Monad

About 3 min

Intent

To provide a mechanism for encapsulating computations or side effects, enabling the chaining of operations while managing context and data flow in a side-effect-free manner.

Explanation

Real-world example

Consider a real-world analogous example of a monad with a restaurant meal ordering process. In this scenario, each step of selecting a dish, adding sides, and choosing a drink can be thought of as a monadic operation. Each operation encapsulates the current state of the order (e.g., main dish chosen) and allows for the next choice (e.g., selecting a side) without exposing the complexity of the entire order's details to the customer.
Just like in a functional monad, if any step fails (like an unavailable dish), the entire process can be halted or redirected without throwing exceptions, maintaining a smooth flow. This encapsulation and chaining allow for a clean, error-managed progression from choosing the main dish to completing the full meal order, akin to how monads handle data and operations in functional programming. This approach ensures a consistent experience, where every choice builds on the previous one in a controlled manner.

In plain words

Monad pattern ensures that each operation is executed regardless of the success or failure of previous ones.

Wikipedia says

In functional programming, a monad is a structure that combines program fragments (functions) and wraps their return values in a type with additional computation. In addition to defining a wrapping monadic type, monads define two operators: one to wrap a value in the monad type, and another to compose together functions that output values of the monad type (these are known as monadic functions). General-purpose languages use monads to reduce boilerplate code needed for common operations (such as dealing with undefined values or fallible functions, or encapsulating bookkeeping code). Functional languages use monads to turn complicated sequences of functions into succinct pipelines that abstract away control flow, and side effects.

Programmatic Example

Here’s the Monad implementation in Java.

The Validator takes an object, validates it against specified predicates, and collects any validation errors. The validate method allows you to add validation steps, and the get method either returns the validated object or throws an IllegalStateException with a list of validation exceptions if any of the validation steps fail.

public class Validator<T> {
  private final T obj;
  private final List<Throwable> exceptions = new ArrayList<>();

   private Validator(T obj) {
    this.obj = obj;
  }
  public static <T> Validator<T> of(T t) {
    return new Validator<>(Objects.requireNonNull(t));
  }

  public Validator<T> validate(Predicate<? super T> validation, String message) {
    if (!validation.test(obj)) {
      exceptions.add(new IllegalStateException(message));
    }
    return this;
  }

  public <U> Validator<T> validate(
      Function<? super T, ? extends U> projection,
      Predicate<? super U> validation,
      String message
  ) {
    return validate(projection.andThen(validation::test)::apply, message);
  }

  public T get() throws IllegalStateException {
    if (exceptions.isEmpty()) {
      return obj;
    }
    var e = new IllegalStateException();
    exceptions.forEach(e::addSuppressed);
    throw e;
  }
}

Next we define an enum Sex.

public enum Sex {
  MALE, FEMALE
}

Now we can introduce the User.

public record User(String name, int age, Sex sex, String email) {
}

And finally, a User object is validated for its name, email, and age using the Validator monad.

public static void main(String[] args) {
    var user = new User("user", 24, Sex.FEMALE, "foobar.com");
    LOGGER.info(Validator.of(user).validate(User::name, Objects::nonNull, "name is null")
        .validate(User::name, name -> !name.isEmpty(), "name is empty")
        .validate(User::email, email -> !email.contains("@"), "email doesn't contains '@'")
        .validate(User::age, age -> age > 20 && age < 30, "age isn't between...").get()
        .toString());
}

Class diagram

Applicability

Consistent and unified error handling is required without relying on exceptions.
Asynchronous computations need clear and maintainable chaining.
State needs to be managed and encapsulated within functional flows.
Dependencies and lazy evaluations are to be handled cleanly and efficiently.

Tutorials

Known Uses

Optional in Java's standard library for handling potential absence of values.
Stream for constructing functional pipelines to operate on collections.
Frameworks like Vavr, providing functional programming enhancements for Java.

Consequences

Benefits:

Increases code readability and reduces boilerplate.
Encourages a declarative programming style.
Promotes immutability and thread safety.
Simplifies complex error handling and state management.

Trade-offs

Can be challenging for developers new to functional programming.
May introduce performance overhead due to additional abstraction layers.
Debugging can be difficult due to less transparent operational flow.

Factory Methodopen in new window: Similar in that it helps instantiate monads, encapsulating object creation logic.
Commandopen in new window: Also encapsulates operations, but monads add context management to the mix.
Decoratoropen in new window: Dynamically enhances functionalities, whereas monads use static typing for consistent composability.