Classes and Objects

Why this matters

So far your data and your functions have lived apart. A class bundles them: it describes a kind of thing, with its own data and its own abilities. Each thing made from it is an object (an instance). You have used objects all along: every list, string and dictionary is one.

The idea

class Dog:
    def __init__(self, name, age=1):   # runs when you create a Dog
        self.name = name               # attributes: data stored on this object
        self.age = age

rex = Dog("Rex", 3)                    # create an instance
fido = Dog("Fido")
print(rex.name, rex.age)               # Rex 3
print(fido.age)                        # 1
rex.age = 4                            # change only Rex

• __init__ is the constructor; it sets up a new object.
• self is the object being created or used. Python passes it in for you, but you must list it first.
• Class names use CamelCase; attributes use snake_case.

Try it

Create two Dog objects, change one’s age, and print both. Then print type(rex) and isinstance(rex, Dog).

Your turn

Write:

• class Point with attributes x and y, both defaulting to 0. Point() is the origin and Point(3, 4) has x == 3.
• class Book(title, author, pages) with those three attributes, plus pages_read starting at 0 and notes starting as a new empty list (every book needs its own list).
• make_library(rows): rows is a list of (title, author, pages) tuples. Return a list of Book objects, one per row, in order.

Recap

• A class describes a kind of object; ClassName(...) creates an instance.
• __init__(self, ...) sets up attributes with self.name = value.
• Each object keeps its own attribute values.

Common mistakes

• Forgetting self in the method definition: TypeError: __init__() takes 1 positional argument but 2 were given.
• Writing x = x instead of self.x = x, so nothing is stored (AttributeError later).
• A mutable default such as def __init__(self, notes=[]), which is shared by every object.