Type annotation in Python

Principle

In Python, you can annotate your code with type annotation. For instance, we can say that variable a is an integer:

a: int = 5

In Python, type checking is decorrelated from the execution. So the following program works:

a: str = 5

You are free to leave variable unanotated.

a: str = "aze"
b = 1

Gradual typing

• C, Java, have static typing.
• Python, Javascript have dynamic typing.
• Python has type anotations. You can annotate only a part of your program. This is called gradual typing. It leverages desirable aspects of both dynamic and static typing.

Javascript and Typescript

The situation is a bit close to Typescript which is Javascript with type annotations.

Declaring constants

A variable is… by definition variable. You can change its value (note that you can change it in Python even if its value is immutable).

To make a constant, we can use the type Final[...].

from typing import Final
a: Final[int] = 2
a = 1 # wrong

from typing import Final
lst: Final[list] = [2]
lst.append(3) # ok
lst = [3] # wrong

This is close to:

• const in C, Javascript
• final in Java
• variable without the keyword mut in Rust

Any

Any is the type of anything in Python.

A: any = 2

• In Typescript, it is also any
• In Pascal, we had Variant but it is a bit different.

Parameterized types

A list of what? A tuple of what? A list of integers is list[int].

L: list[int] = [1, 2, 3]

D: dict[str, int] = {"aze": 2, "a": 1}

Veryfing types

Just run:

mypy mypythonprogram.py