About the Author
David Abrahams is a founding member and moderator of the Boost C++ library development group. Dave has been an ANSI/ISO C++ committee member since 1996, where he is best known for contributing a theory, specification, and implementation of exception handling for the C++ standard library. His company, Boost Consulting, provides Boost-related support and development services and professional training in the art of software construction.
Aleksey Gurtovoy is a technical lead for MetaCommunications and a contributing member of the Boost C++ community. Aleksey is the original author of the Boost Metaprogramming Library. He has been working with C++ since 1993, and holds a M.S. degree in computer science from Krasnoyarsk Technical State University, Russia.
Excerpt. © Reprinted by permission. All rights reserved.
In 1998 Dave had the privilege of attending a workshop in Generic Programming at Dagstuhl Castle in Germany. Near the end of theworkshop, a very enthusiastic Kristof Czarnecki and Ullrich Eisenecker (of Generative Programming fame) passed out a few pages of C++ source code that they billed as a complete Lisp implementation built out of C++ templates. At the time it appeared to Dave to be nothing more than a curiosity, a charming but impractical hijacking of the template system to prove that you can write programs that execute at compile time. He never suspected that one day he would see a role for metaprogramming in most of his day-to-day programming jobs. In many ways, that collection of templates was the precursor to the Boost Metaprogramming Library (MPL): It may have been the first library designed to turn compile-time C++ from an ad hoc collection of "template tricks" into an example of disciplined and readable software engineering. With the availability of tools to write and understand metaprograms at a high level, we've since found that using these techniques is not only practical, but easy, fun, and often astoundingly powerful.
Despite the existence of numerous real systems built with template metaprogramming and the MPL, many people still consider metaprogramming to be other-worldly magic, and often as something to be avoided in day-to-day production code. If you've never done any metaprogramming, it may not even have an obvious relationship to the work you do. With this book, we hope to lift the veil of mystery, so that you get an understanding not only of how metaprogramming is done, but also why and when. The best part is that while much of the mystery will have dissolved, we think you'll still find enough magic left in the subject to stay as inspired about it as we are.
Making the Most of This Book
The first few chapters of this book lay the conceptual foundation you'll need for most everything else we cover, and chapters generally build on material that has come before. That said, feel free to skip ahead for any reason—we've tried to make that possible by providing cross-references when we use terms introduced earlier on.
Chapter 10, Domain-Specific Embedded Languages, is an exception to the rule that later chapters depend on earlier ones. It focuses mostly on concepts, and only appears late in the book because at that point you'll have learned the tools and techniques to put Domain-Specific Embedded Languages into play in real code. If you only remember one chapter by the time you're done, make it that one.
Near the end of many chapters, you'll find a Details section that summarizes key ideas. These sections usually add new material that deepens the earlier discussion,* so even if you are inclined to skim them the first time through, we suggest you refer back to them later.
We conclude most chapters with exercises designed to help you develop both your programming and conceptual muscles. Those marked with asterisks are expected to be more of a workout than the others. Not all exercises involve writing code—some could be considered "essay questions"—and you don't have to complete them in order to move on to later chapters. We do suggest you look through them, give a little thought to how you'd answer each one, and try your hand at one or two; it's a great way to gain confidence with what you've just read.
This book comes with a companion CD that supplies the following items in electronic form
- Sample code from the book.
- A release of the Boost C++ libraries. Boost has become known for high-quality, peer-reviewed, portable, generic, and freely reusable C++ libraries. We make extensive use of one Boost library throughout the book—the Boost Metaprogramming Library (MPL)—and we discuss several others.
- A complete MPL reference manual, in HTML and PDF form.
- Boost libraries discussed in this book that are not yet part of an official release.
index.html file at the top level of the CD will provide you with a convenient guide to all of its contents. Additional and updated material, including the inevitable errata, will appear on the book's Web site: http://www.boost-consulting.com/mplbook. You'll also find a place there to report any mistakes you might find.
Trying It Out
To compile any of the examples, just put the CD's
boost 1 32 0/ directory into your compiler's
The libraries we present in this book go to great lengths to hide the problems of less-than-perfect compilers, so it's unlikely that you'll have trouble with the examples we present here. That said, we divide C++ compilers roughly into three categories.
A. Those with mostly conforming template implementations. On these compilers, the examples and libraries "just work." Almost anything released since 2001, and a few compilers released before then, fall into this category.
B. Those that can be made to work, but require some workarounds in user code.
C. Those that are too broken to use effectively for template metaprogramming.
Appendix D lists the compilers that are known to fall into each of these categories. For those in category B, Appendix D refers to a list of portability idioms. These idioms have been applied to the copies of the book's examples that appear on the accompanying CD, but to avoid distracting the majority of readers they don't appear in the main text.
The CD also contains a portability table with a detailed report of how various compilers are doing with our examples. GCC is available free for most platforms, and recent versions have no problems handling the code we present here.
Even if you have a relatively modern compiler from category A, it might be a good idea to grab a copy of GCC with which to cross-check your code. Often the easiest way to decipher an inscrutable error message is to see what some other compiler has to say about your program. If you find yourself struggling with error messages as you try to do the exercises, you might want to skip ahead and read the first two sections of Chapter 8, which discusses how to read and manage diagnostics.
And now, on to C++ Template Metaprogramming!
— Dave and Aleksey
* We borrowed this idea from Andrew Koenig and Barbara Moo's Accelerated C++: Practical Programming By Example.