[antlr-interest] rule, that consider token, follow the match token

[Andrei Plet]

Bart;

Thanks for fast response.
Your solution works fine.

Thanks and kind regards
Andrei

Von: Bart Kiers [mailto:bkiers at gmail.com]
Gesendet: Donnerstag, 28. Juni 2012 13:47
An: Andrei Plet
Cc: antlr-interest at antlr.org
Betreff: Re: [antlr-interest] rule, that consider token, follow the match token


On Thu, Jun 28, 2012 at 1:20 PM, Andrei Plet <andrei.plet at empolis.com<mailto:andrei.plet at empolis.com>> wrote:
...
OP_AND : ('A'|'a') ('N'|'n') ('D'|'d');
TEXT   : ('a'..'z'|'A'..'Z')+;
PLUS   : '+';

The rule OP_AND should match only, if after the input 'and' isn't a '+', so

'and'  --> should match rule OP_AND
'and+' --> should match 2 rules: the rule TEXT 'and' and rule PLUS '+'

I wouldn't do that in the lexer, but something like this instead:

OP_AND : ('A'|'a') ('N'|'n') ('D'|'d');
TEXT   : ('a'..'z'|'A'..'Z')+;
PLUS   : '+';

any_text : OP_AND | TEXT;
add_expr : any_text PLUS any_text;

Regards,

Bart.