[antlr-interest] probably too obvious to see it - plz help

[mzukowski at yci.com]

By the way, the reason is the optional 'n' and antlr's linear approximate
lookahead.  

Monty

-----Original Message-----
From: mzukowski at yci.com [mailto:mzukowski at yci.com] 
Sent: Thursday, July 31, 2003 4:18 PM
To: antlr-interest at yahoogroups.com
Subject: RE: [antlr-interest] probably too obvious to see it - plz help


So if it sees "n" which rule is it supposed to choose?  You need to handle
the two possibilities prefixed with 'n' in one rule.

Monty

-----Original Message-----
From: tdjastrzebski [mailto:tdjastrzebski at yahoo.com] 
Sent: Thursday, July 31, 2003 4:14 PM
To: antlr-interest at yahoogroups.com
Subject: [antlr-interest] probably too obvious to see it - plz help


I ma probably too tireed to see it. Why following two lexer rules 
cause nondeterminism (k=2).

Identifier
	: ('a'..'z')+
	;
StringLiteral
	: ('n')? '\'' (~'\'')* '\''
	;

StringLiteral rule is supposed to match unicode strings like: 
n'blabla'

Thanks in advance.
Tom Jastrzebski

full test grammar:

options {
	language = "Java";
}

class TestParser extends Parser;
options {
	k = 2;
}

expression
	: (Identifier | StringLiteral) EOF
	;

class TestLexer extends Lexer;
options {
    k = 2;
}

Identifier
	: ('a'..'z')+
	;

StringLiteral
	: ('n')? '\'' (~'\'')* '\''
	;


 

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


 

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


 

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