[antlr-interest] Re: Anyone tried this ANTLR-inspired CC?

lgcraymer lgc at mail1.jpl.nasa.gov
Sun Nov 9 02:17:41 PST 2003


Chris--

Welcome!  The performance differential is not a surprise--that's 
sort of an artifact of ANTLR 2's history.  Ter was too busy with 
jGuru to do much polishing.  That should change with ANTLR 3, which 
will be a major upgrade, both in terms of features and 
implementation.

I've looked over the LLK source, and can see that you have 
incorporated design features from several other language translation 
tools.  I'm curious:  why did you choose the LLK syntax?  It is not 
what I would have chosen.

--Loring


--- In antlr-interest at yahoogroups.com, "leung13512c" 
<leung13512c at y...> wrote:
> Hi, this is Chris Leung, author of LLK. I am just one of
> many to write open source software to kill time (somebody may
> mistaken that as xxx ms).
> 
> Thanks for all the interest in LLK.  LLK is started as I explore
> several design alternatives while I study ANTLR. It turns out
> to have some worthwhile features and improvements.  So I released
> the code and hope that would be useful for others.  I would be
> most glad if anything in LLK would be useful for ANTLR.
> 
> To be honest, I am no expert in this field. LLK is started as a 
hack
> of ANTLR. The basic algorithm is same as ANTLR.  But as more and 
more
> feature and implementation differences comes in, it is almost
> a complete rewrite. Most of the differences from ANTLR are 
explained
> in the LLK documentation at llk.sourceforge.net.
> 
> I ran some preliminary test (with GnuCParser) and found that it 
(LLK
> v0.1) is about 4-5 times faster than ANTLR 2.7.2 (in Java). But I 
> have not dig too deep into it yet. I would certainly be glad to 
> discuss any specific issues about LLK.
> 
> Thanks,
> Chris.
> 
> --- In antlr-interest at yahoogroups.com, Terence Parr <parrt at c...> 
wrote:
> > 
> > On Tuesday, November 4, 2003, at 01:26 PM, Oliver Zeigermann 
wrote:
> > 
> > > Had a quick look at it and found it interesting too. It takes 
up the
> > > idea that certain semantic actions should be executed while 
guessing.
> > >
> > > The code does not seem to be copied at all, but at most 
inspired by
> > > ANTLR. Terence, has this guy never contacted you?
> > 
> > Nope....it took me a while, but I found what I think is an email 
> > address; no name.  I emailed him last night and quickly looked 
at the 
> > source code.  I too verified that he had not just copied the 
source 
> > except for some smaller files.
> > 
> > >  Maybe it could be
> > > interesting to share ideas or experience. I invited him to 
present his
> > > ideas here on the mailing list. Let's see if he does...
> > 
> > Great! :)  Let's see if he likes the LL-regular.  I wonder who 
he is.  
> > The code showed a nontrivial understanding of the area.
> > 
> > Ter
> > --
> > Professor Comp. Sci., University of San Francisco
> > Creator, ANTLR Parser Generator, http://www.antlr.org
> > Co-founder, http://www.jguru.com
> > Co-founder, http://www.knowspam.net enjoy email again!
> > Co-founder, http://www.peerscope.com pure link sharing


 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ 




More information about the antlr-interest mailing list