When I introduced objects previously in the A proper introduction section of the chapter, I said that we use them to represent real-life objects. For example, we sell goods of any kind on the web nowadays and we need to be able to handle, store, and represent them properly. But objects are actually so much more than that. Most of what you will ever do, in Python, has to do with manipulating objects.

So, without going into too much detail (we'll do that in Chapter 6, OOP, Decorators, and Iterators), I want to give you the in a nutshell kind of explanation about classes and objects.

We've already seen that objects are Python's abstraction for data. In fact, everything in Python is an object, infact numbers, strings (data structures that hold text), containers, collections, even functions. You can think of them as if they were boxes with at least three features: an ID (unique), a type, and a value.

But how do they come to life? How do we create them? How do we write our own custom objects? The answer lies in one simple word: classes.

Objects are, in fact, instances of classes. The beauty of Python is that classes are objects themselves, but let's not go down this road. It leads to one of the most advanced concepts of this language: metaclasses. For now, the best way for you to get the difference between classes and objects is by means of an example.

Say a friend tells you, I bought a new bike! You immediately understand what she's talking about. Have you seen the bike? No. Do you know what color it is? Nope. The brand? Nope. Do you know anything about it? Nope. But at the same time, you know everything you need in order to understand what your friend meant when she told you she bought a new bike. You know that a bike has two wheels attached to a frame, a saddle, pedals, handlebars, brakes, and so on. In other words, even if you haven't seen the bike itself, you know the concept of bike. An abstract set of features and characteristics that together form something called bike.

In computer programming, that is called a class. It's that simple. Classes are used to create objects. In fact, objects are said to be instances of classes.

In other words, we all know what a bike is; we know the class. But then I have my own bike, which is an instance of the bike class. And my bike is an object with its own characteristics and methods. You have your own bike. Same class, but different instance. Every bike ever created in the world is an instance of the bike class.

Let's see an example. We will write a class that defines a bike and then we'll create two bikes, one red and one blue. I'll keep the code very simple, but don't fret if you don't understand everything about it; all you need to care about at this moment is to understand the difference between a class and an object (or instance of a class):

# bike.py
# let's define the class Bike
class Bike:

    def __init__(self, colour, frame_material):
        self.colour = colour
        self.frame_material = frame_material

    def brake(self):
        print("Braking!")

# let's create a couple of instances
red_bike = Bike('Red', 'Carbon fiber')
blue_bike = Bike('Blue', 'Steel')

# let's inspect the objects we have, instances of the Bike class.
print(red_bike.colour)  # prints: Red
print(red_bike.frame_material)  # prints: Carbon fiber
print(blue_bike.colour)  # prints: Blue
print(blue_bike.frame_material)  # prints: Steel

# let's brake!
red_bike.brake()  # prints: Braking!

So many interesting things to notice here. First things first; the definition of a class happens with the class statement. Whatever code comes after the class statement, and is indented, is called the body of the class. In our case, the last line that belongs to the class definition is the print("Braking!") one.

After having defined the class, we're ready to create instances. You can see that the class body hosts the definition of two methods. A method is basically (and simplistically) a function that belongs to a class.

The first method, __init__, is an initializer. It uses some Python magic to set up the objects with the values we pass when we create it.

The other method we defined, brake, is just an example of an additional method that we could call if we wanted to brake the bike. It contains just a print statement, of course; it's an example.

We created two bikes then. One has red color and a carbon fiber frame, and the other one has blue color and a steel frame. We pass those values upon creation. After creation, we print out the color property and frame type of the red bike, and the frame type of the blue one just as an example. We also call the brake method of the red_bike.

One last thing to notice. You remember I told you that the set of attributes of an object is considered to be a namespace? I hope it's clearer what I meant now. You see that by getting to the frame_type property through different namespaces (red_bike, blue_bike), we obtain different values. No overlapping, no confusion.

The dot (.) operator is of course the means we use to walk into a namespace, in the case of objects as well.