|
Based on 6 reviews.
 Brain embedding knowledge, July 9, 1999
|
A "must read" book for C++ programmers.,
December 13, 1998
|
Prentice Hall; ISBN: 0023887257 ; Dimensions (in inches): 1.60 x 9.58 x 7.23
The individual features of C++ discussed in the book are not only complex by themselves, but when put together in a program they interact in highly non-intuitive ways. The author discuss each of the features very briefly and as an isolated entity, giving the readers the illusion of understanding but avoiding discussing any non-trivial issues that arise in practice as well as typical mistakes (this fault alone downgraded the book to below average). But when they try to program using the book, they're in for a painful surprise. Actually this book has a misleading title, this an intro to STL, not so much into Data Structures.
In the first three chapters the author delves into C++ arcana and I suspect that a student with just one (and semi-forgotten) course of C++ will be unable to understand this material no matter what. Examples are very sketchy and generally require work to make them run. the first three labs are of low quality and especially out of touch with the college reality when students that take this class barely can program in C++. They are simply unsuitable for most students after a year of C++ experience. And to add insult to injury none of tricky areas covered in the labs are explained well in the book.
Chapter 1 discusses software development. Extremely weak and superficial treatment of a pretty complex topic. It's clear that the author does not understands the subject in depth and the chapter is completely detached from reality. Any student would be better off reading the "Elements of the Program Style" or Fast project Developemnt" instead. No help of dealing with real problem, and no relevance to the rest of course. The waterfall model of software development is introduced without even mentioning alternatives. The "official" list of software development stages provided in a book is highly misleading. In practice, the phases are intermixed and the process in iterative -- often you need to return from coding to specification after discovering that the approach chosen does not work or that there is a better way to specify the task that leads to a cleaner and/or simpler solution.
Chapter 2 discusses relationship between data structures and abstract data types. Discussion of the primitive C++ data types contains some information that is more appropriate for the assembler class (like the way floating representation stores mantissa ). After that the author discusses arrays. Sparse arrays problem and the variant of storage of multidimensional arrays are completely ignored. Some author claims are suspect. For example of p.52 we read:
"One of the principles of object oriented programming is that an object should be self-contained, which means that it should carry within itself all the information necessary to describe and operate on it. Arrays violate this principle. in particular they carry neither their size nor their capacity within them..."
... " we must pass to Print() not only the array to be displayed but also its size, and this is not consistent with the aims of the object-oriented programming" (italics belongs to the Larry Nyhoff -- NNB).
It's because such nonsense I hate the books that mix OOP with data structures :-). It looks like the author does not understand that C++ implementation of arrays is not universal and in some old procedural languages like PL/1 the size of the array is passed as a hidden parameter and can be retrieved using build-n function ;-). Same is true for non-object oriented scripting languages like Perl. In those languages arrays are pretty much self-contain. Then the author tries to explain structures and fail to provide any reasonable treatment of the tickly aspects of this topic. For example is next to impossible to understand how unions work from the text and I highly recommend to get a different textbook for the topic At the end of the chapter the author extols the virtue of OOP in comparison with the procedure oriented programming although is algorithms and data structures this approach is highly suspect and might represents a contemporary religious aberration that will not survive the test of the time.
In Chapter 3 along with explanation of C++ class construct the author managed to discusses strings (without much details. A simple editor listing is provided, data encryption (DES and RSA are mentioned, and, to ensure that the student completely lost interest, the subject pattern matching is added to the mix :-).
After that the book changes the topic and became more introduction of STL library that a data structure course. As an introduction to STL it might makes sense but as an introduction to data structures probably not.
Chapter 4 discusses stacks. Chapter 5 queues. Chapter 6 is devoted to Templates and Standard containers. Chapter 7 is about ADTs.
Actually vectors are strings are explained not that badly and labs are much better than previous.
Chapters 8 and 9 discuss lists. Chapter 10 discusses binary trees. Chapter 11 discusses sorting. Chapter 12 tries to link OOP and abstract data types. Chapter 13 is devoted to trees. Chapter 14 discusses Graphs and Digraphs.
