[antlr-interest] Wrapping my head around ANTLR v3

[Matt Holmes]

First off, let me say hello. This is my first message to the list. I 
recently discovered ANTLR as I was looking for tools to help my 
implement a scripting language in a C#/.NET project I am working on. I 
started writing my lexer and parser with v2, when I realized that v3 was 
available at seemed stable enough to use.

The first thing I noticed is that the lexer and parser in all of the v3 
examples seemed to be combined in a single grammar file now. Is this a 
required situation, or can I split them across multiple grammar files to 
keep the grammar clean and easy to read? Is there any particular reason 
to keep them combined? Perhaps I am too "attached to the old school", 
the ole' yacc/lex days when you not only put them in separate files, you 
used separate tools, hehe. Perhaps I am over analyzing this and keeping 
them in a single grammar file is not that big an issue, I am just 
curious if it's still possible.

Also, I noticed that in many grammars, tokens are no longer defined 
(such as tokens { FOR="for"; WHILE="while"; } ), but instead single 
parser rules are used to define those tokens (e.g. FOR: 'for'; ). Is 
there a technical reason behind this? Is it simply a conscious choice by 
the developer, or is something gained by using this method?

Apologies if this are extremely novice questions, I am just trying to 
understand some of the things that seem "odd" to me as a past user of 
older lexical tools.

Thanks,

Matt Holmes