DeepHierarchy

From APIDesign

Current revision

OOP encourages code reuse and may lead to deep subclassing hierarchies. Beware of that when designing API.

OOP Models the World

Classic object-oriented programming languages are said to be inspired by nature and its evolution. Everyone knows the example defining a Mammal class with basic methods describing the behavior of every mammal, and various subclasses such as Dog or Cat overriding the methods to do what cats and dogs do differently. This example not only demonstrates what object-oriented languages do best, but it also justifies the whole purpose of OOP. By showing how easy it is to describe nature, the object-oriented concept is also elevated to be the right programming technique to describe the real world. OOP languages are supposed to be more suitable than traditional ones for coding and capturing real-world concepts.

Just an Intricate Switch

After maintaining a framework in Java for more than fifteen years, I don’t buy this. Honestly, virtual method dispatch is nothing else than an intricate switch statement! A method call isn’t a direct invocation of some code but rather a switch that chooses from multiple implementations. The switch is based on the list of subclasses and can potentially grow as various developers create more and more subclasses. But it’s still a switch. Of course, explaining OOP as an enhanced switch is much less fancy than presenting it as a technology that helps us model the real world. That’s why computer courses are likely to stick with the Mammal example.

Switch with no Limits!

However, I’ll use the switch explanation to reveal an important hidden catch: when writing an API, the size and behavior of the switch is unknown to the writer of the code. This can be seen as an advantage, increasing the number of ways that a piece of code can be reused. However, a few releases later, if you don’t know the actual participants in the switch, you can have a nightmare on your hands. Simply allowing an unknown party to participate in your code “switch” can be too open to be maintainable.

Substitution Principle

When writing an API, the term subclass shouldn’t be used for someone who can participate in a switch, but for someone who behaves exactly like me, plus adds some additional behavior. True, many would agree that a Human is a subclass (a specialization) of a Mammal. Nice, but simply exposing a DeepHierarchy of classes is unlikely to improve an API’s usability magically.

Especially GUI toolkits (full of widgets inheriting from each other) got this all wrong. They like to claim that a JFrame is a specialization of JComponent, but try to use JFrame where you can use JComponent! It is going to fail. That is not a failure of the usage, but a failure of the API designer!

Do Not Expose Deep Hierarchies!

Rather than exposing DeepHierarchy separate APIvsSPI and stay as flat as possible! If you have to expose subtyping, remember Liskov's substitution principle: if something is a subtype, it has to inherently be ready to be used in place of the original super class or interface.