Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!rpi!crdgw1!camelback!volpe
From: volpe@camelback.crd.ge.com (Christopher R Volpe)
Newsgroups: comp.lang.c
Subject: Re: LL(1) C grammar
Message-ID: <20726@crdgw1.crd.ge.com>
Date: 19 Jun 91 12:17:48 GMT
References: <91-06-021@comp.compilers> <18959@prometheus.megatest.UUCP>
Sender: news@crdgw1.crd.ge.com
Reply-To: volpe@camelback.crd.ge.com (Christopher R Volpe)
Lines: 18

In article <18959@prometheus.megatest.UUCP>, djones@megatest.UUCP (Dave
Jones) writes:
|>Huh? There's nothing that can be done with an LL parser
|>that can't be done with an LR parser. What do you mean by "interpretation
|>versus compilation, incremental parser building etc"?
          
Isn't is just much more convenient to deal with a LR grammar? It's true
that any LR(1) grammar can be converted to an equivalent LL(1) grammar 
that accepts (generates) the same language, but (to me) they are very unnatural
looking and it's harder to tell what's going on. I've LL(1)-ized
grammars before
and the resulting productions don't look much like anything one could associate
meaningful and intuitive action routines to. Am I missing something?

==================
Chris Volpe
G.E. Corporate R&D
volpecr@crd.ge.com