Programming Pearls (2nd Edition) [Paperback]

Jon Bentley (Author)

Book Description

ISBN-10: 0201657880 | ISBN-13: 978-0201657883 | Publication Date: October 7, 1999 | Edition: 2

"The first edition of Programming Pearls was one of the most influential books I read early in my career, and many of the insights I first encountered in that book stayed with me long after I read it. Jon has done a wonderful job of updating the material. I am very impressed at how fresh the new examples seem." --Steve McConnell When programmers list their favorite books, Jon Bentley's collection of programming pearls is commonly included among the classics. Just as natural pearls grow from grains of sand that irritate oysters, programming pearls have grown from real problems that have irritated real programmers. With origins beyond solid engineering, in the realm of insight and creativity, Bentley's pearls offer unique and clever solutions to those nagging problems. Illustrated by programs designed as much for fun as for instruction, the book is filled with lucid and witty descriptions of practical programming techniques and fundamental design principles. It is not at all surprising that Programming Pearls has been so highly valued by programmers at every level of experience.In this revision, the first in 14 years, Bentley has substantially updated his essays to reflect current programming methods and environments. In addition, there are three new essays on / testing, debugging, and timing / set representations / string problems All the original programs have been rewritten, and an equal amount of new code has been generated. Implementations of all the programs, in C or C++, are now available on the Web. What remains the same in this new edition is Bentley's focus on the hard core of programming problems and his delivery of workable solutions to those problems. Whether you are new to Bentley's classic or are revisiting his work for some fresh insight, the book is sure to make your own list of favorites. 0201657880B04062001

Editorial Reviews

Amazon.com Review

Fourteen years after it was first issued, C++ expert Jon Bentley reinvents a true classic with the second edition of his Programming Pearls. Completely revised and brought up to date with all new code examples in C and C++, this book remains an exceptional tutorial for learning to think like a programmer.

The "pearls" in question center not only on choosing the right algorithms (like binary searches, sorting techniques, or sparse arrays) but also on showing how to solve problems effectively. Each chapter frames a particular programming task--such as sorting numbers, creating anagrams, or counting the words in a block of text--many drawn from Bentley's experiences in his long career as a developer. The book traces the process of arriving at a fast, efficient, and accurate solution, along with code profiling to discover what works best. After refining the correct answer, each chapter enumerates programming principles that you can use on your own.

The author also challenges you to think like an engineer, and each chapter ends with about a dozen problems to get you thinking creatively about design issues. (Sidebars on such historical topics as the first computer solutions to computer chess, spell-checking, and even architectural design help create a perspective on successful problem solving and make for a truly educational and enjoyable tour of how to become a better programmer.) Bentley also asks the reader to think analytically about the world with "back of the envelope" estimation techniques drawn from engineering. Appendices list the algorithms and code rules covered in the book, plus some sample solutions.

Fans of the first edition of this title will be pleased to see this favorite computer text brought up to date for today's faster hardware. Whether you want to improve your command of algorithms or test your problem-solving skills, the new version of Programming Pearl is a challenging, instructive, and thoroughly entertaining resource. --Richard Dragan

Topics covered: Programming and problem-solving tutorial, sorting algorithms, merge sort, bit vectors, binary searches, program correctness and testing, improving performance, engineering and problem-solving techniques, performance estimates, designing for safety, divide-and-conquer and scanning algorithms, tuning code, tips for more efficient memory usage, insertion sort, quicksort algorithms, sparse arrays, searching algorithms, binary search trees, heaps, priority queues, searching text, and generating random text.

From the Inside Flap

Computer programming has many faces. Fred Brooks paints the big picture in

The Mythical Man Month; his essays underscore the crucial role of management

in large software projects. At a finer grain, Steve McConnell teaches good programming

style in Code Complete. The topics in those books are the key to good software

and the hallmark of the professional programmer. Unfortunately, though, the

workmanlike application of those sound engineering principles isn't always thrilling

-- until the software is completed on time and works without surprise.

About the Book
The columns in this book are about a more glamorous aspect of the profession:

programming pearls whose origins lie beyond solid engineering, in the realm

of insight and creativity. Just as natural pearls grow from grains of sand that

have irritated oysters, these programming pearls have grown from real problems

that have irritated real programmers. The programs are fun, and they teach important

programming techniques and fundamental design principles.

Most of these essays originally appeared in my ''Programming Pearls'' column

in Communications of the Association for Computing Machinery. They were

collected, revised and published as the first edition of this book in 1986.

