[antlr-interest] [ANTLRWorks 1.4][Code Generation] [C target] Bug??

[Sam Harwell]

I'd initially write it as (b* a+ b*) and handle the error of b elements
located at both locations with a user-understandable message at a later
point. My end goal would be reducing rule1 to the following as long as it
stays unambiguous.

rule1 : (rule2 (a|b)*)*

Sam

-----Original Message-----
From: antlr-interest-bounces at antlr.org
[mailto:antlr-interest-bounces at antlr.org] On Behalf Of Kevin J. Cummings
Sent: Wednesday, December 01, 2010 10:21 AM
To: antlr-interest at antlr.org
Subject: Re: [antlr-interest] [ANTLRWorks 1.4][Code Generation] [C target]
Bug??

On 12/01/2010 10:32 AM, Amr Muhammad wrote:
> Hello,
> 
> I have a rule like this:
> 
> version1 ---  rule1: (rule2 (a+ b* | b* a+) )* (rule2 a*)*
                       ^^^^^^^^^^^^^^^^^^^^^^^^^ Hmmm, 1 or more a followed
by 0 or more b OR zero or more b followed by
1 or more a.

What happens when there are 0 b.  Your code reduces to 1 or more a OR 1 or
more a.

I would re-write this as:

rule1: (rule2 ( (a+ b*) | (b+ a+)) )* (rule2 a*)*

as the case of just matching a+ is handled by the first alternative.

Now you need to handle the problem of the which rule2 case to match given
that the first one could be empty and you want to match the second.  I think
you need to disambiguate your grammar further.

--
Kevin J. Cummings
kjchome at rcn.com
cummings at kjchome.homeip.net
cummings at kjc386.framingham.ma.us
Registered Linux User #1232 (http://counter.li.org)

List: http://www.antlr.org/mailman/listinfo/antlr-interest
Unsubscribe:
http://www.antlr.org/mailman/options/antlr-interest/your-email-address