Book Review: C Network Programming, Volume 2 by Matthew Wilson
Title: C NetWork Programming, Volume 2: Systematic Reuse With Ace and Frameworksauthors: Douglas C. Schmidt and Stephen D. Huston Publisher: Addison Wesley Professional, 2002 Pages: 384 Price: $ 39.99
中文 书 Title: "C Network Programming: System Chemical Reuse Based on ACE and Frames" Translator: Mavida http://www.flyingdonkey.com/ Publishing House: Huazhong University of Science and Technology Publishing House Published: To be determined (after Diction) Price: to be determined
By the Douglas Schmidt and Stephen Huston written by C Network Programming, Volume 2: Systematic Reuse with ACE and Frameworks, their C Network Programming, Volume 1: Mastering Complexity with ACE and Patterns of the sequel. Volume 1 focuses on the low-level, specific OS network primitive package and abstraction into the ACE OS adapter layer and C Wrapper Facade (packaging appearance). Volume 2 is constructed on this basis, and the various points of the framework are processed, and how to use ACE for system implementation - this is done by enhancing and expanding the log service examples described in volume 1.
Chapter 1 acts as a book, and an introduction to the mode / framework method used by the ACE. The progress of this chapter is difficult and slow until the two authors began to discuss some details; my mental fog is raised, I can't wait to read Chapter 3.
Chapter 2 continues to introduce some basic concepts by comprehensively covering services, servers, and some detail issues. The service is characterized for a short duration of VS. Duration, the internal VS. external, stateful VS. stateless, hierarchical VS. integral. The author gave the server and the configuration to be similar in depth. The progress in this chapter also has a little difficult, but it is also interesting and useful.
From Chapter 3, it is the actual content section of this book, covering the various frameworks in ACE. From this time, I began to feel the true interest. Each frame-including Reactor, Service Configurator, Task, Acceptor-Connector, and proactor (front-sleeper) - has received comprehensive Discussion, and combined with the problems that need to be resolved in the real world, the design of the framework, and the necessary constraints or shortcomings of portability and efficiency.
The best chapter is the last chapter: Chapter 9 on the Ace Streams framework. By demosing how to use the ACE_TASK, ACE_MODULE and ACE_STREAM classes, this chapter shows the effectiveness and strength of the ACE. This is a powerful, almost Plug-Play network technology, which helps to answer any questions we may have, such as whether it is necessary to invest, not too insignificant, to learn ACE.
When discussing various OS, the two authors have no bias, which makes me feel happy and fresh (on this issue, the Timer class, select (), and the description of the ACE_WFMO_REACTOR class). Contact to the method of realizing the effectiveness of the focus of ACE and the description of it, this is greatly attractive and shows the author's confidence: they have created a high quality and independent. The solution for the platform. In general (although not all), this book is very good. The author uses UML a lot of UML, which promotes the coherence of this book (although this has made me rushed to my bookshelf multiple times). In terms of disadvantages, the font seems to be too big, especially in the figure, and in the extracted code; this is the first time I read such a book, it seems to be "explosive" from many cracks. Although the author (or their editor) is stunning the code does not cross the page, I still hope to get more things (of course, books cannot be like a 1600 * 1200 screen, but if you use a smaller font, you can print More code.)
There are quite a few (61) sub-bars in the book, which offer delightful, small information fragments. However, this is also a distinction. Because they have so much, and more easily read more than the text, they are too easy to let people throw away. Oh, you'd better like to stay with Reify (let ...) this word, because I have never seen anything in this book.
There is also a problem in the books, which always emphasize a basic view of this book, that is, the reuse is good, ACE is driven. This is not necessary, because the first two chapters have clearly expressed this view.
The index is very good, the reference is impressive; my book "Wish List" has doubled. The glossary is very good. All books should have such a glossary!
It is very strange, although this book is published by Addison-Wesley, there are some charts in the books. For example, the line segments between the boxes and boxes have just recognized it. Moreover, there are also some errors: such as the palette header starting with "ACE_SERVICE_USCOREHANDAL", should be "ace_service_handler". I hope that I have only a gift version, and your book will not have a similar problem.
In general, it is impossible to think that it is impossible to leave this book (if this is not a proud). I have found that I have a strong wish, I want more (re-) reading books about the model (more don't say those valuable wealth in the reference), and want to rush out to use ACE to start developing a distribution Project. I expect to test the ACE in real environments: Great income in productivity, and no sacrificing flexibility and (especially) efficiency. Quoted in Chapter 1 Summary: "Continuous redishes and inventions ... make the expense ... too long, it has always been high."
I want to say that this book and volume 1 are useful for anyone who wants to master distributed network programming - whether they are ultimately using ACE. (And if the ACE claims is that it is true, I think ACE should be the preferred development library, especially in the case of multiple hardware platforms). It is not a book of beginners, not particularly easy to read (very different from Meyers / Sutter / Dewhurst's easy digestive style), but it is rich, interesting, and does not give people an impression of any important content. However, it seems to rely on a large number of preparatory knowledge, especially about the knowledge of the model method, so first read the Design Pattern [1] and Pattern-Oriented Software Architecture [2] will be smart. Note note
[1] Erich Gamma, Richard Helm, Ralph Johnson, And John Vlissides. Design Patterns (Addison-Wesley, 1995).
[2] Douglas C. Schmidt. Pattern-Oriented Software Architecture, Volume 2: Patterns for Concurrent and NetWorked Objects (John Wiley & Sons, 2000).
About author
Matthew Wilson holds a degree in Information Technology and a PhD in Electrical Engineering and is a software development consultant for Synesis Software. Matthew's work interests are in writting bullet-proof real-time, GUI, and software-analysis software in C, C , and .