Twelve of the thirteen pieces in the first edition have been edited substantially

for this edition, and three new columns have been added.

The only background the book assumes is programming experience in a high-level

language. Advanced techniques (such as templates in C++) show up now and then,

but the reader unfamiliar with such topics will be able to skip to the next

section with impunity.

Although each column may be read by itself, there is a logical grouping to

the complete set. Columns 1 through 5 form Part I of the book. They review programming

fundamentals: problem definition, algorithms, data structures and program verification

and testing. Part II is built around the theme of efficiency, which is sometimes

important in itself and is always a fine springboard into interesting programming

problems. Part III applies those techniques to several substantial problems

in sorting, searching and strings.

One hint about reading the essays: don't go too fast. Read them carefully,

one per sitting. Try the problems as they are posed -- some of them look easy

until you've butted your head against them for an hour or two. Afterwards, work

hard on the problems at the end of each column: most of what you learn from

this book will come out the end of your pencil as you scribble down your solutions.

If possible, discuss your ideas with friends and colleagues before peeking at

the hints and solutions in the back of the book. The further reading at the

end of each chapter isn't intended as a scholarly reference list; I've recommended

some good books that are an important part of my personal library.

This book is written for programmers. I hope that the problems, hints, solutions,

and further reading make it useful for individuals. The book has been used in

classes including Algorithms, Program Verification and Software Engineering.

The catalog of algorithms in Appendix 1 is a reference for practicing programmers,

and also shows how the book can be integrated into classes on algorithms and

data structures.

The Code
The pseudocode programs in the first edition of the book were all implemented,

but I was the only person to see the real code. For this edition, I have rewritten

all the old programs and written about the same amount of new code. The programs

are available at this book's web site. The code includes much of the scaffolding

for testing, debugging and timing the functions. The site also contains other

relevant material. Because so much software is now available online, a new theme

in this edition is how to evaluate and use software components.

The programs use a terse coding style: short variable names, few blank lines,

and little or no error checking. This is inappropriate in large software projects,

but it is useful to convey the key ideas of algorithms. Solution 5.1 gives more

background on this style. The text includes a few real C and C++ programs, but

most functions are expressed in a pseudocode that takes less space and avoids

inelegant syntax. The notation for i = 0, n) iterates i from

0 through n-1. In these for loops, left and right parentheses denote

open ranges (which do not include the end values), and left and right square

brackets denote closed ranges (which do include the end values). The phrase

function(i, j) still calls a function with parameters i and j,

and arrayi, j still accesses an array element.

This edition reports the run times of many programs on ''my computer'', a 400MHz

Pentium II with 128 megabytes of RAM running Windows NT 4.0. I timed the programs

on several other machines, and the book reports the few substantial differences

that I observed. All experiments used the highest available level of compiler

optimization. I encourage you to time the programs on your machine; I bet that

you'll find similar ratios of run times.

To Readers of the First Edition
I hope that your first response as you thumb through this edition of the book

is, ''This sure looks familiar.'' A few minutes later, I hope that you'll observe,

''I've never seen that before.''

This version has the same focus as the first edition, but is set in a larger

context. Computing has grown substantially in important areas such as databases,

networking and user interfaces. Most programmers should be familiar users of

such technologies. At the center of each of those areas, though, is a hard core

of programming problems. Those programs remain the theme of this book. This

edition of the book is a slightly larger fish in a much larger pond.

One section from old Column 4 on implementing binary search grew into new Column

5 on testing, debugging and timing. Old Column 11 grew and split into new Columns

12 (on the original problem) and 13 (on set representations). Old Column 13

described a spelling checker that ran in a 64-kilobyte address space; it has

been deleted, but its heart lives on in Section 13.8. New Column 15 is about

string problems. Many sections have been inserted into the old columns, and

other sections were deleted along the way. With new problems, new solutions,

and four new appendices, this edition of the book is 25 percent longer.

Many of the old case studies in this edition are unchanged, for their historical

interest. A few old stories have been recast in modern terms.

Acknowledgments for the First Edition
I am grateful for much support from many people. The idea for a Communications

of the ACM column was originally conceived by Peter Denning and Stuart Lynn.

Peter worked diligently within ACM to make the column possible and recruited

me for the job. ACM Headquarters staff, particularly Roz Steier and Nancy Adriance,

have been very supportive as these columns were published in their original

form. I am especially indebted to the ACM for encouraging publication of the

columns in their present form, and to the many CACM readers who made

this expanded version necessary and possible by their comments on the original

columns.

