What is an Object in Python?

Question

I am surprised that my question was not asked (worded like the above) before. I am hoping that someone could break down this basic term "object" in the context of a OOP language like Python. Explained in a way in which a beginner like myself would be able to grasp.

When I typed my question on Google, the first post that appears is found here.

This is the definition: An object is created using the constructor of the class. This object will then be called the instance of the class.

Wikipedia defines it like this: An object is an instance of a Class. Objects are an abstraction. They hold both data, and ways to manipulate the data. The data is usually not visible outside the object.

I am hoping someone could help to break down this vital concept for me, or kindly point me to more resources. Thanks!

Answer 1

Everything is an object

An object is a fundamental building block of an object-oriented language. Integers, strings, floating point numbers, even arrays and dictionaries, are all objects. More specifically, any single integer or any single string is an object. The number 12 is an object, the string "hello, world" is an object, a list is an object that can hold other objects, and so on. You've been using objects all along and may not even realize it.

Objects have types

Every object has a type, and that type defines what you can do with the object. For example, the int type defines what happens when you add something to an int, what happens when you try to convert it to a string, and so on.

Conceptually, if not literally, another word for type is class. When you define a class, you are in essence defining your own type. Just like 12 is an instance of an integer, and "hello world" is an instance of a string, you can create your own custom type and then create instances of that type. Each instance is an object.

Classes are just custom types

Most programs that go beyond just printing a string on the display need to manage something more than just numbers and strings. For example, you might be writing a program that manipulates pictures, like photoshop. Or, perhaps you're creating a competitor to iTunes and need to manipulate songs and collections of songs. Or maybe you are writing a program to manage recipes.

A single picture, a single song, or a single recipe are each an object of a particular type. The only difference is, instead of your object being a type provided by the language (eg: integers, strings, etc), it is something you define yourself.

Answer 2

To go deep, you need to understand the Python data model.

But if you want a glossy stackoverflow cheat sheet, let's start with a dictionary. (In order to avoid circular definitions, let's just agree that at a minimum, a dictionary is a mapping of keys to values. In this case, we can even say the keys are definitely strings.)

def some_method():
    return 'hello world'

some_dictionary = {
    "a_data_key": "a value",
    "a_method_key": some_method,
}

An object, then, is such a mapping, with some additional syntactic sugar that allows you to access the "keys" using dot notation.

Now, there's a lot more to it than that. (In fact, if you want to understand this beyond python, I recommend The Art of the Metaobject Protocol.) You have to follow up with "but what is an instance?" and "how can you do things like iterate on entries in a dictionary like that?" and "what's a type system"? Some of this is addressed in Skam's fine answer.

We can talk about the python dunder methods, and how they are basically a protocol to implementing native behaviors like sized (things with length), comparable types (x < y), iterable types, etc.

But since the question is basically PhD-level broad, I think I'll leave my answer terribly reductive and see if you want to constrain the question.