Path: utzoo!yunexus!ists!jarvis.csri.toronto.edu!mailrus!cornell!uw-beaver!ubc-cs!alberta!calgary!cpsc!aycock
From: aycock@cpsc.ucalgary.ca (John Aycock)
Newsgroups: comp.lang.modula2
Subject: Re: Modula-2 compiler for UNIX
Summary: has needed features, some oddities
Message-ID: <2078@cs-spool.calgary.UUCP>
Date: 15 Nov 89 01:32:54 GMT
Article-I.D.: cs-spool.2078
References: <1726@quiche.cs.mcgill.ca>
Sender: news@calgary.UUCP
Lines: 22

In article <1726@quiche.cs.mcgill.ca>, sami@quiche.cs.mcgill.ca (Sami BOULOS) writes:
>  I am seeking opinions on Modula-2 compilers for our SUN UNIX system.
>  Does anyone have any experience with the SUN Modula-2 compiler? We 
>  require that the compiler support both calls from Modula-2 to C and vice
>  versa. The compiler should also be able to support tools such as dbx, and
>  easily interface to window systems such as X Windows. 

I'm a computer science student at the University of Calgary, and we've had to
use Sun's version of M2 for the sun4.  It supports interfacing to C, through
a type of definition module "DEFINITION FOR C MODULE <name>", and generates
dbx information through a -g command line option.  It has a few quirks, 
however; an implementation module is required for *all* definition modules,
regardless of whether one is actually needed.  The compiler also does not
seem to have any error recovery... I haven't found a non-fatal error to date.
And, if you use variant records, it does not store the variant fields in a
common space, instead storing ALL the fields ALL the time.  However, it does
provide an adequate interface to UNIX, all things considered.

::: John

  //\|_||\|	...{alberta|ubc-cs|utai}!calgary!aycock
\/ \/| || |