[antlr-interest] Re: Syntactic predicates and ()+ rules

[Eric Mahurin]

Try this instead:

expr: exp1 ( (OP ~(NL|SPC)) => OP exp1 )* ;

antlr complains that the syntactic predicate is redundant, but this
warning is a bug.  Ignore it and you should be good.  If you look back
in this yahoo group you'll find some of my postings on the topic and a
little wrapper around antlr to remove these bogus messages.

Eric

--- In antlr-interest at yahoogroups.com, Joan Pujol <joanpujol at g...> wrote:
> Hi,
> 
> After a lot of time I have found my problems with indeterminism and
> syntactic predicates ;)
> But I have a question:
> 
> I have a grammar like
> 
> inst: instx (NL|SPC);
> instX:  ... expr OP;
> 
> expr: exp1
>         (
>            (OP ~(NL|SPC)) =>  OP exp1
>            |
>          )
> 
> And it works well and it has no indeterminism. I'm lucky because the
> OP can't be associated in form      exp1 OP exp1 OP exp1 OP (is a
> comparison operator)
> 
> But my question is If I suppose that exp1 OP exp1 OP exp1 is valid
> and then  I need to do:
>  
> expr: exp1
>         (
>            (OP ~(NL|SPC)) =>  (OP exp1)+
>            |
>          )
> 
> This doesn't work, because the syntactic predicate is only evaluated
> once and when I enter in the ()+ block I don't solve the
> indeterminism.
> My question how can I manage this If sometimes i need a grammar like
> this. I tried to refactor, but with no exit.
> 
> A lot of thanks,
> 
> 
> -- 
> Joan Jesús Pujol Espinar



 
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/