Customer Reviews

Mastering Regular Expressions, Second Edition

5 star:		(102)
4 star:		(24)
3 star:		(10)
2 star:		(6)
1 star:		(2)

 

 

Jeffrey Friedl's "Matering Regular Expressions" does a facinating job in taking you through the jungle (and I mean jungle) of RegExp.

I am a Perl/CGI programmer, and I had considered myself good at RegExp even before I read this book. Most of the things I knew were from Programming Perl, 3rd edition (chapter 5, Pattern Matching). But I still decided to give Jeffrey a chance since I was having some trouble with my Parse::Syntax module, which is designed to parsing *any* programming language and highlighting the syntax accordingly (provided it has a syntax/grammer file written for the specific language). The accuracy of the parser (and more importantly the speed) does depend on well crafted regular expressions.

As I started reading the book, I couldn't stop. I took it to my school's cafeteria with me and no one could make me leave untill I finished the whole book. I was excited. I was pleased! Here is the outline of tha chapters:

Chapter 1 and 2 introduce you to regular expressions and give some basic regex examples. Mail utility and date matching is two of them.

Chapter 3 mostly talks about conventions that all the regex tools follow and their differences.

Chapter 4 deals with Traditional NFA, POSIX NFA and DFA regex engines and their pros and cons. What you'll like the most is the details provided by the author on each and every single example. He also uses a lot of step-by-step illustrations to take you deeper into the regex engine itself and see/feel how it works. He shows the point of backtrackings and provides awesome benchmarks. He uses such examples of matching a quote, allowing escaped quotes inside the pattern, matching C-style comments, IP addresses and many more.

Chapter 5 deals with writing efficient regular expressions for NFA engines. It also re-vists some of the examples provided in the previous chapter and fine-tunes them.

Chapter 6 and 7 deals with Tool-Language specific features of Regex engines. Chapter 6 is dedicated to Awk, Tcl and GNU Emacs, whereas chapter 7 is entirely dedicated to Perl, good over 100 pages of Perlism.

It's true that there're features that Perl 5.6 offers when it comes to regex that didn't exist at the time this book went to press, (lookbehinds, for example). But this no way makes this book dated. Just take my word for it. Jeffrey put together a great masterpiece that will not die for many years, no matter how fast the technology tends to enhance
Haven't read anything more exciting than this for many years.

I got Mastering Regular Expressions in 1998 and I peek back regularly and it am proud to say is one of most used and useful books I own.

Regular expressions are for everyone. They are simple to write and can be close to impossible to read and even unimaginably harder to understand. But reading comes after you can craft one.

And this art of crafting RE is explained in astonishing detail and analysis in this book. Reader is carefully guided through basics, differences and common and uncommon pitfalls. Some of the parts are definitely not for the faint hearted! And this is especially true for parts that cover Perl RE. While main topic is Perl RE engine, a deep-enough travel through different RE dialects is made to help RE-crafting in almost any tool that supports REs.

I've used the methods described in the book in tools as different as MS VC++, various editors, search engines and programming (Perl, PHP, C++, ...). The ability to use RE usually impressed people to the point they were sure I am using some incredible magic.

But I must say that the most useful feature delivered in this book was ability to PARSE, UNDERSTAND and FIX a (slow, broken, strange acting) RE. This can be slowly absorbed by reader's (open) mind while carefully reading the book. The difference between meant/written can be as extreme as it can get in RE. I can (proudly) say I've been able to fix several complex REs, by simply (one could say blindly) following the rules laid in the book.

My opinion is thath without this book real understanding of RE is almost impossible. I recommended it all of those who use programs that support (any form) of RE engine. It will help them solve some (hard) problems incredibly fast!

In spite of the vow that Friedl would not put himself (and his family) through the rigors of writing a followup to his first edition, I sure am glad he made the decision to write a second edition. Over 5 years ago, I got the first such book and found it to be a humbling experience -- realizing just how little I *really* knew about regular expressions (in spite of thinking my skills were quite advanced in that area).

