[antlr-interest] Re: problem parsing numbers

[William Lam]

Here's a grammar that works for me.

class Zoo extends Lexer;

options {
    exportVocab = Zoo;
    testLiterals = false;
    k = 2;
    caseSensitive = false;
    caseSensitiveLiterals = false;
//    charVocabulary = '\3' .. '\177';
    charVocabulary='\u0000'..'\uFFFE';
}

PLUS            :  '+'     ;
MINUS           :  '-'     ;

NUMBER
    {boolean isDecimal=false;}
    :    '.' {_ttype = DOT;}
            (('0'..'9')+ (EXPONENT)? (FLOAT_SUFFIX)? { _ttype = 
NUMBER_FLOAT; })?
    |    (  ('0'..'9')+  {isDecimal=true;}        // non-zero decimal
        )
        (   ( 'l'! ) { _ttype = NUMBER_LONG; } // strips the L/l
        
        // only check to see if it's a float if looks like decimal so 
far
        |   {isDecimal}?
            (    '.' ('0'..'9')* (EXPONENT)? (FLOAT_SUFFIX)?
            |    EXPONENT (FLOAT_SUFFIX)?
            |    FLOAT_SUFFIX
            )
            { _ttype = NUMBER_FLOAT; }
        )?
    ;

/**
 * From java.g
 */
WS  :   (   ' '
        |   '\t'
        |   '\f'
        |   ( "\r\n"
            | '\r'
            | '\n'
            )
            { newline(); }
        )
        {$setType(Token.SKIP);}
    ;

// a couple protected methods to assist in matching floating point 
numbers
protected
EXPONENT
    :    ('e') ('+'|'-')? ('0'..'9')+
    ;

protected
FLOAT_SUFFIX
    :    'f'|'d'
    ;

protected
N : '0' .. '9' ( '0' .. '9' )* ;


--- In antlr-interest at yahoogroups.com, "isabelle_muszynski" 
<isabelle_muszynski at y...> wrote:
> Hi everyone,
> 
> A typical newbie question: the following grammar won't parse 
numbers 
> correctly (for ex. give 123.456). I have simplified the grammar to 
> the extreme to show only the number problem.
> 
> class ExpressionLexer extends Lexer;
> options {
>     k = 2; // needed for newline stuff
>     filter = true; 
>     charVocabulary='\3'..'\177';  // allow ASCII
> }
> 
> tokens {
>     LONG;
> }
>  
> PLUS   : '+' ;
> MINUS  : '-' ;
> 
> protected
> DIGIT
> 	:	'0'..'9'
> 	;
> 
> protected
> EXPONENT
> 	:	('e' | 'E') ('+' | '-')? (DIGIT)+
> 	;
> 
> NUMBER
>     :   ( (DIGIT)+ ('.' | 'e' | 'E') )=> (DIGIT)+ ( '.' (DIGIT)* 
> (EXPONENT)? | EXPONENT	)
> 	|	'.'	(DIGIT)+ (EXPONENT)?
> 	|	'0' ('0'..'7')* {_ttype = LONG;}
> 	|	'1'..'9' (DIGIT)* {_ttype = LONG;}
> 	|	'0' ('x' | 'X') ('a'..'f' | 'A'..'F' | DIGIT)+ 
> {_ttype = LONG;}
> 	;
> 
> // Whitespace -- ignored
> WS	:	(	' '
> 		|	'\t'
> 		|	'\f'
> 			// handle newlines
> 		|	(	options {generateAmbigWarnings=false;}
> 			:	"\r\n"  // Evil DOS
> 			|	'\r'    // Macintosh
> 			|	'\n'    // Unix (the right way)
> 			)
> 			{ newline(); }
> 		)+
> 		{ _ttype = Token.SKIP; }
> 	;
> 
> {import java.lang.Math;}
> class ExpressionParser extends Parser;
> options 
> { 
>     buildAST=false; 
> 	exportVocab=Eval;
> }
> 
> statement returns [double r = 0]
>     : r=constant
>     ;
> 
> constant returns [double r = 0]
> 	:	n:NUMBER {r=Double.parseDouble(n.getText());}
> 	|	l:LONG {r=Long.parseLong(l.getText());}
> 	;


 

Yahoo! Groups Links

To visit your group on the web, go to:
 http://groups.yahoo.com/group/antlr-interest/

To unsubscribe from this group, send an email to:
 antlr-interest-unsubscribe at yahoogroups.com

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