Python Type Hints Basics

Type Hints¶

Type Hints were first introduces in Python 3.5
Python’s type hints provide you with optional static typing to leverage the best of both static and dynamic typing.
With type hints you don't have to write doc strings
Until a popular package incorporates type hints, we can use Type Hinting Package Stubs (THPS).

Type hints do not affect the Python interpreter. Type Hints are optional.

You can put them in or leave them out of your code with no effect. As far as you are concerned, the Python interpreter ignores type hints.

Type checking tools¶

To check the syntax for type hints, we need to use a static type checker tool.
Some tools are:
- mypy (Dropbox)
- pytype (Google)
- pyre (Facebook)
- pyright (Microsoft)
- PyCharm (JetBrains)
Installing and using the tool: pip install mypy, mypy app.py

Basics¶

Syntax of type hints: <variable name> : <type name>
The basic types are int, float, str, bool, bytes, list, tuple, dict, set, frozenset, None.
- Some of them like dict, list and etc may have become a basic type after python 3.9.

def say_hi(name: str) -> str: 
    return f'Hi {name}'

Type Hinting Functions¶

Default values

def f(num1: int, my_float: float = 3.5) -> float:
    return num1 + my_float

If a function doesn’t explicitly returns a value, you can use None to type hint the return value.
```
def log(message: str) -> None:
    print(message)
```

Generators

# A generator function that yields ints is secretly just a function that
# returns an iterator of ints, so that's how we annotate it
def g(n: int) -> Iterator[int]:
    i = 0
    while i < n:
        yield i
        i += 1

Function annotation can be split over multiple lines

def send_email(address: Union[str, list[str]],
               sender: str,
               cc: Optional[list[str]],
               bcc: Optional[list[str]],
               subject='',
               body: Optional[list[str]] = None
               ) -> bool:

Fixing the type of the variable¶

name: str = 'Hello'
name = 100

""" 
error: Incompatible types in assignment (expression has type "int", variable has type "str")

Adding a type to a variable is unnecessary because static type checkers typically can infer the type based on the value assigned to the variable.

name = 'Hello' 
name = 100

This will give the same error as above
But the first approach is more descriptive and better IMO
"""

Multiple Types¶

To set type hints for multiple types, you can use Union from the typing module.
- from typing import Union

from typing import Union 

def add(x: Union[int, float], y: Union[int, float]) -> Union[int, float]: 
    return x + y

Starting from Python 3.9, you can use the X | Y syntax to create a union type.

def add(x: int | float, y: int | float) -> int | float:
    return x + y

Type Aliases¶

from typing import Union

number = Union[int, float]

def add(x: number, y: number) -> number:
    return x + y

Genome = list[int] # this is a type alias
Population: list[Genome] # this is a variable

Its tricky to decide when type hints should be used.
They should not be used everywhere since they decrease the readability of the code.
Use them when you have to repeat the same complex type over and over again or if the type is becoming too long.
- A good example would be using type aliases for callable functions.

Type Hinting Dictionaries and Lists¶

In newer versions of python we can use

myList: list[str] = ["hello", "there"]
myDict: dict[str, int] = {"1": 32, "3": 45}

In older versions of python you will have to use the typing module

from typing import Dict, List

dict_of_users: Dict[int,str] = {
    1: "Jerome",
    2: "Lewis"
}

list_of_users: List[str] = [
    "Jerome", "Lewis"
]

Type Hinting in Classes¶

class User:
    def __init__(self, name:str, address:"Address"):
        self.name = name
        self.address = address

class Address:
    def __init__(self, owner:User, address_line:str):        
        self.owner = owner
        self.address_line = address_line

We can define the data types of the attributes

class Test:
    name: str

    def __init__(self, name: str) -> None:
        self.name = name

In case of recursive data structures we have to enclose the type in quotes if they are of the same class.

from typing import Optional

class BinaryTree:
    leftNode: Optional["BinaryTree"] = None
    # leftNode: "BinaryTree" | None = None 
    # the above would have also worked
    rightNode: Optional["BinaryTree"] = None

In the above example Optional is needed because

Generics¶

Generics are useful when we don't know the the exact type but we know it will be the same

from typing import TypeVar

T = TypeVar("T")

def myFunction(lst: list[T], index: int) -> T:
    return lst[index]

myFunction([1, 2, 3], 2)

We could have passed it a list of strings or list of actually anything and it would have worked fine.

`Any` type¶

from typing import Any

def myFunction(a: Any) -> None:
    print(a)

myFunction(45)

`Optional` type¶

We use optional for types that could be None

from typing import Any, Optional

# this means our function can return either string or None
def myFunction(a: Any) -> Optional[str]:
    print(a)

myFunction("test")

We can also use optional for variables: x : Optional[int] = 34
It is badly named. It is not the optional type in functions. It is just a shortcut for Union between None and the type in between the brackets. In the above example it is int.

`Sequence` type¶

It is a list of strings, integers etc.
Anything that can be iterated over.

from typing import Sequence

# this is a sequence of unknown values
# we can also be more specific Sequence[int]
def test(lst: Sequence) -> None:
    for i in lst:
        print(i)

test([1, 2, 34])
test("hello world") # since string is a sequence of characters

Dictionaries and sets are not considered as sequence.

`Callable` type¶

from typing import Callable

def firstFunc(a: int, b: str) -> bool:
    print(a, b)
    return True

def secondFunc(func: Callable[[int, str], bool]) -> None:
    func(4, "te")

secondFunc(firstFunc)

References¶

Type hints cheat sheet (Python 3) - mypy 0.961 documentation

Last updated: 2022-10-23