Customer Reviews

Enough Rope to Shoot Yourself in the Foot: Rules for C and C++ Programming (Unix/ C S)

5 star:		(5)
4 star:		(2)
3 star:		(1)
2 star:		(2)
1 star:		(2)

 

 

I almost passed on this book because of some of the negative reviews.

Okay, it is no Code Complete, but this book is an absolute gem. I have been programming C++ for about 6 years now (VB users be darned, it is my favorite programming language, period) and I was amazed how much I agreed with this author's advice.

Yeah, I learned a lot from Scott Meyers' canonical "Effective" books. But what I like about this book is that it is a no-nonsense, in-your-face, tell-it-like-it-is book that I think is essential for those wet-behind-the-ears C++ programmers (or those whiners who claim that C++ is too hard).

It will probably piss off a lot of Windows programmers as well as he is somewhat anti-Microsoft. However, I am primarily a MFC programmer and
what he says about MFC (earlier editions) and Windows programming in general is absolutely correct.

This book is a must for every novice C/C++ programmer. The experienced C/C++ programmers will probably tell you that they know all about this, so most of them will give it a pass. However, as a experienced C++ programmer, I think that they are making a mistake. They need to get this book on the shelf as well.

At the very least they can loan it to the beginners.

I have very much enjoyed reading this book. I found Mr. Holub's style engaging and entertaining. A lot of the other reviewers disliked his style, but I liked his content & was not offended, except when he expressed negative opinions about Math grads (I'm one). The table of contents serves as a list of all his rules -- very useful. There are a few typos in the book, so you have to know C++ to catch them. E.g., on p.119 the keyword "inline" is missing in 2 spots. If you do what the book literally showed, you'd have problems. I checked. I removed the inline keyword from a function declaration in the class and the function definition in the header, below the class definition. It compiled, but when I tried linking an application to this library code, it failed to link. So this is not a book for beginners. (I found the McGraw-Hill web page for Errata, but it only had a handful of books and this book was not one of them.) But I enjoyed the book and valued his insights. Other reviewers prefer Scott Meyers. I don't like Dr. Meyers books. Meyers rambles and beats around the bush. "Code Complete" by McConnell is just way too long. Time is money. I started to read Maguire's "Writing Solid Code" and closed the book after the 1st chapter. Not impressive at all. I have "Industrial Strength C++" by Henricson and found it dry - no fun to read, but nice reference. "C++ Distilled" by Pohl is nice reference, but also not fun. Have I told you yet that I enjoyed reading Holub's book?

I'm a 11-year mostly C programmer and I AGREE WITH EVERYTHING IN THIS BOOK. I've been a consultant for 5 years - and fixed a lot of code. Beginner programmers should read this and follow it. Computer Science and MIS degrees focus on algorithm design and efficiency and theoretical programming. THIS IS NOT THEORETICAL - this is PRACTICAL. Please, people, write code that's easy to read for the idiot that will come after you; over 50% of the cost of a progam is MAINTENANCE and if you write a simple routine that takes twice as long to execute, OH Well, unless it's a visible delay, no one cares about efficiency.

Use this book as a CORPORATE STYLE GUIDE FOR C & C++ PROGRAMMING. If you follow it and write easy-to-read programs, your employer will SAVE MONEY IN MAINTENANCE. I literally agreed with everything in this book - with one exception. He poo-poos working over 8 hours/day. I disagree. True, programming is creative work and you can burn out, but I find 10 or 12 hour days highly productive because people don't interrupt me, and I try to task switch enough to give myself breaks (or stare out the window blankly to do some subconcious-problem-solving).

In Short - Buy This Book, or borrow it - and save your employers money by creating simpler programs. You'll look like a star by creating something even a junior programmer can modify easily. If you're an expert C or C++ programmer, it's very worth it, too - you get another perspective on ease-of-use. I changed my mind and now will consider using const (I had only used #defines to simplify things - he made a good argument that the compiler is better than the preprocessor at catching brainfarts).

Also, If you can, give it to your manager. They need to know this stuff, too.

I read this book before I read "C Programming FAQs". While "Enough Rope to Shoot Yourself in the Foot" does a good job of pointing out trobule areas in C, the FAQ book does a much better job of point out more general problems because its content is taken from a newsgroup where real people as real questions.

The advantage to this book is that C++ issues are also discussed, which make it a good resource.

I really enjoy this book - it is one of the most enjoyable programming books I have found. Yes, I will agree with most of the points - that he is opinionated, and sometimes wrong, but he is just giving advice, and you are free to examine his coding practices and pick and choose what you like. Yes, Code Complete is a good book, but it is generic, whereas this book is all C/C++ advice, which I find more insightful, where McConnell is more obvious. And it is just plain fun to read.

Allen Holub's book should be on every C/C++ programmer's shelf.
He is a master of his craft and one of the few writers in the
field that is not afraid of being critical of the latest and
greatest fad. His experience comes from years of work in the
trenches and it shows. Ignore his guidelines at your own risk.
You may not agree with everything he writes (I didn't) but it's
all food for thought.

This is very much in the same vein as Scott Meyers 'Effective C++' and 'More Effective C++' books. The Meyers books are probably better, but if you already have them and enjoyed them, you'll like this one. He actually fits imore in his book than either of the Meyers books and covers plain C as well as C++. However, some of his points are more contentious than Meyers'. Holub clearly knows his stuff - there is nothing that is plain wrong - it's just that certain advice isn't appropriate to apply in every case. To be fair, he often points out when one of his rules over-rides another. Another point worth considering is the frequent comments about how bad certain Microsoft programming idioms are. In most cases I agree with him, but couldn't help feeling he goes overboard with this at times e.g. complaining that prefixing member variables with 'm_' is nothing but clutter. Many programmers are forced to work within this framework - it would have been more constructive to offer! ! tips about how best to do this instead of just dismissing it out of hand. Nonetheless, even if you don't always agree with him, Holub makes you think about why you program the way you do. That alone makes this book valuable.

Imagine someone took the Ten Commandments and tossed in some of his own commandments, and published it; of course, it would have some GREAT parts, but it would lend authority to the author's additions which they don't deserve.

Yes, he includes the venerable rules of programming style, as laid out by the classic authors, then tosses in some of his own that are self-indulgent and provincial. Good luck to a beginner trying to sort them out.

Pulling rules out of thin air makes you an autocrat, not an authority. Get The Practice of Programming, Code Complete.

Read McConnell's "Code Construction" instead. Covers the same ground, but better, more informative, and less opinionated. Nothing wrong with opinionated. Just that Holub tends to present his views as the de-facto best practice when other views are just as valid or more valid. He tends to neglect any supporting reasons for his views. For instance, his section on tab stops ("40. Use four-space indents") presents the following reason logic: a) 1 space indents are unreadable and b) 4 space indents are ideal. That's it.

What this book is, is a codification of Holub's prejudices regarding writing C & C++ code. I agree with most of them personally. In fact, he's the only coding style writer I've seen that agrees with me on some, but nevertheless this appears to be simply a vanity book in style and substance.

McConnell's book is more thorough in style and substance, even if I think he's wrong in many cases.

To write programs in C is a nice but also brutal way of commanding your computer to do what you want it to do. I've gone from BASIC to Assembler to Pascal to C to C++. As I've taken impression from all of these languages I found this book very informative and useful when I tried to write better and cleaner C-code. The 3-star rating of the book comes from the "I know best, because I've coded alot"-attitude and the lack of pros and cons in the book.

This book can make your coding style better but as always, use your good judgement. Holub's advices aren't perfect.