane";
 
    Console.WriteLine(x);
    Console.WriteLine(y.GetType());
    Console.WriteLine(name.GetType());
  }
}

We have a small example, where we use the var type.

int x = 34;
var y = 32.3f;

The first variable is explicitly typed, the second variable is implicitly typed. The compiler will look at the right side of the assignment and infer the type of the variable.

Console.WriteLine(y.GetType());
Console.WriteLine(name.GetType());

These two lines will check the type of the two variables.

$ ./itlv.exe 
34
System.Single
System.String

As we can see, the two variables use the familiar, built-in data types.

Implicitly typed arrays are arrays, in which the type of the array is inferred from the elements of the array in the array initializer by the compiler. The rules are the same as for implicitly typed local variables. Implicitly typed arrays are mostly used in query expressions together with anonymous types and object and collection initializers.

using System;

public class CSharpApp
{
  static void Main()
  {
    var items = new[] { "C#", "F#", "Python", "C" };

    foreach (var item in items)
    {
      Console.WriteLine(item);
    }
  }
}

An example demonstrating the implicitly typed arrays.

var items = new[] { "C#", "F#", "Python", "C" };

We again use the var keyword. The square brackets on the left side are omitted.

foreach (var item in items)
{
  Console.WriteLine(item);
}

The foreach loop is used to traverse the array. Note the use of the local implicitly typed item variable.

$ ./ita.exe 
C#
F#
Python
C

This is the output of the example.

Object initializers

Object initializers give a new syntax for creating and initiating objects. Inside a pair of curly brackets {} we initiate members of a class through a series of assignments. They are separated by comma character.

using System;

public class Person 
{
  private string _name; 
  private int _age;

  public string Name
  {
    get { return _name; }
    set { _name = value;} 
  }

  public int Age
  {
    get { return _age; }
    set { _age = value;} 
  }

  public override string ToString()
  {
    return String.Format("{0} is {1} years old", _name, _age);
  }
}

public class ObjectInitializers
{
  static void Main()
  {
    Person p1 = new Person();
    p1.Name = "Jane";
    p1.Age = 17;

    Person p2 = new Person { Name="Becky", Age=18 };

    Console.WriteLine(p1);
    Console.WriteLine(p2);
  }
}

In the above example, we have a Person class with two properties. We create two instances of this class. We use the old way and we also use the object initializer expression.

public string Name
{
  get { return _name; }
  set { _name = value;} 
}

This is a Name property with set a get accessors.

Person p1 = new Person();
p1.Name = "Jane";
p1.Age = 17;

This is the old way of creating an object and initiating it with values using the field notation.

Person p2 = new Person { Name="Becky", Age=18 };

This is the object initializer. Inside curly brackets, the two members are initiated.

Collection initializers

Collection initializers are a way of initiating collections where the elements of a collection are specified inside curly brackets.

List <string> planets = new List<string>();

planets.Add("Mercury");
planets.Add("Venus");
planets.Add("Earth");
planets.Add("Mars");
planets.Add("Jupiter");
planets.Add("Saturn");
planets.Add("Uranus");
planets.Add("Neptune");

This is the classic way of initiating a collection. In the following example, we will use a collection initializer for the same generic list.

using System;
using System.Collections.Generic;

public class CollectionInitializers
{
  static void Main()
  {
    List <string> planets = new List <string> 
      {"Mercury", "Venus", "Earth", "Mars", "Jupiter",
        "Saturn", "Uranus", "Neptune"};


    foreach (string planet in planets)
    {
      Console.WriteLine(planet);
    }
  }
}

We create a generic list of planets.

List <string> planets = new List <string> 
  {"Mercury", "Venus", "Earth", "Mars", "Jupiter",
    "Saturn", "Uranus", "Neptune"};

This is the collection initializer expression. The planet names are specified between the curly brackets. We save a some typing. We do not need to call the Add() method for each item of the collection.

$ ./collectioninitializers.exe 
Mercury
Venus
Earth
Mars
Jupiter
Saturn
Uranus
Neptune

Here we can see the output of the collectioninitializers.exe program.

Automatic properties

In a software project, there are lots of simple properties that only set or get some simple values. To simplify programming and to make the code shorter, automatic properties were created. Note that we cannot use automatic properties in all cases. Only for the simple ones.

using System;

public class Person 
{
  public string Name { get; set; }
  public int Age { get; set; }
}

public class AutomaticProperties
{
  static void Main()
  {
    Person p = new Person();
    p.Name = "Jane";
    p.Age = 17;
    
    Console.WriteLine("{0} is {1} years old", 
      p.Name, p.Age);
  }
}

This code is much shorter. We have a person class in which we have two properties.

public string Name { get; set; }
public int Age { get; set; }

Here we have two automatic properties. There is no implementation of the accessors. And there are no member fields. The compiler will do the rest for us.

Person p = new Person();
p.Name = "Jane";
p.Age = 17;

Console.WriteLine("{0} is {1} years old", 
  p.Name, p.Age);

We normally use the properties as usual.

$ ./automatic.exe 
Jane is 17 years old

This is the output of the example.

Anonymous types

The Person class that we have used in the previous example is said to be a type. More precisely, it is a user defined type. The type has a name and we can explicitly create an instance of it by referring to its name. On the other hand, anonymous types are types that do not have a name. They are class types that consist of one or more public read-only properties. No other kinds of class members are allowed.

Anonymous types are allow data types to encapsulate a set of properties into a single object without having to first explicitly define a type. Anonymous types must be stored in variables declared with the var keyword, telling the C# compiler to use type inference for the variable.

Anonymous types are created with the new keyword and object initializer. The compiler will infer the types of the properties itself. It will give the type a name, but it is only used by the compiler; it is not available to the programmer.

An anonymous type is another syntactic sugar to reduce typing. Anonymous types are often used in LINQ expressions.

using System;

public class AnonymousType
{
  static void Main()
  {
    var p = new { Name="Jane", Age=17 };

    Console.WriteLine("{0} is {1} years old", p.Name, p.Age);
  }
}

This is an anonymous type example.

var p = new { Name="Jane", Age=17 };

An anonymous object initializer declares an anonymous type and returns an instance of that type. We use the var keyword because we do not know the type.

Console.WriteLine("{0} is {1} years old", p.Name, p.Age);

We access the properties created using the field access notation.

$ ./anonymoustype.exe 
Jane is 17 years old

This is the output of the example.

Extension methods

Developers often face situations in which they would like to extend an existing type, but it is not possible. For example, the class is sealed. The extension method is a workaround for such cases. Extension methods are a special kind of a static method, but they are called as if they were instance methods on the extended type. To create an extension method, we need a static class and a static method. When we call our extension method on a class, the compiler does some behind the scenes processing; it appears as if we have called the extension method on the object of a type.

It is also possible, and it was a common workaround in the past, to create special utility classes for such methods. These were mostly created as static methods of static classes. Both approaches have their advantages. It is also advised to use extension methods sparingly.

using System;

public static class Util
{
  public static string Reverse(this string input)
  {
    char[] chars = input.ToCharArray();

    Array.Reverse(chars);
    return new String(chars);
  }
}

public class ExtentionMethod
{
  static void Main()
  {
    string str1 = "Jane";
    string str2 = str1.Reverse();

    Console.WriteLine(str2);  
  }
}

In our case, we would like to add a new method to a string class. The string class is a built-in class and it is sealed. No inheritance is possible. This is why we need an extension method.

public static class Util
{
  public static string Reverse(this string input)
  {
    char[] chars = input.ToCharArray();

    Array.Reverse(chars);
    return new String(chars);
  }
}

We have a Reverse() extension method. This method reverses characters of a string variable. The method and its class must be static. The static Reverse() method becomes an extension method, when we put the this modifier as the first parameter of the method.

string str1 = "Jane";
string str2 = str1.Reverse();

We define and initialize a string variable. We call the Reverse() method on this variable. Even though the method is static, we use the instance method call syntax.

$ ./extentionmethods.exe 
enaJ

This is the output of the extentionmethods.exe program.

In this part of the C# tutorial we have talked about new features of the C# 3.0.