Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!spool.mu.edu!uunet!world!iecc!compilers-sender
From: toebes@sas.com (John Toebes)
Newsgroups: comp.compilers
Subject: Re: PL/1 grammar
Keywords: PL/I, parse
Message-ID: <1991May08.175108.19777@iecc.cambridge.ma.us>
Date: 8 May 91 14:35:14 GMT
Article-I.D.: iecc.1991May08.175108.19777
References: <9105061719.AA06183@eds.com>
Sender: compilers-sender@iecc.cambridge.ma.us
Reply-To: toebes@sas.com (John Toebes)
Organization: SAS Institute Inc.
Lines: 17
Approved: compilers@iecc.cambridge.ma.us

If you are going to do anything with PL/I, the definitive source is the
TSI compiler from Translation Systems Incorporated.  With the exception
of the IBM offering, all of the PL/I (Subset G) compilers that I have
encountered have been based on the TSI implementation.  As has already been
noted here, PL/I does not have any reserved words making yacc/lexx a
formidable task.  The TSI technology is well tuned to PL/I and most
importantly has all the aspects of the language layed out.

I have no association with TSI but have worked on a PL/I compiler from
the TSI code in the past.
--
========= Working for but not representing SAS Institute Inc. ============
John A. Toebes, VIII     toebes@sas.com

-- 
Send compilers articles to compilers@iecc.cambridge.ma.us or
{ima | spdcc | world}!iecc!compilers.  Meta-mail to compilers-request.