Velamuvi AIAcademy

Lesson 2 of 15 · Classes and Objects

Defining a Class

Write your first Python class step by step: the class keyword, creating objects, giving them attributes, adding methods and understanding self.

  • 9 min read
  • Intermediate
  • Reviewed
In this lesson you will learn
  • How to write a class with the class keyword
  • How to create objects from a class
  • How to store data on an object
  • How to add a method and what self means

The smallest possible class

A class starts with the keyword class, then a name, then a colon. Everything that belongs to the class is indented underneath, just like the body of a function or a loop.

Try it yourselfPython
class Dog:
    pass


rex = Dog()
bella = Dog()

print(rex)
print(type(rex))

pass is a keyword that means "do nothing". Python needs at least one line inside a class, and pass is a placeholder.

Two important things happened:

  • class Dog: created the blueprint. On its own, it does nothing.
  • Dog() creates an object from the blueprint.

The first print shows something like <__main__.Dog object at 0x7f...>. That is Python's default way of describing an object: its class and its address in the computer's memory. You will learn to replace it with something friendlier later in the course.

Naming classes

Examples: Dog, BankAccount, ChatMessage. This makes classes easy to tell apart from variables and functions, which use snake_case.

Every object is separate

rex and bella were made from the same class, but they are two different objects.

Try it yourselfPython
class Dog:
    pass


rex = Dog()
bella = Dog()

print(rex is bella)
print(rex is rex)

is checks whether two names point to the same object. They do not: each call to Dog() builds a brand-new one, just as each press of a cookie cutter makes a new cookie.

Giving objects data

You can attach data to an object with a dot, like sticking a label on it: object.name = value. Remember, a piece of data stored on an object is called an attribute.

Try it yourselfPython
class Dog:
    pass


rex = Dog()
rex.name = "Rex"
rex.age = 3

bella = Dog()
bella.name = "Bella"
bella.age = 5

print(rex.name, rex.age)
print(bella.name, bella.age)

Each dog keeps its own attributes. Changing rex.age has no effect on bella.

Setting attributes from outside like this is fine for learning, but it is error-prone: nothing guarantees every dog gets a name. In the next lesson you will learn the proper way, where the class itself sets up every new object.

Adding a method

A method is a function written inside a class. It is an action the class's objects can perform.

Try it yourselfPython
class Dog:
    def bark(self):
        print("Woof!")


rex = Dog()
rex.bark()

You call a method with a dot, on an object: rex.bark().

What is self?

Look at the method again: def bark(self):. When you call rex.bark(), Python automatically passes rex in as self. You never pass it yourself.

That is how a method can use the object's own data:

Try it yourselfPython
class Dog:
    def bark(self):
        print(f"{self.name} says Woof!")


rex = Dog()
rex.name = "Rex"

bella = Dog()
bella.name = "Bella"

rex.bark()
bella.bark()

When rex.bark() runs, self is rex, so self.name is "Rex". When bella.bark() runs, self is bella. One method, written once, works for every dog.

A helpful way to picture it: rex.bark() is a shortcut for Dog.bark(rex). Both lines do exactly the same thing.

Try it yourselfPython
class Dog:
    def bark(self):
        print(f"{self.name} says Woof!")


rex = Dog()
rex.name = "Rex"

rex.bark()
Dog.bark(rex)

You will always write the short form, but knowing the long form explains where self comes from.

A method with extra parameters

Methods can take more parameters after self. You pass only those when you call the method. Python fills in self for you.

Try it yourselfPython
class Dog:
    def describe(self):
        return f"{self.name} is {self.age} years old"

    def birthday(self):
        self.age = self.age + 1

    def fetch(self, item, times):
        for _ in range(times):
            print(f"{self.name} brings back the {item}")


rex = Dog()
rex.name = "Rex"
rex.age = 3

rex.fetch("ball", 2)
rex.birthday()
print(rex.describe())

Notice three different kinds of method:

  • describe() returns information about the object.
  • birthday() changes the object's data.
  • fetch() takes extra arguments: rex.fetch("ball", 2) passes "ball" as item and 2 as times.

Common mistakes

Practice exercise

Write your first class, with one method that changes the object and one that reports on it.

Your solutionPython
# 1. Write a class called Lamp with two methods:
#    switch_on(self)  sets self.is_on to True
#    status(self)     returns "on" or "off" depending on self.is_on
# 2. The code below must then print exactly:
# Desk lamp is off
# Desk lamp is on


lamp = Lamp()
lamp.name = "Desk lamp"
lamp.is_on = False
print(lamp.name, "is", lamp.status())
lamp.switch_on()
print(lamp.name, "is", lamp.status())
Knowledge check

When you call rex.bark(), what is self inside the bark method?

Summary

  • class Name: defines a class. Class names use PascalCase.
  • Name() creates a new, separate object from the class.
  • object.attribute = value stores data on an object.
  • A method is a function inside a class. Its first parameter is always self.
  • self is the object the method was called on, so self.name reads that object's data.

Next: the special method that sets up every new object automatically.

Finished reading? Tick it off to track your progress.