Question

I have a Token OR:’OR’; that I use for evaluating a boolean expression( a==b OR a==c) I have another rule for parsing state abbreviation that are in a char list AZ,AK,OR,GA…
What I am finding is that antlr has error on the state list thinking OR should be an or token rather then

stateName   
    : CHAR CHAR (','|EOF) ->^(STATE CHAR+)
    ;   

how would I go about resolving this ambiguity?

here are some of the rules I am trying to parse

• THEN STATE_LICENSE
  AL,AK,AS,AZ,AR,CT,DE,DC,FM,FL,GA,GU,HI,ID,IL,IN,IA,KS,KY,LA,ME,MH,MD,MA,MI,MN,MS,MO,MT,NE,NV,NH,NJ,NM,NY,NC,ND,MP,OH,OK,OR,PW,PA
• IF 1198 == “x” OR 1190 != “x” THEN DISABLE 800
• IF 801 >= “1000000” THEN DISPLAY_ERROR “It’s+too+expensive.+Go+and+get+cheaper+one+!!!”

Here is the gramar that I am using

grammar PointFieldRule;


options 
{
//language = 'CSharp3'; 
output=AST; 
ASTLabelType=CommonTree;
} 
tokens{
STATE;

}


rule : ifExpression?  actionExpression EOF!
;
ifExpression 
    :'IF'! logicalConditionExpression
    ;

logicalConditionExpression
    : booleanAndConditionExpression ( BigOR^ booleanAndConditionExpression)*
    ;


booleanAndConditionExpression
    : logicalCondition ( BigAND^ logicalCondition )*
    ;

BigAND : 'and'|'AND';

logicalCondition
    : booleanAndCondition ( OR^ booleanAndCondition )*
    ;

OR:'||';

booleanAndCondition
    : evalCondition ( AND^ evalCondition)*
    ;

AND: '&&';

evalCondition 
    : FieldID OPERATOR^ (FieldID|STRING)
    ;

 actionExpression 
    : 'THEN'! (actionMessage | fieldAction | stateAction )  
    ;

actionMessage 
    : ('DISPLAY_WARNING' | 'DISPLAY_ERROR')^ STRING
    ;

fieldAction 
    : ('DISABLE' | 'REQUIRED')^ FieldID ( ','! FieldID )* 
    ;


stateAction
    : 'STATE_LICENSE'^ stateName+ //(','! stateName)*
    ;

FieldID
    :'0'..'9'+;

 /* item : FIELD 
    | CHAR CHAR
    ;
*/
//class csharpTestLexer extends Lexer; 

stateName   
    : CHAR CHAR (','|EOF) ->^(STATE CHAR+)
    ;   
CHAR:  ('a'..'z'|'A'..'Z')
    ;

WS : (' ' 
| '\t' 
| '\n' 
| '\r') 
{ $channel = HIDDEN; } 
//{ $channel = Hidden; } 
; 



OPERATOR 
    :   '=='
    | '!='
    |    '<='
    |    '>='
    |    '<'
    |    '>'
    | 'TD'
    | 'FD'
    | 'PD'
    | 'TY'
    | 'LY'
    | 'TM'
    | 'LM'
    | '+(DELTA%)>'
    | '-(DELTA%)>'
    | '+(DELTA)>'
    | '-(DELTA)>'
    | 'LIKE'
    ;


STRING
    :  '"' ( ESC_SEQ | ~('\\'|'"') )* '"'
    ;


fragment
HEX_DIGIT : ('0'..'9'|'a'..'f'|'A'..'F') ;

fragment
ESC_SEQ
    :   '\\' ('b'|'t'|'n'|'f'|'r'|'\"'|'\''|'\\')
    |   UNICODE_ESC
    |   OCTAL_ESC
    ;

fragment
OCTAL_ESC
    :   '\\' ('0'..'3') ('0'..'7') ('0'..'7')
    |   '\\' ('0'..'7') ('0'..'7')
    |   '\\' ('0'..'7')
    ;

fragment
UNICODE_ESC
    :   '\\' 'u' HEX_DIGIT HEX_DIGIT HEX_DIGIT HEX_DIGIT
    ;

//fragment
BigOR: 'or'|'OR';

Answer 1

The lexer creates tokens independently from the parser. So it doesn’t matter if the parser might “need” two CHAR tokens at a given point, if the lexer “sees” the text "OR", it will always create a BigOR token. There’s nothing you can do about that.

In your case, you can simply let stateName match two CHAR tokens, or a single OR token like this:

stateName   
 : name (','|EOF) ->^(STATE name)
 ;   

name
 : CHAR CHAR
 | BigOR
 ;

Parsing input "THEN STATE_LICENSE AL,OR,PA" will result in the following AST:

Note that the OR is a single token, unlike the others, whose type are CHAR and have their chars separated. If you want your OR node to behave like that too, do something like this:

name
 : CHAR CHAR
 | BigOR     -> CHAR[""+$BigOR.text.charAt(0)] CHAR[""+$BigOR.text.charAt(1)]
 ;

resulting in:

Or if you want the two separate chars to be concatenated, do:

name
 : (CHAR CHAR | BigOR) -> CHAR[$text]
 ;

resulting in: