Misplaced Pages

Revision as of 18:06, 15 September 2006

In object-oriented programming, a constructor (sometimes shorted to ctor) in a class is a special method (function) called when an oject is declared and never has a return type. Constructors are special instance methods that are called automatically upon the creation of an object (instance of a class). They are often distinguished by having the same name as the class of the object they're associated with. Its main purpose is to pre-define the object's data members and to establish the invariant of the class, failing if the invariant isn't valid. A properly written constructor will leave the object in a 'valid' state.

Java

Example

public class Example 
{ 
  //declaration of instance variable(s).
  protected int data;
  //definition of the constructor. 
  public Example()
  {
     data = 1;
     data = 2;
  }
}

REALbasic

Example

Constructors in REALbasic can be in one of two forms. Each form uses a regular method declaration with a special name (and no return value). The older form uses the same name as the Class itself, and the newer form uses the name "Constructor." The newer form is the preferred one because it makes refactoring your class easier.

Class Foobar
  // Old form
  Sub Foobar( someParam as String )
  End Sub
  // New form
  Sub Constructor()
  End Sub
End Class

C# (C Sharp)

Example

class myClass
{
  public myClass(int a, string b)
  {
    a = 1;
    b = 2;
  }
}

Constructors Simplified (with pseudocode)

Constructors are always part of the implementation of classes. A class (in programming) refers to a specfication of the general characteristics of the set of objects that are members of the class rather than the specific characteristics of any object at all. A simple analogy follows. Consider the set (or class, using its generic meaning) of students at some school. Thus we have

class Student {
	// refers to the class of students
	// ... more omitted ...
}

However, the class Student is just a generic prototype of what a student should be. To use it, the programmer creates each student as an object or instance of the class. This object is a real quantity of data in memory whose size, layout, characteristics, and (to some extent) behavior are determined by the class definition. The usual way of creating objects is to call a constructor (classes may in general have many independent constructors). For example,

class Student {
	Student (String studentName, String sex, String Address, int ID) {
		// ... storage of input data and other internal fields here ...
	}
	// ... more omitted ...
}
// elsewhere:
Student john = Student("John Doe", "male", "under the sea", 42); 

It is up to the class to decide what to do with the information provided (storing it within the new object is very common, but it might be stored in an entirely different format, passed to other constructors or functions and then forgotten, or even disregarded altogether). The constructed object is persistent, and (depending on the class) may be augmented, examined, or modified in ways related or not to the constructor:

	print(john.studentName); // output "John Doe", the internal data
	print(john.getFirstName()); // object calculates just "John", which is printed
	john.addNickName("Johnny"); // object remembers this

Here it is assumed that the class provides the studentName variable and the getFirstName() and addNickName() functions, and that the variable is simply assigned the value of the first argument to the constructor. Note that this code does not explicitly mention the class Student; it is known that the variable john is an instance of the class, so the elements of the variable are understood to be members of the same class. This implicit specification is a powerful mnemonic, although it is not fundamentally more powerful than function overloading.

See also

• Destructor
• Object lifetime
• Initializer
• Copy constructor

References

• Bjarne Stroustrup: The C++ Programming Language, Addison-Wesley, ISBN 0-201-70073-5

• Programming constructs