Using @classmethod for Multiple Class Constructors

00:00 Using @classmethod for Multiple Constructors. A powerful technique for providing multiple constructors in Python is to use the @classmethod decorator.

00:12 This allows you to turn a regular method into a class method. Unlike a regular method, a class method doesn’t take the current instance, self, as an argument.

00:23 Instead, it takes the class itself, which is commonly passed in as the cls argument. Using cls to name this argument is a popular convention in the Python community.

00:36 On-screen, you can see the basic syntax to define a class method.

00:43 DemoClass defines a class method using Python’s built-in @classmethod decorator. The first argument of .class_method() holds the class itself.

00:53 Through this argument, you can access the class from inside itself. In this example, you access the .__name__ attribute, which stores the name of the underlying class as a string.

01:04 It’s important to note that you can access a class method using either the class or a concrete instance of the class at hand.

01:17 No matter how you invoke .class_method(), it will receive DemoClass as its first argument.

01:26 The ultimate reason why you can use class methods as constructors is that you don’t need an instance to call a class method. Using the @classmethod decorator makes it possible to add as many explicit constructors as you need to a given class.

01:40 The @classmethod decorator returns the instance of the class. It’s a Pythonic and popular way to implement multiple constructors, and you can also call this type of constructor an alternative constructor.

01:55 Now, how can you use a class method to customize Python’s instantiation process? Instead of fine-tuning .__init__() and the object initialization, you’ll control both steps: object creation and initialization. Through the following examples, you’ll learn how to do just that, starting with an example where you construct a circle from its diameter.