Python Classes
In Python, you use classes to describe objects. Think of a class as a tool you use to create your own data structures that contain information about something; you can then use functions (methods) to perform operations on the data you describe. A class models how something should be defined and represents an idea or a blueprint for creating objects in Python.
Creating a Class
Say that you want to create a class to describe a router. The first thing you have to do is define it. In Python, you define a class by using the class keyword, giving the class a name, and then closing with a colon. Pep8 (introduced in Chapter 3) recommends capitalizing a class name to differentiate it from a variable. Here is a simple example of creating a class in Python:
>>> class Router:
pass
This example uses pass as a sort of placeholder that allows the class to be defined and set up to be used as an object. To make the class more useful, you can add some attributes to it. In the case of a router, you typically have some values that you want to have when you instantiate the class. Every router has a model name, a software version, and an IP address for management. You also need to pass some values to get started. The first value is always self. The reason for this will become obvious when you instantiate the class: The self value passes the object name that you select to instantiate the class. In the following example, the object you will create is rtr1:
class Router:
'''Router Class'''
def __init__(self, model, swversion, ip_add):
'''initialize values'''
self.model = model
self.swversion = swversion
self.ip_add = ip_add
rtr1 = Router('iosV', '15.6.7', '10.10.10.1')
After defining the class, you add a docstring to document what the class is for and then you create a function that calls __init__, which is a special case that is used for the setup of the class. (In __init__, the double underscores are called dunder or magic methods.) Functions that are within the class are called methods and become actions that you can perform on the object you are creating. To store attributes, you map the name self and the values you pass to it become variables inside the object, which then store those values as attributes. The last bit of code instantiates the object itself. Up until now, you have been creating a template, and by assigning data to the variables within the class, you have been telling Python to build the object. Now you can access any of the stored attributes of the class by using dot notation, as shown here:
>>> rtr1.model 'iosV'
When you call rtr1.model, the interpreter displays the value assigned to the variable model within the object. You can also create more attributes that aren’t defined during initialization, as shown in this example:
>>> rtr1.desc = 'virtual router' >>> rtr1.desc 'virtual router'
This example shows how flexible objects are, but you typically want to define any attributes as part of a class to automate object creation instead of manually assigning values. When building a class, you can instantiate as many objects as you want by just providing a new variable and passing over some data. Here is another example of creating a second router object rtr2:
>>> rtr2= Router('isr4221', '16.9.5', '10.10.10.5')
>>> rtr2.model
'isr4221'
Methods
Attributes describe an object, and methods allow you to interact with an object. Methods are functions you define as part of a class. In the previous section, you created an object and applied some attributes to it. Example 4-1 shows how you can work with an object by using methods. A method that allows you to see the details hidden within an object without typing a bunch of commands over and over would be a useful method to add to a class. Building on the previous example, Example 4-1 adds a new function called getdesc() to format and print the key attributes of your router. Notice that you pass self to this function only, as self can access the attributes applied during initialization.
Example 4-1 Router Class Example
class Router:
'''Router Class'''
def __init__(self, model, swversion, ip_add):
'''initialize values'''
self.model = model
self.swversion = swversion
self.ip_add = ip_add
def getdesc(self):
'''return a formatted description of the router'''
desc = f'Router Model :{self.model}\n' f'Software Version :{self.swversion}\n' f'Router Management Address:{self.ip_add}'
return desc
rtr1 = Router('iosV', '15.6.7', '10.10.10.1')
rtr2 = Router('isr4221', '16.9.5', '10.10.10.5')
print('Rtr1\n', rtr1.getdesc(), '\n', sep='')
print('Rtr2\n', rtr2.getdesc(), sep='')
There are two routers instantiated in this example: rtr1 and rtr2. Using the print function, you can call the getdesc() method to return formatted text about the object’s attributes. The following output would be displayed:
Rtr1 Router Model :iosV Software Version :15.6.7 Router Management Address:10.10.10.1 Rtr2 Router Model :isr4221 Software Version :16.9.5 Router Management Address:10.10.10.5
Inheritance
Inheritance in Python classes allows a child class to take on attributes and methods of another class. In the previous section, Example 4-1 creates a class for routers, but what about switches? If you look at the Router class, you see that all of the attributes apply to a switch as well, so why not reuse the code already written for a new Switch class? The only part of Example 4-1 that wouldn’t work for a switch is the getdesc() method, which prints information about a router. When you use inheritance, you can replace methods and attributes that need to be different. To inherit in a class, you create the class as shown earlier in this chapter, but before the colon, you add parentheses that include the class from which you want to pull attributes and methods. It is important to note that the parent class must come before the child class in the Python code. Example 4-2 shows how this works, creating a second class named Switch, using the Router class as parent. In addition, it creates a different getdesc() method that prints text about a switch rather than about a router.
Example 4-2 Router Class and Switch Class with Inheritance
class Router:
'''Router Class'''
def __init__(self, model, swversion, ip_add):
'''initialize values'''
self.model = model
self.swversion = swversion
self.ip_add = ip_add
def getdesc(self):
'''return a formatted description of the router'''
desc = (f'Router Model :{self.model}\n'
f'Software Version :{self.swversion}\n'
f'Router Management Address:{self.ip_add}')
return desc
class Switch(Router):
def getdesc(self):
'''return a formatted description of the switch'''
desc = (f'Switch Model :{self.model}\n'
f'Software Version :{self.swversion}\n'
f'Switch Management Address:{self.ip_add}')
return desc
rtr1 = Router('iosV', '15.6.7', '10.10.10.1')
rtr2 = Router('isr4221', '16.9.5', '10.10.10.5')
sw1 = Switch('Cat9300', '16.9.5', '10.10.10.8')
print('Rtr1\n', rtr1.getdesc(), '\n', sep='')
print('Rtr2\n', rtr2.getdesc(), '\n', sep='')
print('Sw1\n', sw1.getdesc(), '\n', sep='')
You can add another variable named sw1 and instantiate the Switch class just as you did the Router class, by passing in attributes. If you create another print statement using the newly created sw1 object, you see the output shown in Example 4-3.
Example 4-3 Code Results of Using Class Inheritance
Rtr1 Router Model :iosV Software Version :15.6.7 Router Management Address:10.10.10.1 Rtr2 Router Model :isr4221 Software Version :16.9.5 Router Management Address:10.10.10.5 Sw1 Switch Model :Cat9300 Software Version :16.9.5 Switch Management Address:10.10.10.8
To learn more about classes, methods, and inheritance, you can refer to the Python documentation. https://docs.python.org/3/tutorial/classes.html