[antlr-interest] Re: lexical nondeterminism

[howardk at fatdog.com]

Another nice solution! 

I'm working on an XQuery grammar, but the snippet causing problems is 
from the Namespaces specification and could easily show up as part of 
almost any XML-related grammar I'd think. What's yours (if you don't 
mind saying)?

Howard

--- In antlr-interest at y..., "John Merrells" <merrells at y...> wrote:
> > QName: (NCName COLON)? NCName;
> 
> Howard,
> 
> I have the same grammar... we must be working on the same
> problem ;-) I solved this problem by having the QName rule
> in the parser instead of the Lexer. This removes the ambiguity
> with k==2.
> 
> John



 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/