[antlr-interest] Problem with LT

Ric Klaren klaren at cs.utwente.nl
Thu Sep 23 05:24:06 PDT 2004

On Mon, Sep 20, 2004 at 10:39:02AM +0200, Xavier Décoret wrote:
> I cannot find (I kow I saw it, but the search function is broken on the 
> website) the documentation on LT()

Try and see if there's a comment describing it in the java code.

> In my parser, I have a rule like this:
> i:id EQUAL value { if (notExist(i) cout<<"Error on line 
> "<<LT(0)->getLine(); }
> where EQUAL is a lexer token.
> It creates segfault from times to times. I am not sure why I used LT(0). 
> I think I believed (at the time I wrote this grammar) that LT(0) 
> indicates the first token of the grammar rule.

LT(0) would refer to the token that has just now been read. E.g. it's not
doing what you want probably.

> Is that correct? Should I use LT(1) to refer to the EQUAL instead? Is it 
> simply a bad practice to try to get a line number for a parser rule?

Just tag the token:

i:id eq:EQUAL value { if (notExist(i) cout<<"Error on line" << eq->getLine(); }


    ---- Ric Klaren ----- j.klaren at utwente.nl ----- +31 53 4893755  ----
     "Evil will always triumph, because Good is dumb." ---  Spaceballs

Yahoo! Groups Links

<*> To visit your group on the web, go to:

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

<*> Your use of Yahoo! Groups is subject to:

More information about the antlr-interest mailing list