Course contentsShow
← Python
Lesson 645 of 1,06010. Advanced Language FeaturesFree lesson

Creating Read-Only Computed Properties

You will design descriptors that provide computed values while raising an AttributeError if someone attempts to modify the underlying attribute.

In this lesson, you will learn how to use the descriptor protocol to create attributes that are calculated on-the-fly and protected from being overwritten.

The Power of Read-Only Descriptors

In Python, a descriptor is an object that defines how attribute access is handled. While we often use __get__ to fetch data, we can leverage the __set__ method to enforce immutability. If a class defines __set__ and explicitly raises an AttributeError, it effectively prevents any attempts to modify that attribute from the outside.

A "computed" property is one that doesn't exist as a static value in the __dict__ of an object, but is instead generated whenever it is accessed. By combining this logic with an error-raising setter, we create a robust, read-only interface.

Implementing the Read-Only Descriptor

Consider a scenario where you want a Circle object to provide an area that is calculated based on its radius, but you don't want anyone to manually "set" the area to a wrong value:

class ComputedArea:
    def __get__(self, instance, owner):
        if instance is None:
            return self
        return 3.14159 * (instance.radius ** 2)

    def __set__(self, instance, value):
        raise AttributeError("Cannot set read-only attribute 'area'")

class Circle:
    area = ComputedArea()

    def __init__(self, radius):
        self.radius = radius

# Usage
c = Circle(10)
print(c.area)  # Output: 314.159
c.area = 50    # Raises AttributeError: Cannot set read-only attribute 'area'

Why This Matters

By implementing __set__, the descriptor becomes a data descriptor. Even if a user tries to assign a value to c.area, the interpreter will call our __set__ method rather than adding a new entry to the instance's __dict__. This ensures that your calculated business logic remains the "source of truth" and cannot be tampered with by external code. This pattern provides a clean API while maintaining strict internal integrity.

Key Takeaway: By defining both __get__ and __set__ in a descriptor, you create a data descriptor that can compute values dynamically while enforcing immutability by explicitly raising an AttributeError within the setter.