Engineering: A Compiler

As a professor and former compiler writer, I'm a big fan (and owner) of compiler books (including those by Aho, Ullman, Goos, Waite, Wirth, ...). The last years have seen a resurgence of really good books covering modern compiler engineering. This one (first and, gladly, second edition) is by far the best in its class. The contemporary competition consists of* Aho/Sethi/Ullman/Lam: nice book, great follow-on from the earlier dragon books, but is so thick that it's tough to teach from, and to be honest, it's getting convoluted.* Allen/Kennedy: another great book, covering some of the best results in optimization (of well behaved languages like Fortran). It is, just like the latest dragon book, heavy slogging and not digestible by many students.* Muchnick: yet another excellent book, but it hasn't (AFAIK) been updated.By contrast, this book (Cooper/Torczon) is not only digestible (nice presentation, not overly terse), but it also covers new and interesting algorithms and data-structures. This applies not only to today's hot topics like optimization (and related data-structures like SSA) and code-generation, but also to front-ends. For example, the chapter on lexical analysis covers Brzozowski's minimization algorithm. (As a specialist in such minimization algorithms, it's very encouraging to see compiler writers/text book authors now embracing such an easily taught algorithm.) All in all, a very nice book on modern compiler engineering.

I grew up with the first edition of the Dragon book, a fine book for its time, especially when coupled with YACC and LEX. Lately I've been contributing to LLVM and I needed something a little more up to date as a refresher. To that purpose, Engineering A Compiler, second edition, is an outstanding modern text. It uses a pleasant+simple pseudo code for its examples as opposed to Java or C++. It's emphasis is on English text and it reads well. I'd put it on the Hennessy and Patterson shelf.

textbook came in good condition, appears to be unmarked

This book is pure theory. The title didn't convey this to me.If you haven't yet created your own compiler, then I would look elsewhere. The best bet would be too get "Brinch Hansen on Pascal Compilers", which contains a small amount of theory but heaps of code (fully commented and understandable) which you can easily digest and then modify for your custom compiler.The compilers I've written (based largely on Hansen) used the "top down" method, which can easily be coded by hand. Although a great introduction to compilers, there is not much discussion on bottom-up parsing or code optimization.The first half of the book reviewed here was a good refresher for me about compilation techniques. I could follow it easily, but I knew most of the material beforehand (from Hansen). The other part was different because it explained bottom-up parsing well. I never "got it" when reading the Dragon book or others. So, I was impressed by the first half of the book.The second half is about optimization. The topics here were either briefly mentioned in Hansen (but no implementation was given), or else were absent. Hansen's approach was to use a stack-based machine, which is simple to do but not good for optimizing code. In this new book, the authors don't use a stack-based approach, but rather a register approach. This allows for lots of types of optimization. It's heavy going. There are some diagrams, but not enough for me. Pseudo code was given to explain each optimization technique, but there were always special cases that threw a spanner in the works. I liked the constant summaries, but when I faced the questions at the end of the chapters, I quickly realized I hadn't digested the material fully! I also realized that I'd never be able to implement the optimizations from the pseudo-code presented.I learned that there is almost an infinite numbers of combinations of optimization code. This shows that there will always be areas of research in compilers. I got stuck in a lot of places, but still got a good understanding of optimization theory. Many techniques were written in acronyms to save space, but I kept forgetting what the names stood for, and that hindered my learning.For a single-semester course, Hansen's book is better, as it's practical. This new book would be a great way to consolidate your knowledge and let you prepare you for further study, or research. There are other books on optimization, but this one has enough topics and theory for me!It's a good book, but I've taken a star away because I feel it's too theoretical.

Great book. The authors are very knowledgeable on the subject of compiler design and theory. This book is structured better than the dragon book. It shouldn't be a surprise that UC Berkeley is using it in their curriculum.

(Haven't finished the book yet) Everything is well presented in a theoretical manner. While the subject matter is abstract, the author keeps the reader right on track.

Great book for a compilers class, especially upper level undergrad/new grad student. I would not have survived building my own compiler for my class had it not been for this book. I also have a copy of the dragon book but I read this first and learned from it therefore I have a bias towards it.

Great Book, use this to learn for my compiler designs course.

Wazige letters op dunne pagina's, waardoor het erg onfijn lezen is.De inhoud daarentegen is erg goed.

Comprehensive but a fair bit of maths. If you want something more hands on then go for the tiger/ML book.

Very thorough, and consistent introduction into compilers (I think it could be called more than introduction..).Goes from implementing scanners to optimization, and most efficient data structures. At each point, advantages and disadvantages of some technique X are exposedI highly recommendI have not read Dragon Book, so I can't compare.On a side note, the format is kinda odd (huge pages, but only filled half-width, very wide margins, I suppose that's for reader's notes), book is really heavy, and the "pseudo-assembly" (ILOC) used is annoying

Das Buch bietet einen guten Überblick, ohne jedoch allzu sehr in die Tiefe zu gehen. Beim Parsing hätte mir etwas mehr Praxisbezug gefallen, es werden CFGs und Attributgrammatiken verwendet und die Autoren haben offenbar ein Faible für Parsergeneratoren. In der Praxis werden solche Generatoren aber eher weniger eingesetzt, vor allem keine für Attributgrammatiken, und da wäre mir etwas mehr über Tricks für handgeschriebene Parser mit Lookahead lieber gewesen. Dafür glänzt das Buch an anderen Stellen, IR, Laufzeitstrukturen und Activation Records werden recht ausführlich erklärt, wobei Optimierungen später getrennt behandelt werden. Auf moderne Tools wie LLVM wird praktisch nicht eingegangen, es geht also vor allem um die Grundlagen und weniger den Praxisbezug, denn wer würde heute noch einen Compiler schreiben, ohne LLVM oder ähnliches zu verwenden? Alles in allem ist das Buch eine Einführung in Standardmethoden, und darin ist es gut und leicht verständlich. Allein für die Erklärung der Vorteile der SSA-Form von machinennahen internen Repräsentationen kann ich das Buch empfehlen. Die Kapitel sind in kurze, oft beinahe abgeschlossene Unterabschnitte eingeteilt, sodass das Lesen Spaß macht und auch nicht ermüdet.Fazit: Ein empfehlenswertes Buch für alle, die sich für Compiler- oder Interpreterdesign interessieren. Ich empfehle generell indische Ausgaben, sie haben etwas dünneres Papier und sind dafür aber auch viel billiger. An dem Gefasel auf dem Einband, dass der Verkauf anderswo verboten ist, ist selbstverständlich überhaupt nichts dran. Natürlich können Sie das Buch jederzeit in Indien bestellen - sonst wäre das Buch auch zu teuer.

Probablement l’un des meilleurs livres sur le sujet. Toutes les notions fondamentales y sont abordées avec une grande clarté. A la fois plus simple à appréhender et parfois plus exhaustif que le « Dragon Book » (Aho, Lam, Sethi & Ullman) je ne saurais que trop recommander ce livre aux étudiants en informatique suivant des cours en théorie des langages et compilation.