Object-Oriented Python

After this lesson, you will be able to: Write Python classes with inheritance, dunder methods, properties, and dataclasses.

Python is multi-paradigm; you don't need OOP for everything. But when modeling real domains, classes + dataclasses are the cleanest path.

Prerequisites:Python Fundamentals

Classes the modern way

Dataclasses cut 80% of boilerplate.

python

from dataclasses import dataclass, field
from typing import Optional

@dataclass
class User:
    name: str
    email: str
    age: int = 0
    tags: list[str] = field(default_factory=list)  # mutable default trick

    def adult(self) -> bool:
        return self.age >= 18

u = User(name="Alex", email="[email protected]", age=30, tags=["admin"])
print(u)                # auto __repr__ via @dataclass
print(u == User("Alex", "[email protected]", 30, ["admin"]))  # auto __eq__

# Frozen (immutable)
@dataclass(frozen=True)
class Point:
    x: float
    y: float

Dunder methods (the magic methods)

__init__: constructor. __repr__: debug str. __str__: human str. __eq__: equality. __hash__: for set/dict keys. __len__: enables `len(obj)`. __iter__/__next__: iteration. __enter__/__exit__: context manager (`with obj:`). __call__: makes instance callable.

Properties + inheritance

Use properties for computed values; inheritance sparingly.

python

class Animal:
    def __init__(self, name: str): self._name = name

    @property
    def name(self) -> str: return self._name

    @name.setter
    def name(self, value: str):
        if not value.strip(): raise ValueError("empty")
        self._name = value

    def speak(self) -> str: return "..."

class Dog(Animal):
    def speak(self) -> str: return "Woof"

class Cat(Animal):
    def speak(self) -> str: return "Meow"

# Polymorphism
for pet in [Dog("Rex"), Cat("Mimi")]:
    print(f"{pet.name}: {pet.speak()}")

Common mistakes only experienced Python devs avoid

Reaching for inheritance when composition would do. 'Has-a' beats 'is-a' more often than not. Skipping @dataclass. Hand-writing __init__ + __repr__ + __eq__ is yesterday's pattern. Storing _private attributes without good reason. Python has no real privacy; convention only. Forgetting `field(default_factory=list)` for mutable dataclass defaults.

Quick Check

Why use @dataclass over a hand-written class?

Pick the cleanest answer.

←Python Fundamentals (Modern Python 3.12+)

Back to Python (Professional)

Pythonic Code→