[antlr-interest] ANTLRWorks hiccup?

Jean Bovet antlr-list at arizona-software.ch
Thu Dec 15 17:01:38 PST 2005


No, you are not blind... it is just that ANTLRWorks is doesn't fully  
handle pure lexer/parser/tree grammar (removing the lexer transform  
the grammar to a "combined" grammar that is currently correctly  
handled by ANTLRWorks). Note that ANTLRWorks will also report some  
syntax errors if you are using some rewrite or tree syntax (which  
doesn't prevent all the other features to work well - it's only  
esthetic).

Most of these issues will be fixed in the next release (I will  
working more on pure lexer/parser grammar in the following weeks).

Thanks for reporting the problem.

Jean Bovet

Le 15 déc. 05 à 16:50, Kay Roepke a écrit :

> Excuse if I'm being stupid, too tired, or just plain blind, but  
> shouldn't this be a valid lexer grammar:
>
> ------
> lexer grammar Lexer;
>
> START	:	ID WS+ ID
> 	;
>
> ID  :   ('a'..'z'|'A'..'Z'|'_') ('a'..'z'|'A'..'Z'|'0'..'9'|'_')*
>     ;
>
> WS	:	(' ' | '\t' | '\n' | '\r')
> 	;
>
>
> -----
>
> ANTLRWorks keeps telling me that it can't show the DFA/NFA diagram  
> for the rules, because there's no start rule to be found.
> This is certainly not the case, and ANTLR happily produces a lexer  
> for me.
> Manually deleting the "lexer" part from the first line and then  
> typing it again fixes the issue?!?
>
> The generated source (by ANTLR) seems to be ok to me...
>
> puzzled,
>
> Kay



More information about the antlr-interest mailing list