[antlr-interest] semantic differentiation

[Ulrich Frank]

Hello,
I've a new question: when I define something like

IDENT :  ( 'a'..'z' | '0'..'9' )*

in my lexer and then I have rules in my grammar like


{
         private String first = "first";
	public String getFirst() { return first; }
}
query  : { first=$FIRST.toString(); } "select" query2 ;
query2 : { first=$FIRST.toString(); } (IDENT)? query3 ;
query3 : { first=$FIRST.toString(); } "from" query4 ;
query4 : { first=$FIRST.toString(); } IDENT query5 ;
query5 : { first=$FIRST.toString(); } ("where" query6)? ;
query6 : { first=$FIRST.toString(); } IDENT query7 ;
query7 : { first=$FIRST.toString(); } ("<" | "<=" | ">" | ">=") query8 ;
query8 : { first=$FIRST.toString(); } IDENT ;


The problem I have is that I want to distinguish the IDENT in query2  
and the IDENT in query4. I have to do something in the parser, but  
what (semantic action, but how)?? Any idea? I need this because when  
I analyze $FIRST in another java class, I have to know if the IDENT I  
got from getFirst() is the IDENT from rule query2 or the IDENT from  
rule query4.

Greets, Uli
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.antlr.org/pipermail/antlr-interest/attachments/20060210/ff8d4a15/attachment-0001.html