What does the @dataclass decorator give you?

Short answer

It automatically generates __init__, __repr__ and __eq__ (and optionally ordering, hash and freezing) from the fields declared with type annotations.

Explanation

dataclass reduces boilerplate for data-holding classes. The key options are frozen=True to prevent modification and make the object hashable, slots=True to reduce memory, kw_only=True to force keyword arguments, and order=True to generate comparison operations. Use field(default_factory=list) for mutable defaults, and put validation in __post_init__. Note that dataclass does not check types at runtime.

Code example

from dataclasses import dataclass, field

@dataclass(frozen=True, slots=True)
class Order:
    id: int
    items: tuple[str, ...] = ()

@dataclass
class Cart:
    lines: list[str] = field(default_factory=list)

    def __post_init__(self):
        if len(self.lines) > 100:
            raise ValueError("too many lines")

Common mistakes

• Writing lines: list = [] directly, which dataclass rejects with a ValueError.
• Assuming dataclass validates types at runtime like Pydantic does.

Follow-up questions

• What is the relationship between __eq__ and __hash__ in Python?

Sources

• Python documentation — dataclasses