Now, years later and as an instructor of UNIX at North Lake College in Irving, Texas, I highly recommend this book to even our first year students. Friedl's clear explanation of this topic and the manner in which he presents the material makes it comprehensible to even those that have never had experience with regex's before. Like Perl's Larry Wall, Jeffrey Friedl has a strong background in natural languages. That contributes not only to bhis lucid writing style but also helps in terms of understanding regex's as a "little language".

This 2nd edition is particularly welcome because of its extensive coverage of regular expressions in the context of Java. That ons aspect alone is sufficient to get the updated second edition (to say nothing of the enhancements in Perl since the first edition of the book).

I can't say enough good things about this book.

Regular expressions are such a powerful concept, but some of that power would be easy to miss if not guided correctly through the topic. Jeffrey Freidl does a great job of making a potentially very dry subject interesting, even while getting very involved in all the complexities that are inherent in such a powerful abstraction. He also does a good job of presenting both the general topic of regular expressions and the specific characteristics of the various tools available to process them.

This is a book that you will come back to many times. I actually read it as a precursor to learning Perl several years ago. Then just recently I revisited it while taking a class on compiler construction and found that it still had useful insights for me. If you're a committed programmer, this book should be on your shelf.

Also, the criticisms I've read in reviews here are pretty misguided. It is "chatty", because the author has a genuine enthusiasm for his subject. It is also not a "teach yourself regular expressions in 24 hours" kind of book. There are actually lots of cookbook style examples, but the main point of the book is to give you a solid enough understanding of the general topic that you don't need examples to craft your regexes. In this case, little patience as a reader will be much rewarded.

I made the mistake of buying a bunch of books on Perl to try to learn Perl programming. I made progress, but it was slow -- and it beat me down. Then I bought THIS book. Perl's implementation of Regular Expressions is a great deal of the functionality of that language. This is true to the extent that trying to read Perl script without knowing Regular Expressions is tedious at best. This is not just another computer book. Jeffrey Friedl has put his heart and soul into this work, and it shows. He even offers free updates (in the truest, most honest sense of the word) from his own web page. This is the best programming book I own.

After reading this book, Perl script reads like USA Today! Save yourself lots of heartache. Buy this book, Mastering Regular Expressions -- read it -- THEN learn Perl!
Jeff Morris

I bought this book on a whim, mainly because I try to buy (and read... ugh!) a hard-core technical book every month or so, but mainly because my UNIX scripting abilities have become rusty with disuse. I used to be able to write a tight, 10-line csh script to mangledit thousands of files at a single time. Now I hack away at files manually with vi. The other day, I even forgot how to search & replace. My kung fu isn't nearly what it used to be.

It usually takes me about a month to slog through a new book (especially academic texts, which are great but make you want to have a stiff martini before each new chapter) but I tore through Friedl's book in a few days. It's an outstanding reference for understanding & learning to use regular expressions.

Regexes can be cryptic to say the least, but Friedl offers many examples, broken down into step-by-step instructions and explanations of how each regex works (in many cases, right down to the individual character level). He covers a variety of platforms and languages - the hardcore Perl enthusiast will enjoy this book greatly, but he offers fairly equal time to alternative languages like Java and the "grep" family.

All that said, this book is an outstanding technical reference, pure and simple, for two reasons:

- Friedl uses an interesting new typesetting convention to illustrate which sections of text are regular expressions, and which sections are not. It's hard to describe (and impossible to reproduce here) but they look like 90-degree braces at the upper-left and lower-right corners. This is a FANTASTIC approach and I for one would love to see this extended to other technical books.

- Speaking of other things I would love to see extended to other technical books, THIS BOOK HAS ALMOST NO ERRORS! This is even more impressive considering the fact that, with regular expressions, screwing up EVEN A SINGLE CHARACTER is significant. Nothing [upsets]memore than spending $50-100 perfectly good dollars on a sloppy, error-laden, grammatically-challenged, poorly-edited, ill-reviewed and badly-structured book (which pretty much describes 90% of all the technical books out there). He made a few mistakes, the vast majority of which are extremely trivial in nature, and all of which were quickly posted to his website as errata. If only the other 90% of the technical authors out there were even half as diligent.

