Python Morsels: Classes are Everywhere

Related article:

• Is it a class or a function? It's a callable!
• Classes
• Variables Are Pointers

Transcript:

Classes are everywhere in Python. Even if you never define your own type of class, you will certainly use classes in Python.

Using a class object

We've imported class Point here, that we're calling. When you call a class, you get back an instance of that class:

>>> from point import Point
>>> p = Point(1, 2, 3)

We can access the x attribute on this Point object (aka Point instance) or call the is_origin method on it:

>>> p.x
1
>>> p.is_origin()
False

Python data structures are classes

When we check the type of p (which is a Point instance) it will tell us I am an object of type "class Point":

>>> type(p)
<class 'point.Point'>

The type of something is the same as the class of that thing. So class instances (aka objects of some class/type) can have attributes and they can have methods.

We've seen attributes and methods in other places in Python too. For example, lists have an append method:

>>> numbers = [1, 2, 3]
>>> numbers.append(4)
>>> numbers
[1, 2, 3, 4]

If you ask lists for their type, you'll see some kind of class. You always get a class when you ask for the type of something in Python:

>>> type(numbers)
<class 'list'>

The type of a list is the list class. So list is not a function in Python, it's a class. When you call list with parentheses after it, you get an empty list back because we're constructing a new object of type list:

>>> list()
[]

All the built-in data structures in Python are classes:

>>> tuple
<class 'tuple'>
>>> dict
<class 'dict'>
>>> set
<class 'set'>

Everything that has a "type" has a "class"

Interestingly, some of the even more fundamental types of objects are actually classes.

Anything that has a type has a class.

When you ask for the type of a string you'll see str:

>>> name = "Trey"
>>> type(name)
<class 'str'>

We may have thought of str as a function that can convert something to a string:

>>> str(4)
'4'

But it's actually a class!

int is also a class:

>>> x = 4
>>> type(x)
<class 'int'>

in fact, there are even attributes we can access on instances of the int class:

>>> x.real
4
>>> x
4

We're accessing the real attribute of object to get the real component of this number, which is the same as the number itself.

So, Integers are a class, which means if we call the int function, we're actually calling a class to get back an integer instance:

>>> int('4')
4

Built-in functions are class

A lot of the built-in functions in Python are actually classes too.

For example, when we call the enumerate function, we'll get back an enumerate object:

>>> enumerate(numbers)
<enumerate object at 0x7f58e8228c00>

The enumerate "function" is actually a class:

>>> enumerate
<class 'enumerate'>

It's not incorrect to say "the enumerate function" even though enumerate is implemented as a class because the distinction between classes and functions isn't very important in Python.

Summary

The distinction between classes and functions doesn't really matter that much in Python. The thing that I care about is the behavior: I care that I can call it.

When you call a function, we'll get the return value of that function. When you call a class, we'll get an instance of that class. Both functions and classes are callables and that's the thing we usually care about ("can I call it?").

So classes are everywhere in Python! Even if you never make your own class, you will certainly use classes.

from Planet Python
via read more