Al Aho, Peter Denning, Mike Garey, David Johnson, Brian Kernighan, John Linderman,

Doug McIlroy and Don Stanat have all read each column with great care, often

under extreme time pressure. I am also grateful for the particularly helpful

comments of Henry Baird, Bill Cleveland, David Gries, Eric Grosse, Lynn Jelinski,

Steve Johnson, Bob Melville, Bob Martin, Arno Penzias, Marilyn Roper, Chris

Van Wyk, Vic Vyssotsky and Pamela Zave. Al Aho, Andrew Hume, Brian Kernighan,

Ravi Sethi, Laura Skinger and Bjarne Stroustrup provided invaluable help in

bookmaking, and West Point cadets in EF 485 field tested the penultimate draft

of the manuscript. Thanks, all.

Acknowledgments for the Second Edition
Dan Bentley, Russ Cox, Brian Kernighan, Mark Kernighan, John Linderman, Steve

McConnell, Doug McIlroy, Rob Pike, Howard Trickey and Chris Van Wyk have all

read this edition with great care. I am also grateful for the particularly helpful

comments of Paul Abrahams, Glenda Childress, Eric Grosse, Ann Martin, Peter

McIlroy, Peter Memishian, Sundar Narasimhan, Lisa Ricker, Dennis Ritchie, Ravi

Sethi, Carol Smith, Tom Szymanski and Kentaro Toyama. I thank Peter Gordon and

his colleagues at Addison-Wesley for their help in preparing this edition.

0201657880P04062001

See all Editorial Reviews

Product Details

• Paperback: 256 pages
• Publisher: Addison-Wesley Professional; 2 edition (October 7, 1999)
• Language: English
• ISBN-10: 0201657880
• ISBN-13: 978-0201657883
• Product Dimensions: 9.2 x 6.3 x 0.7 inches
• Shipping Weight: 12.8 ounces (View shipping rates and policies)
• Average Customer Review: 4.2 out of 5 stars  See all reviews (33 customer reviews)
• Amazon Best Sellers Rank: #19,136 in Books (See Top 100 in Books)

More About the Author

Discover books, learn about writers, read author blogs, and more.

Customer Reviews

Most Helpful Customer Reviews

64 of 66 people found the following review helpful:

5.0 out of 5 stars A course in how to think like an experienced programmer, June 17, 2000

By 

Mike Christie (Austin, TX USA) - See all my reviews
(REAL NAME)   

This review is from: Programming Pearls (2nd Edition) (Paperback)

The thirteen columns in this book appeared in the Communications of the ACM between 1983 and 1985. There can't be more than a couple of technical books on computing from that era that are still worth reading. Kernighan & Ritchie's book, "The C Programming Language", is one that springs to mind; this book is definitely another, and will probably outlast K&R as it has almost no ties to existing or past hardware or languages.

What Bentley does in each of these columns is take some part of the field of programming--something that every one of us will have run into at some point in our work--and dig underneath it to reveal the part of the problem that is permanent; that doesn't change from language to language. The first two parts cover problem definition, algorithms, data structures, program verification, and efficiency (performance, code tuning, space tuning); the third part applies the lessons to example pseudocode, looking at sorting, searching, heaps, and an example spellchecker.

Bentley writes clearly and enthusiastically, and the columns are a pleasure to read. But the reason so many people love this book is not for the style, it's for the substance--you can't read this book and not come away a better programmer. Inefficiency, clumsiness, inelegance and obscurity will offend you just a little more after you've read it.

It's hard to pick a favourite piece, but here's one nice example from the algorithm design column that shows how little the speed of your Pentium matters if you don't know what you're doing. Bentley presents a particular problem (the details don't matter) and multiple different ways to solve it, calculating the relationship between problem size and run time for each algorithm. He gives, among others, a cubic algorithm (run time equal to a constant, C, times the cube of the problem size, N--i.e. t ~ CN^3), and a linear algorithm with constant K (t ~ KN). He then implemented them both: the former in fine-tuned FORTRAN on a Cray-1 supercomputer; the latter in BASIC on a Radio Shack TRS-80. The constant factors were as different as they could be, but with increasing problem size the TRS-80 eventually has to catch up--and it does. He gives a table showing the results: for a problem size of 1000, the Cray takes three seconds to the TRS-80's 20 seconds; but for a problem size of 1,000,000, the TRS-80 takes five and a half hours, whereas the Cray would take 95 years.

