Creating a Simple Class

Design Class

The basic purpose of a class is to define an object that encapsulates data and the operations performed on that data. For example, BasicClass defines a property and two methods that operate on the data in that property:

Value — Property that contains the numeric data stored in an object of the class
roundOff — Method that rounds the value of the property to two decimal places
multiplyBy — Method that multiplies the value of the property by the specified number

Start a class definition with a classdef ClassName...end block, and then define the class properties and methods inside that block. Here is the definition of BasicClass:

classdef BasicClass
   properties
      Value {mustBeNumeric}
   end
   methods
      function r = roundOff(obj)
         r = round([obj.Value],2);
      end
      function r = multiplyBy(obj,n)
         r = [obj.Value]*n;
      end
   end
end

For a summary of class syntax, see classdef.

To use the class:

Save the class definition in a .m file with the same name as the class.
Create an object of the class.
Access the properties to assign data.
Call methods to perform operation on the data.

Create Object

Create an object of the class using the class name:

a = BasicClass

a = 

  BasicClass with properties:

    Value: []

Initially, the property value is empty.

Access Properties

Assign a value to the Value property using the object variable and a dot before the property name:

a.Value = pi/3;

To return a property value, use dot notation without the assignment:

a.Value

ans =

    1.0472

For information on class properties, see Property Syntax.

Call Methods

Call the roundOff method on object a:

roundOff(a)

ans =

    1.0500

Pass the object as the first argument to a method that takes multiple arguments, as in this call to the multiplyBy method:

multiplyBy(a,3)

ans =

    3.1416

You can also call a method using dot notation:

a.multiplyBy(3)

Passing the object as an explicit argument is not necessary when using dot notation. The notation uses the object to the left of the dot.

For information on class methods, see Method Syntax.

Add Constructor

Classes can define a special method to create objects of the class, called a constructor. Constructor methods enable you to pass arguments to the constructor, which you can assign as property values. The BasicClass Value property restricts its possible values using the mustBeNumeric function.

Here is a constructor for the BasicClass class. When you call the constructor with an input argument, it is assigned to the Value property. If you call the constructor without an input argument, the Value property has a default value of empty ([]).

methods        
    function obj = BasicClass(val)
        if nargin == 1
            obj.Value = val;
        end
    end
end

By adding this constructor to the class definition, you can create an object and set the property value in one step:

a = BasicClass(pi/3)

a = 

  BasicClass with properties:

    Value: 1.0472

The constructor can perform other operations related to creating objects of the class.

For information on constructors, see Class Constructor Methods.

Vectorize Methods

MATLAB^® enables you to vectorize operations. For example, you can add a number to a vector:

[1 2 3] + 2

ans =

     3     4     5

MATLAB adds the number 2 to each of the elements in the array [1 2 3]. To vectorize the arithmetic operator methods, enclose the obj.Value property reference in brackets.

[obj.Value] + 2

This syntax enables the method to work with arrays of objects. For example, create an object array using indexed assignment.

obj(1) = BasicClass(2.7984);
obj(2) = BasicClass(sin(pi/3));
obj(3) = BasicClass(7);

These two expressions are equivalent.

[obj.Value] + 2
[obj(1).Value obj(2).Value obj(3).Value] + 2

The roundOff method is vectorized because the property reference is enclosed in brackets.

r = round([obj.Value],2);

Because roundOff is vectorized, it can operate on arrays.

roundOff(obj)

ans =

    2.8000    0.8700    7.0000

Overload Functions

Classes can implement existing functionality, such as addition, by defining a method with the same name as the existing MATLAB function. For example, suppose that you want to add two BasicClass objects. It makes sense to add the values of the Value properties of each object.

Here is an overloaded version of the MATLAB plus function. It defines addition for the BasicClass class as adding the property values:

methods
   function r = plus(o1,o2)
      r = [o1.Value] + [o2.Value];
   end
end

By implementing a method called plus, you can use the “+” operator with objects of BasicClass.

a = BasicClass(pi/3);
b = BasicClass(pi/4);
a + b

ans =

    1.8326

By vectorizing the plus method, you can operate on object arrays.

a = BasicClass(pi/3);
b = BasicClass(pi/4);
c = BasicClass(pi/2);
ar = [a b];
ar + c

ans =

    2.6180    2.3562

Related Information

For information on overloading functions, see Overload Functions in Class Definitions.

For information on overloading operators, see Operator Overloading.

`BasicClass` Code Listing

Here is the BasicClass definition after adding the features discussed in this topic:

classdef BasicClass
    properties
        Value {mustBeNumeric}
    end
    methods
        function obj = BasicClass(val)
            if nargin == 1
                obj.Value = val;
            end
        end
        function r = roundOff(obj)
            r = round([obj.Value],2);
        end
        function r = multiplyBy(obj,n)
            r = [obj.Value] * n;
        end
        function r = plus(o1,o2)
            r = [o1.Value] + [o2.Value];
        end
    end
end