Question

I am wrting a small csharp console application which reads text from console, do some some manipulations and returns a string. For this, I am using Antlr. The grammar file is shown below.

    grammar test;
options {

  language = CSharp2;
  output = AST;
}
start returns [String res]: expression EOF
{ 
$res=$expression.res;  
} ;

expression returns [String res]
: Identifier  {$res=$Identifier.text}
|Num {$res=$num.text;
|function {$res=function.res}
;

function: 'left'  '( Identifier ')'{some code here}
        | 'right' '( Identifier ')'{some code here}
        |..........
         ;
Num : (Minus)?('0'..'9')+   ;
Identifier  :  ('a'..'z'|'A'..'Z'|'\\'|'/'|'_'|':'|';'|'?'|'.'|'0'..'9')('a'..'z'|'A'..'Z'|'\\'|'/'|'_'|':'|';'|'.'|'?'|'0'..'9')*; 

I have several such functions which do some string manipulations. Now, I want Antlr to identify these function names irrespective of case. At present, it accepts only lower case letters as function names like.. upper(asdf). I cannot convert every token to lower case in my application as it changes the case of Identifiers also. How can I achieve this ?

Answer 1

Simply define the appropriate token. So, for the code above:

function : Left  '(' Identifier ')' {some code here}
         | ..........
         ;

Left : ('l'|'L')('e'|'E')('f'|'F')('t'|'T');

or, if you are sure you want have cases like lEfT:

function : Left  '(' Identifier ')' {some code here}
         | ..........
         ;

Left : ('left'|'LEFT');