Object Oriented Python

Python is an object oriented language. Objects go to the heart of the language in the sense that all of the built in container types (list, tuple, dict string) are objects and all of the modules in the standard library use objects and present object based interfaces. If you write any non-trivial Python program you will be using objects. Unlike Java however, you can write Python programs without using an object oriented style. Python programs can be purely procedural and just make use of the built in objects. This flexibility supports different programming styles with Python, from quick and dirty scripts to solve a simple problem to fully engineered applications.

In this course, you won't need to write programs with objects except in a few cases - notably when you are writing and using unit tests. This chapter summarises the main features of objects in Python so that you can understand programs if you see them and begin writing object based code if you wish to do so. It isn't an introduction to object oriented programming - but if you've already written in that style, then you should be able to pick up enough information here to write this way in Python.

Python Classes

The class definition in Python defines the content and behaviour of an object in terms of member variables (to hold data) and methods (to define behaviour). In an application program, we would make an instance of a class which we could use to call the methods. An example of creating and using an instance can be seen in this string example:

>>> l = list()
>>> l.append(3)
>>> l.append(4)
>>> print(l)
[3, 4]
>>> l.reverse()
>>> l
[4, 3]

Here I've created an instance of the built in list class, which gives me an empty list, and then called the append method to add two integers to the list. I print the list and finally call the reverse method to reverse the elements of the list.

Most classes that we deal with will be like the list class in that they will hold some data and present an interface to updating or accessing the data.

Here is a simple class definition that might represent an on/off switch:

class Switch:

    # switch state, True == on, False == off
    state = False

    def on(self):
        """Turn the switch on"""
        self.state = True     

    def off(self):
        """Turn the switch off"""
        self.state = False

    def toggle(self):
        """Turn on if we're off, off if we're on"""

        if self.state:
            self.off()
        else:
            self.on()

In the example, the class definition is introduced with the keyword class and the name of the class. By convention, class names are capitalised but this is not a requirement of the language. The class definition is a code block and so is introduced by a colon at the end of the class statement and indented by four spaces. In the body of the class definition we have two kinds of things: data attributes (state in the example) define variables to hold the data belonging to the class instances and methods (on, off and status) define the operations that are supported on instances of the class.

The method definitions in the example are really just the same as the procedure definitions that we've already seen, except that they have a special argument called self. This argument is required for any method and its value is the instance of the class on which the method is being called. It is equivalent to the this variable in C++ and Java, except that it must be explicitly included as a parameter where those languages define it automatically for any method. Inside the body of the methods we refer to the data attribute via the self variable using the dot notation, as in self.state. Similarly, we call methods on the instance in the same way, as in the definition of the toggle method which calls self.on() or self.off().

To create an instance of this new class we use the name of the class like a procedure, in this case with no arguments. We can then call the methods of the class. Here's an example:

>>> sw = Switch()
>>> print(sw.state)
False
>>> sw.on()
>>> print(sw.state)
True
>>> sw.toggle()
>>> print(sw.state)
False

In this example we create a new Switch instance and assign it to the variable sw. We then print the value of the state attribute for this instance and call the on and toggle methods. The output will show that the state of the switch is changed by the methods (try it for yourself to confirm this).

The __init__ Method

In the Switch example, there was no initial information needed to initialise the instance. The starting state was just defined as False for all new instances. It is possible to define an explicit constructor for a class that may take additional arguments. The name of the constructor in Python is __init__ and it is defined like any other method. To extend our example, here is a version with a constructor that takes an initial state argument:

class Switch:

    def __init__(self, state=False):
        """Initialise the switch"""

        self.state = state

    def on(self):
        """Turn the switch on"""
        self.state = True     

    def off(self):
        """Turn the switch off"""
        self.state = False

    def toggle(self):
        """Turn on if we're off, off if we're on"""

        if self.state:
            self.off()
        else:
            self.on()

Note the differences here. The data attribute state is not explicitly initialised in the class body, instead it is set in the constructor to have the value of the state parameter (which defaults to False if not provided). Any method can create new data instances like this but it is most common to do so in the constructor. We can now create instances of Switch in one of two ways:

sw1 = Switch()      # initial state will default to False
sw2 = Switch(True)  # initial state set explicitly to True  

This example also shows how additional parameters are declared for methods, they are just added after the self parameter. Note that when we call the method (or the constructor) we provide one less parameter than in the method declaration - the initial self parameter is added by the Python runtime.

__init__ is one example of a special method that has a defined meaning in Python. There are others, all beginning and ending with two underscores, that define special behaviour. An example is the __str__ method which is called whenever Python wants a string representation of the instance. We won't go into these here but you can follow this up in the Python documentation.

\ Python Web Programming by Steve Cassidy is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.