Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!rutgers!ucla-cs!zen!ucbvax!decvax!ima!johnl
From: johnl@ima.UUCP
Newsgroups: comp.compilers
Subject: Re: recursive-descent error recovery
Message-ID: <664@ima.ISC.COM>
Date: Sun, 16-Aug-87 12:37:29 EDT
Article-I.D.: ima.664
Posted: Sun Aug 16 12:37:29 1987
Date-Received: Wed, 19-Aug-87 01:26:15 EDT
References: <634@ima.ISC.COM> <642@ima.ISC.COM> <651@ima.ISC.COM> <655@ima.ISC.COM>
Sender: johnl@ima.ISC.COM
Reply-To: scott@cs.rochester.edu (Michael Scott)
Organization: U of Rochester, CS Dept, Rochester, NY
Lines: 22
Approved: compilers@ima.UUCP


In article <655@ima.ISC.COM> chuck@amdahl.amdahl.com (Charles Simmons) writes:
| I claim it is impossible to write a reasonable compiler
| using YACC.  I base this claim on the fact that ... it is
| impossible to produce reasonable error messages using YACC.

Well, it's certainly difficult, but that's not because YACC is a
bottom-up parser.  It's because you're trying to do the error recovery
by hand.  Bernard Dion's thesis work at Wisconsin (with Charlie
Fischer) demonstrated how to perform locally least-cost *automatic*
error *correction* in bottom-up parsers, with modest space and
negligible time overhead.  [Compiler writer gives the parser generator
a table of insertion and deletion costs for tokens; error corrector
makes the least cost set of insertions and deletions that allows one
more real token to be shifted.]  Good quality corrections make pretty
good error messages.  Dion's techniques are incorporated in the ECP
(error correcting parser) package distributed by Fischer's group.
-- 
Michael L. Scott
University of Rochester    (716) 275-7745
scott@cs.rochester.edu     scott%rochester@CSNET-RELAY
{decvax, allegra, seismo, cmcl2}!rochester!scott

--
Send compilers articles to ima!compilers or, in a pinch, to Levine@YALE.ARPA
Plausible paths are { ihnp4 | decvax | cbosgd | harvard | yale | cca}!ima
Please send responses to the originator of the message -- I cannot forward
mail accidentally sent back to compilers.  Meta-mail to ima!compilers-request