C++ Programming Style

C++ Programming Style still gets regularly mentioned as an important guide to intermediate C++. However, from the perspective of 2007, it's looking long in the tooth. The book was published in 1992, so the modern reader will notice the old fashioned C++ - no templates (therefore no STL), no exceptions, no strings, old style .h headers. The implementation of the assignment operator relies on a check for self assignment (rather than merely using it as an optimisation), and it uses arrays polymorphically. Readers of Exceptional C++ and Effective C++ will know that these are not recommended practices in modern C++. However, I assume that you aren't going to read this without a good grounding in basic and intermediate C++ and can spot the parts which require tweaking. Fortunately, the book is structured as a series of code reviews, so it stands out from the glut of mini-essay type books, and the general principles of class design that the book propounds remain useful. And even fairly advanced programmers probably won't spot all the problems that Cargill highlights, so you'll definitely learn something. I'm giving it four stars because is still covers a core of C++ that is relevant and you can pick it up cheaply. But don't expect it to be fully up to date.

I have read tons of C++ programming books. A lot of them lack of originality. This is where Tom Cargill book shines. The author of this book first presents a small C++ program listing for each chapter and then ask the readers to take few minutes to try to identify the errors or the aspects that could be improved. It is really instructive to find out all the things you have not identified yourself and this is what makes this format so interesting. Some people says that this book is for novice programmers but I disagree. To my opinion, almost all experienced programmers will miss at least half of the problems present in the sample programs.

It is unfair to judge this book from the perspective of the "average" C++ programmer. Tom goes at least three steps further to treat programmers/readers as intelligent beings of the same species who already have the fundamental programming language "mechanics" skills. The reviewer who spewed forth about "coding style" really doesn't "get it." The whole issue of "where you put your braces" and naming conventions isn't what Tom or Tom's book is about. He already assumes that if you're programming C++ you have some idea of when you're going to press enter on the keyboard. (To make whitespace, in case you were wondering...)The inferior thinking that confounds the world of programming is that C++ is an easy language to master. Very few programmers have much hope of aspiring to learn even 80% of the language and use it effectively. Thinking otherwise is like saying that everyone who wants to run a foot race can be Jessie Owens. Tom starts by treating readers as programming peers. That alone is an incredible benefit anytime programming is being done. Prima donas and those guys who always seem too busy to provide their "public interface" are the ones to avoid in learning anything of use regarding C++.I pick up Tom's book every couple of months and browse it. His noted "brevity" is like a good RPG that gives subtle hints that incite thinking for yourself without following what many other books do by drawing a roadmap to one solution that worked for this one situation but may never again apply to anything useful. In my opinion, Tom's "lessons" are appropriately concise.If you haven't read Tom's book, buy it, read it...if you're serious about your C++ skills. There is a *good* reason why Scott Meyers recommends Tom's book. It is something of a unique and interesting perspective on C++, which is really all that any of us can hope to give back to the language. The book is an interesting, insightful perspective that has pragmatic commentary that will help you be a better C++ programmer. At the last (ever?) C++ World conference in December of 1999, a discussion of which books to read evolved out of some other spew. Cargill's book came up as necessary reading, as it always seems to, for the simple fact that it comes from a respected industry professional with an uncany ability to boil out the meat of the topic without overcooking the stew. C++ is, at least, also an art form. Tom's ability with the art of C++ is inspiring. Scott Meyers is another artist. So is Angelika Langer and Herb Sutter, and Andy Koenig, Stan Lippman, Doug Lea and Erich Gamma and Jim Copelien and numerous others. But, for each of them, there are 10,000 very so-so programmers out there spewing forth complete nonsense. Help de-nonsense your world with Tom's book. One person indicated that it is somewhat stale. It is really like fine wine. It just gets better with age.

I've understood the syntax of C++ and the basic ideas of OO for some time now, but I have not used these tools professionally because OO design is so hard, and I have little experience. I have often wished I had a book that would help me evaluate an OO design, and determine what its weaknesses were. This is that book.I especially like the format; the author presents an OO class and then step by step tears it down and shows you what's broken about it. For each problem and its corresponding solution, the author presents a rule of thumb that can help you avoid similar design mistakes in the future. Some of the material is specific to C++, but a lot of it is applicable to any OO language.

This isn't just a book for C++ programmers, it deals with architectural decisions in object-oriented programs whose example base uses C++ programs. There is much valuable information here for object developers of any hat.