Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!cmcl2!rutgers!sri-spam!ames!sdcsvax!ucbvax!decvax!ima!johnl
From: johnl@ima.istc.sri.com
Newsgroups: comp.compilers
Subject: Re: error recovery in yacc
Message-ID: <706@ima.ISC.COM>
Date: Fri, 4-Sep-87 10:42:31 EDT
Article-I.D.: ima.706
Posted: Fri Sep  4 10:42:31 1987
Date-Received: Fri, 11-Sep-87 07:23:56 EDT
References: <689@ima.ISC.COM>
Sender: johnl@ima.ISC.COM
Reply-To: harvard!seismo!sun!academ!nuchat!steve (Steve Nuchia)
Organization: Public Access - Houston, Tx
Lines: 33
Approved: compilers@ima.UUCP
Summary: error recovery for binary licensed yacc

In article <689@ima.ISC.COM>, gupta%cs.unc.edu@RELAY.CS.NET (Gopal Gupta) writes:
> About 1-2 years ago someone at UNC-Chapel Hill did a study
> of error recovery mechanism in YACC. The paper is titled
> "Error Recovery in LR Parsing: A Case Study Using YACC"

For those interrested in coaxing acceptable error handling out
of yacc without being able to hack the source or having to
digest academic papers (not that it would hurt you...) there
is hope.

_Introduction_to_Compiler_Construction_with_Unix_
Schreiner and Friedman, Prentice-Hall, ISBN 0-13-474396-2

This is a usable tutorial introduction to getting results from
lex/yacc.  I bought it to learn the tools, since they had resisted
attack using just the unix manuals.  The book is organized around
the evolution of a C subset compiler (roughly small-C) which I have
since extended and retargeted and I'm using it to generate code
for my architecture simulation.

Anyway, they include simple yacc source construction rules for
error recovery and a simple modification to /usr/lib/yaccpar
that you can make at home.  The result is truly usable error
handling, better than a lot of commercial compilers.
-- 
Steve Nuchia
{soma,academ}!uhnix1!nuchat!steve
(713) 334 6720
--
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