Different ways to declare internal class variables in Python. Which is the best one?

Question

The following code includes several different ways of declaring a static variable used by a class. Are there any functional differences between them? Are there any advantages/disadvantages for each one? Is there a better way that I don't know about?

# 1st way
class ApplePie:

    type = "apple"

    def __init__(self):
        print(f"I'm an {ApplePie.type} pie!")



# 2nd way
class ApplePie:

    @property
    def type(self) -> str:
        return "apple"

    def __init__(self):
        print(f"I'm an {self.type} pie!")


# 3rd way
from functools import cached_property

class ApplePie:

    @cached_property
    def type(self) -> str:
        return "apple"

    def __init__(self):
        print(f"I'm an {self.type} pie!")

Which method would you guys use and why?

Answer 1

Your first example doesn't work. You can define a static variable like this:

class ApplePie:

    type = "apple"

    def __init__(self):
        print(f"I'm an {ApplePie.type} pie!")

This is a class property (i.e. it's shared among all instances of this class), as opposed to an instance property, accessed through self, like in your second example. These can differ among multiple instances of a class.

Your third example is

Useful for expensive computed properties of instances that are otherwise effectively immutable.

as described in the official documentation.