Xref: news2.mv.net comp.os.msdos.djgpp:5622
From: Eric Rudd <rudd AT cyberoptics DOT com>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Does DJGPP conform to ANSI-C standards with the for () ?
Date: 27 Jun 1996 13:11:15 GMT
Organization: CyberOptics Corp
Lines: 34
Message-ID: <4qu19j$d5v@hagar.cyberoptics.com>
References: <199606250925 DOT NAA27938 AT video DOT yars DOT free DOT net> <31D15300 DOT 129C86C6 AT laden DOT ilk DOT de>
NNTP-Posting-Host: rudd.cyberoptics.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

There have been plenty of opinions offered here, but has anyone actually 
consulted the ANSI spec?  The relevant section is 5.1.2.2.1, where it is 
explained that "the implementation defines no prototype for this function 
[main]..."  The ANSI document goes on to give an example where "main" can 
be defined

   int main(void) { /*...*/ }

and specifications for the arguments argc and argv *if they are defined*.
(Since they are local to main, they don't even have to be named argc and 
argv, but only an IOCCC entry would name them anything else.)  Since the 
prototype is not defined in the ANSI spec, one is free to declare it

   void main()

or even

   main()

wherupon it defaults to type "int".

In section 5.1.2.2.3, the ANSI document states "If the main function 
executes a return that specifies no value, the termination status 
returned to the host environment is undefined."  It is therefore implied 
that "void main" is not an error.

The Borland C++ compiler will flag void main() as an error if the -A 
(strict ANSI) switch is specified, but this is an error, since, as you 
see, the return type of main is *not specified* in the ANSI standard.

Good programming practice is another matter.

-Eric Rudd