All told, this book belongs on your bookshelf.

This is a book about Regular Expressions 'for Perl and Other Tools' but really it's a book about Perl. Other tools are covered, but coverage is a bit woolly, so if you are reading it to get the low-down on Python's RegEx capabilities etc. this book won't get you far. I can't help that feel that if the book had just concentrated on Perl, rather than giving sketchy coverage of `Other Tools', it could have been even better (and would have definitely got five starts). That said, the extra coverage (particularly the RegEx engine material) was very interesting and has turned out invaluable in practice.

If you are an absolute beginner to Perl or programming you will need another book(s) to get the most out of this one, but it is a gentle and thorough introduction that won't leave you scratching your head, which is a feat in itself considering the complexity of Regular Expressions.

This is an excellent book for getting the most out of Perl's RegEx capabilities, you will close the back cover feeling that you genuinely have Mastered the subject. It is without doubt the best book available on the subject, nicely written, with a friendly and un-patronising tone (grammatical errors aside), you can't really go wrong with this one.

There could have been some more useful real world examples/projects included, and the book could probably do with an overhaul (it was published in 97) to accommodate Perl 5.6 and other developments (O'Reilly are you listening?), but it is still the best out there.

If RegEx is a subject you need to get to grips with, this book is the solution.

Whether you are searching text documents with BBedit or flagging Usenet articles with Newswatcher, learning to use regular expressions can make your life a lot easier. Regular expressions ("regex") are text strings that allow you to search for complex variations of words and phrases. For instance, with the simple regex "gr[ea]y" you can flag any instances of the words "gray" or "grey". While a lot of applications support regex, finding out how to use this powerful function can be very difficult. There is suprisingly little online documentation of regex.

Friedl's excellent book fills this void. In the introductory material, he covers what a regex is and how to use the regex language to do your own searches. The rest of the book, however, is aimed at programmers only, dealing with how various programming languages (such as Perl) implement regex. If all you want to do is use regex for searching text files or flagging Usenet articles, most of the book will not be useful to you. However, until someone comes out with a pocket manual of regex, Friedl's book is the only game in town.

If you are an intermediate-or-better PHP programmer who wants to get a handle on regular expressions, you'll find this book so incredibly handy. I'm only a few chapters into the book and already I get the gist of regexs and how to use them with PHP... plus the Perl examples are very, very clear (the author stays focused on regex, Perl is just used to put it in a program's context, so he explains clearly what is going on with each line and you can see how it would translate easily to PHP)... the first chapter of this book did more to explain regexs than spending weeks searching the web for a *clear* tutorial.

I suspect I'll skip most of the second half of the book-- after all, it's called "mastering regular expressions," and I don't really want to master them, I just want to be really good and then keep going with other stuff. (The author spends chapters looking at nearly EVERY regex tool out there, etc... which is not useful to me since I know I'm going to be using regexs 99% of the time in PHP.) If there is anything I could change about this book, it would be to put the first five-or-so chapters into a separate slim volume and *PLEASE* include a quick reference card for all the metachars! (that's my next purchase...)

But listen, if you want to LEARN regexs, this book does the job, the writing is crystal-clear, the mini-quizzes every few pages are excellent, the typographic conventions (used to really make the expressions clear) are clever and smart, and this book has EVERYTHING you will EVER need on regular expressions, in your entire life. <g> Well worth the $; you won't ever need another regex book again.

§
The third edition just came out -- and that makes 10 years since this book was first published. That statistic alone indicates the steady value of "Mastering Regular Expressions" across the constant changes in applied software technology over that time. Regular expressions, still the most under-used power tool, needs a book like this.

Specifically, this third edition features enhanced coverage of PHP in the early tutorial chapters, plus an all new chapter devoted entirely to PHP regular expressions.

Also new in this edition, the Java chapter has been rewritten and expanded to reflect new features as Java itself has developed.
§