D is a programming language designed to fulfill the rigorous demands of large-scale application design without sacrificing the productivity and convenience of high-level languages. Using D’s modeling capabilities, teams of programmers can build large systems out of modular, separately tested parts without having to fear undue complexity, stilted expressiveness, and abstraction penalty.

Although it belongs to the C-related family of languages (C, C++, Java, and C#), by syntax and many semantic facilities D makes definite design departures in the wake of modern experience with these languages.

Advantages specific or unique to D include:

• D’s concurrency model is well-suited for today’s and tomorrow’s highly parallel architectures by favoring message passing, inexpensive immutable sharing, and compiler-verified contention control.

• Built-in mechanisms for unit testing, programming contract enforcement, and modularity make it easy to create and combine high-quality encapsulated components. An expressive type system helps library and system designers create modules that are easy to use correctly and hard to use incorrectly.

• Powerful code introspection and generation mechanisms make generic and generative programming in D not only possible, but also simple and easy. This translates into less code needed for achieving a task and better reusability of the resulting code.

• D doesn’t push an ideological agenda, and instead fosters a multi-paradigm approach within an integrated and remarkably small package. D allows programmers to seamlessly mix and match the advantages of object-oriented, generic, functional, and procedural styles.

• The approach to safety is layered. D has a proper subset that enables development of applications with compiler-verified memory safety. In addition, D acknowledges the occasional necessity of software to tap into memory manipulation or into the speed of assembly code. A large system can have the majority of its code written in D’s safe dialect and a small performance-sensitive core using more dangerous features. The compiler verifies the former part, leaving only a minority of code to human scrutiny.

• Numeric programming is given special attention in the form of high-precision intrinsics, a fast math library, and hardware-assisted vector operations.

D has been implemented for Windows, major Linux distributions, and Mac.

The D Programming Language is a systematic, terse, and authoritative introduction to the language’s concepts. The intended readership consists of intermediate to advanced programmers who are conversant with the essentials of computing and already know one or more other languages (knowledge of some specific language is not assumed).

Building on the successful classic by Kernighan and Ritchie, The D Programming Language progressively introduces language elements along with practical, useful examples in the form of small programs or modules. Special attention is given to fostering sound, principled coding practices in D and beyond. On occasion, the book stresses particularly interesting points by revealing practical-implementation or theoretical-design aspects of the language. Various contemporary paradigms (functional, object-oriented, generic) are discussed within the context of D, as well as notions and idioms original to the D language. A treatment of higher-level concerns, such as concurrency, large-scale design, standard library elements, optimization, and operating system interface, completes the book.

The writing style, which is casual but precise, the clarity of explanations, and, last but not least, the author’s understated sense of humor, all contribute to an enjoyable reading experience.