Which part of the dragon book is outdated? [closed]

Question

I've encountered this review, where the reviewer says:

Given multi processors and the now proven success of LH parsing and much more, much of the old dragon series is out of date and Modern Compiler Design (MCD) is well positioned to fill the void, along with Cooper.

Given that the review is very long and detailed, I assume he is knowledgeable about the subject. So specifically, in which part of the dragon book is outdated?

(What is LH parsing anyway? I tried googling but couldn't find anything relevant. Maybe a typo?)

Answer 1

I've been building parser and program analysis/transformation rules for 40+ years. The Dragon book is really good at describing the basics of compiling. There's little reason to believe it has suddenly gotten less good.

Regarding the review: I've never heard of "LH" parsing, and I think I keep track. (Really good parsers these days are "GLR"). If your quote is accurate, I wouldn't put a lot of stock in it. (This is on the Springer page that is trying to "sell" the book; looks like marketing fluff).

There are other compiler books (yes, including Torczon/Cooper) that are really good and go into topics beyond what the Dragon Book covers. But complementing isn't the same as replacing. You'd be foolish to decide not to use it, if you do anything serious in compiling. My solution: just buy them all (if you are building compilers you are going to have a hard time convincing me you don't have the money); most have something useful to say that the other ones don't. Even the reviewed book is likely to be pretty good; Dick Grune is one of the editors, and he reallys knows his parsing stuff, and has a good rep in the academic space on compilers.

If you insist only buying one, the Dragon book is still pretty good.