Return-Path: samiam@pacbell.net Date: Tue, 17 Dec 1996 22:30:13 -0800 (PST) To: J.J.vanderHeijden@student.utwente.nl Newsgroups:

comp.lang.pascal.misc,comp.lang.pascal.borland,comp.lang.pascal.ansi-iso,gnu .gcc.announce

Subject: Re: ANNOUNCE: GNU Pascal 2.0 (2.7.2.1) From: samiam@pacbell.net (Scott A. Moore) X-Newsreader: WinVN 0.99.2 Distribution: world Keywords: GNU,Pascal,Free Software,FreeWare References: gnusenet32b12cac.763542@news.student.utwente.nl

In article gnusenet32b12cac.763542@news.student.utwente.nl,

J.J.vanderHeijden@student.utwente.nl says...

...

This is the announcement of version 2.0 of

 GNU Pascal 

corresponding to GCC version 2.7.2.1.

What is GNU Pascal?

GNU Pascal is, as the name says, the Pascal compiler from the GNU compiler family. This means:

• 32 bit compiler, no limits, highly optimizing,
• runs on all operating systems supported by GCC (including DOS, Win95/NT, OS/2, Linux, FreeBSD, many other Unix dialects),
• Free Software according to the GNU General Public License,
• compatible to other GNU languages and tools such as GNU C and the GNU debugger.

The compiler integrates the following language standards:

• ISO 7185 Standard Pascal,
• ISO 10206 Extended Pascal (90% implemented),
• Borland Pascal (80% compatible).

I know this is going to start something of a fight, but when I last checked, this compiler did not in fact comply with ISO 7185 (the original Pascal standard). I know that you folks put a lot of free time into this compiler, and that is appreciated, but I also believe that it hurts everyone and undermines the standard to falsely claim compliance with it.

To me, the difficulty lies in the fact that there are no real compliance tests for standard Pascal. The BSO had a PAID one, but the problem with that is that to validate a compiler maker's claim to obey that standard required a >$1000 investment, hardly within the means of the average user. The only other way such a test can work is if the originators would have taken issue with false claims of compliance. The BSO didn't do that, and to this day at least one "standard" pascal is shipped and advertised falsely claiming to have passed it (Microway).

For the obvious question, yes, I have pointed out the standards compliance difficulties to the originators of GCC Pascal, and offered to spend my time free of charge to help find compliance problems with the compiler (and I still do).

What say we get this compiler the final few steps to meet the minimum standard, or stop claiming that it is standard. Obviously I prefer the former. Otherwise, when the above statement (IMHO false) is published, I feel it is my duty to provide equally public opposition.

                                               [sam]

-- "Nothing shocks me, I'm a scientist", Indiana Jones