What are the differences between namedtuples and dataclasses in Python?

What are the differences between namedtuples and dataclasses in Python?

Namedtuples are immutable and provide a lightweight way to create tuple-like objects with named fields. Dataclasses, introduced in Python 3.7, are mutable and offer more features like default values, type annotations, and methods.

Can you give me an example of how to use both?

Sure! Here's a namedtuple example:

from collections import namedtuple
Point = namedtuple('Point', ['x', 'y'])
point = Point(1, 2)

And a dataclass example:

from dataclasses import dataclass
@dataclass
class Point:
    x: int
    y: int
point = Point(1, 2)

Which one should I use for my project?

It depends on your needs. If you need immutability and a simple structure, use namedtuples. If you need mutability and more features, go for dataclasses.

Are there performance differences between them?

Yes, namedtuples are generally faster due to their simplicity, but dataclasses offer more flexibility and features, which might be worth the trade-off.