You cannot compare this text to Knuth were you really feel the class of author even if you do not understand half of the material (still you definitely feel the respect, that might help to return to the subject later in one's professional career.)
In this book a typical student probably will never understand two-thirds of the material because data structures are obscured by object-oriented arcana. And he/she will not receive anything in return other then strong desire not have anything in common with this subject for the rest of his/her professional life :-).
Less than helpful, May 13, 2000
Reviewer:
liz mccraven (see more about me) from New Jersey
I used this book as a student in an online version of a Data Structures class.
None of us liked this book. The examples are vague and some of the exercises
are misleading. I found the organization of the book to be confusing as well.
Not good for learning Data Structures on your own.
Not for the faint of heart, January 31, 2000
Reviewer:
Jim Hare (see more about me) from St Louis, MO
This book, while perhaps suitable for a more agressive Data Structures course.
Was unsuitable for most students in my college Data Structures class after a
year of C++ experience. The code presented in the text is very skeletal, with
far too many "left as an exercise to the reader" for my taste.
Paperback - 256 pages 1st edition (December 17, 2001)
Addison-Wesley Pub Co; ISBN: 020170434X
Amazon.com Sales Rank: 5,740
Filled with forty all-new working strategies and solutions, February 9,
2002 Reviewer: Midwest Book Review (see more about me) from Oregon, WI USA
Part of the outstanding Addison-Wesley "C++ In-Depth Series", Herb Sutter's More Exceptional C++: 40 New Engineering Puzzles, Programming Problems, And Solutions is the sequel to his popular and "user friendly" Exceptional C++: 47 Engineering Puzzles, Programming Problems, And Solutions (0201615622). Filled with forty all-new working strategies and solutions to solve real-world problems, More Exceptional C++ presents its valuable information in problem-and-solution format so that experienced programmers can streamline their code and improve their results. Chapters cover such topics as code optimization and so-called "lazy optimization"; the uses and pitfalls of multiple inheritance; how to properly use auto_ptr and much, much more. A valuable and informative resource for experienced C++ programmers.
Effective C++'s Big Brother, February 1, 2002 Reviewer: A reader from Sim City, CA (Somewhere in the Bay Area)
Addison Wesley's "C++ In-Depth Series" is showing an amazing consistency in producing one winner after the other. Any C++ programmer could almost blindly pick up any book in the series and learn some new valuable insights.
I like "More Exceptional C++" even more than the original. It's not clear to me whether this is because the book is better or because the subject matter has become more important to me. The "Exceptional C++" series is shaping up to be a big brother to the "Effective C++" series, covering areas somewhat more advanced than those in the Effective series, such as exceptions, templates, and namespaces.
One aspect of the book I don't particularly care for is the quizzes/points format that, I suspect, is due to the origins of the book in the author's "Guru of the Week" series.
This is a great book and should belong in every
advanced C++ programmer's personal library.
Picks Up Where The First Book Left Off, January 30, 2002
Reviewer:
Philip R. Heath (see more about me) from Plano, TX United States
More Exceptional C++ is every bit as good as the first offering from Sutter.
Like the first, this is an advanced text, and a solid working knowledge of C++
is necessary to get the most out of this book.
For those without experience with Sutter's previous book, this is divided into "Items" grouped together by broad subject area. Unless the the items make up a series such as Items 13-16, they can be read independently and in any order. This layout is helpful to the reader who doesn't have a lot of time to read a book from cover to cover. One can sit down and spend 30 minutes with an item and gain valuable insight into the specific subject matter Sutter deals with.
I enjoy the author's writing style because he tends to be more conversational than lecturing. He interjects humor - albeit it geek humor - from time to time. The presentation makes learning advanced techniques, dare I say, fun rather than dry and cumbersome.
It is also worth noting that being advanced
doesn't preclude being practical. Sutter deals with everyday topics such as
the STL, exception safety, and inheritance. If you are ready to make the step
to advanced C++ programmer, this book will guide you on your way in a
practical, enjoyable manner.
Paperback - 889 pages (September 1999)
MacMillan Computer Pub; ISBN: 0789721449 ; Dimensions (in inches): 2.09 x 9.27 x 7.42
