Xref: utzoo comp.lang.misc:7976 misc.wanted:15612 comp.unix.sysv386:8859 comp.os.cpm:5306 comp.os.msdos.programmer:5569
Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!bellcore!uunet!cadence!acae037!eversole
From: eversole@acae037.cadence.com (Richard Eversole; x6239)
Newsgroups: comp.lang.misc,misc.wanted,comp.unix.sysv386,comp.os.cpm,comp.os.msdos.programmer
Subject: Re: PL/I compiler
Message-ID: <1991Jun10.163332.29768@cadence.com>
Date: 10 Jun 91 16:33:32 GMT
References: <1991Jun4.073924.2610@weyrich.UUCP> <60LR3ND@math.fu-berlin.de> <1991Jun6.144734.244@batcomputer.tn.cornell.edu> <1991Jun7.120942.3626@world.std.com> <1991Jun7.152725.819@news.larc.nasa.gov>
Sender: usenet@cadence.com (USENET News)
Reply-To: eversole@acae037.cadence.com (Richard Eversole; x6239)
Distribution: usa
Organization: Cadence Design Systems, Inc.
Lines: 27

As of 1988 the VAX VMS PL/I compiler was not a full implementation of the
language as defined by IBM.

I had a 65,000 line PL/I program for the IBM 3090 that required
15,000 lines of changes just to get it to compile under VAX PL/I.
Unfortunately, the program also required a CODASYL compliant database
which was not available for the VAX at our site at that time.

I do not remember the exact problems but they were subtle uses of 
the language. 

The only company I'm aware of that has a PL/I compiler that may be
of any value is LPI (I do not have the address). Unfortunately, I
think their compilers only run under UNIX and maybe a handful of
other operating systems (But not MESSDOS).

DR's PL/I compiler is garbage. All "Subset G" PL/I compilers are going
to be worthless if you are transferring from an IBM mainframe compiler.


-- 
  
  =====================================================================

    eversole@cadence.com
  
    Live long and prosper !