Functional Patterns in Java #

Conventional programming languages are growing ever more enormous, but not stronger.

In his Turing-Award lecture in 1977, John Backus asked: Can programming be liberated from the von Neumann style? He gave a damning report on the state of conventional programming languages at the time.

Inherent defects at the most basic level cause them to be both fat and weak: their primitive word-at-a-time style of programming inherited from their common ancestor—the von Neumann computer, their close coupling of semantics to state transitions, their division of programming into a world of expressions and a world of statements, their inability to effectively use powerful combining forms for building new programs from existing ones, and their lack of useful mathematical properties for reasoning about programs.

This assessment is decades old. The functional programming language Lisp is two decades older but was not conventional at the time, and still isn’t today. Functional programming languages focus on expressions rather than statements, which enables them to provide powerful combining forms for programs with useful properties for reasoning. As a result, conventional programming languages are incorporating patterns from functional programming. Java is no exception.

The word pattern is used here in the sense of design pattern (and not as in pattern matching.) In functional programming, design patterns often can be implemented in terms of so called higher-order functions, which are functions that take other functions as arguments or return other functions as results. In modern Java, functions can be represented using functional interfaces and corresponding instances can be written concisely using lambda expressions or method references.

Representing functions as values (or objects in Java) allows them to be used in expressions without executing them right away as they would be in statements including procedure (or method) calls. In expressions, functions can be more flexibly combined for building new programs from existing ones. Algebraic properties, similar to primary school arithmetic, can be used to manipulate combined expressions, including those containing functions.

In this tutorial, we will look at combining forms, or combinators, that are particularly prevalent: map, filter, and flatMap. All of them are provided for different types by Java’s standard library. We will discuss how to use them for Streams and Optionals and then implement our own type providing these combinators.

The underlying source code is available online, and this tutorial includes tasks to extend it. If you want to follow along, you can use your own Java development environment or install Docker and VS Code with the Remote Development Extension Pack to use a predefined environment without creating (or adjusting) your own. To use the predefined environment in VS Code download and unpack the source code (or clone the repository using git), open the repository folder in VS Code, click on the Remote Containers icon in the bottom-left corner, and select Reopen Folder in Container.