Compiler Design Theory (The Systems programming series)

Rating:
Summary: The good starting point, is embracing, neat and actual.
Review: This book is a good starting point for anyone who needs to create a compiler, parser or scanner, but didn't read anything about compiler design theory yet. This book is completely self-contained and assumes only the familiarity with programming languages and the mathematical sophistication commonly found in juniors or seniors.

The material in this book has been taught for several years in one-semester first-year graduate courses at Rensselaer Polytechnic Institute in Troy, N.Y., the State University of New York at Albany, and many other institutions. The bewildered looks of students in these institutions have motivated the authors to do several rewrites of the materials. That's why the book is pleasant to read and easy to understand.

The book contains a good introduction to state machines and all modern grammars, including, but not limited to, LALR(1).

Since the book doesn't cover code generation, but only lexical and syntax part, it is still actual since published 1976. For example, the most popular nowadays LALR(1) grammar, supported by Yacc and Bison, is a core of the GNU compiler and many other commercial compilers. But the book itself is tool-free, it explains the background that never expires, rather than bothering the reader with the tools which may exist today and vanish tomorrow. The example of such book with a mess of theory and tools is "Compilers: Principles, Techniques, and Tools" by by Alfred V. Aho.

The book "Compiler Design Theory" is embracing, neat and actual.

Rating:
Summary: great book
Review: This book is the perfect storm of compiler theory books. Forget about the Dragon Book and pick this one up used for 10% of the "D" book's cost. Lewis et al lead you from an exploration of FSMs through PDMs(both D and ND)to CFGs,Top Down(LL(k)) and Bottom Up(LR(k)) grammars.The first 13 chapters of the book are worth the cost alone. Unlike other compiler books this one gives you a deep understanding of how to specify and develop a compiler.They don't gloss over things because they'll be using automated tools(LEX,YACC,etc.) to actually develop model compilers. You're down on "bare metal" here and you'll be the one specifying LL(1) Selection Sets or LR(1) Push Tables.
In short, if you're looking to obtain a detailed theoretical and practical understanding of Compiler design this is the book for you.