The book is informative, entertaining, and will painlessly make you a better programmer. What more can you ask?

80 of 85 people found the following review helpful:

5.0 out of 5 stars The Pearls Still Glitter After a Decade, March 2, 2000

By 

Charles Ashbacher (Marion, Iowa United States) - See all my reviews
(TOP 500 REVIEWER)    (VINE VOICE)    (HALL OF FAME REVIEWER)   

This review is from: Programming Pearls (2nd Edition) (Paperback)

Without any doubt, my favorite article in _Communications of the ACM_ in the 1980's was the regular `Programming Pearls' articles by Jon Bentley. When the first edition of these collected gems was published, I read it with great delight. Now, over a decade later, a second edition has been published, containing the same problems with additional modifications and notations. Given the enormous changes in programming since the mid 80's, your first reaction might be that this book is dated and therefore irrelevant. Nothing could be further from the truth.
Elegant solutions to complex programming problems are free from the rot of time. Programming is a thought process largely independent of the notation used to write it down. The solutions are sketched and explained rather than coded, and the solutions are complete. There is a certain mystique about taking a complex problem, finding an initial solution and then refining it down until it kicks some big time. There are some major lessons in program refinement explained in these solutions.
Coding a binary search is covered quite extensively, which may seem like a waste of space, as this problem was solved decades ago. However, that solution took decades to get right, and this is one of those "separates the coders from the key bangers" type of problem. Other problems examined include performance tuning, squeezing space and program correctness. While the improvement in the performance of the hardware has been astounding since these solutions were written, that does not make them obsolete. The complexity of the programs that we now build has risen even faster, so performance and space considerations are just as critical.
Some problems were here at the beginning and will still be here at the end. Even though there may be canned code to handle them, these problems are generic enough that the solutions can be applied elsewhere, so we must learn how to solve them. Understanding these problems and their solutions will give you a fundamental skill set that will serve you well for a long time.

24 of 28 people found the following review helpful:

5.0 out of 5 stars The how-to for profile-based tuning, July 27, 2002

By 

Bob Carpenter (New York, NY) - See all my reviews

This review is from: Programming Pearls (2nd Edition) (Paperback)

Bentley's classic, "Programming Pearls", makes an important point, namely that you won't get good performance without careful coding and profile-based tuning. And it's made clearly, concisely and with compelling examples. The choice of language (C), and the choice of problems (those from computer science 101 we all think we know cold) betrays the sophistication of Bentley's analyses.

Suppose, for the sake of argument, that you have a binary search that's holding up your loop. Or your Huffman coding just isn't snappy enough? "How is that possible?", you might say, fresh out of computer-science 201, "Didn't we just prove these algorithms are optimal?" Well yes, asymptotically up to an arbitrary constant multiplier. But this is the real world, and your code needs to go faster. If this sounds like your predicament, pull up a chair and read "Programming Pearls"; if it's not, you might wonder what all the fuss is about.

Next, fire up your favorite hardware (Sparc or x86 or PowerPC), favorite language (Perl, Java, or even C), favorite release of that language, along with your favorite interpreter or compiler (Hotspot or standard? GCC or Visual C++). And you'll need a profiler; might as well treat yourself to a good one if you're serious. Then fire up your code with a representative range realistic test data and observe what happens. Function by function, byte by byte. Then try to be as clever as Bentley in (a) figuring out why, (b) trying a range of alternatives, and (c) making it all go faster with minor tuning. Typically, you'll find a single bottleneck taking an order of magnitude more time than everything else, and work on that. Repeat until fast enough.

As well as this simple, yet surprisingly effective and realistic methodology, Bentley provides a range of concrete tips on making things go faster, from tweaking data structures to unfolding loops (especially precomputing low-order cases) to using accumulators and caching, all with an eye to underlying memory, communication and CPU resources.

Real code that has to run fast, like the code that we write at my current company for signal processing, speech recognition and speech synthesis, typically looks like the end-product of Bentley's refactorings. And it gets that way following exactly the path he lays out: analyze the problem, choose the right algorithm (or the right few to evaluate), and then tune it up using profiling.

"Programming Pearls" is the beginning of the road. You will need to look elsewhere for topics such as compression for memory saving, numerical algorithms, effective concurrency and memory sharing, efficient buffered I/O, garbage collection, and the wide range of dynamic programming and heuristic techniques.