From:	Erik Max Francis <max AT alcyone DOT com>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: Question on pointers and arrays
Date:	Mon, 24 Feb 1997 00:19:20 -0800
Organization:	Alcyone Systems
Lines:	23
Message-ID:	<33114F08.7FB7F03A@alcyone.com>
References:	<32f92a6c DOT 0 AT ntnews DOT compusmart DOT ab DOT ca> <32FA7146 DOT 3883 AT cam DOT org> <32fa7242 DOT 998097 AT news DOT walrus DOT com> <5dsgcv$5m4 AT star DOT cs DOT vu DOT nl> <3303DF62 DOT 21E0 AT cs DOT com> <5e1f1u$m07 AT star DOT cs DOT vu DOT nl> <33092376 DOT 1E3 AT cs DOT com>
NNTP-Posting-Host:	newton.alcyone.com
Mime-Version:	1.0
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

John M. Aldrich wrote:

> No, but apparently this sort of thing is system-specific.  When you do
> that with DJGPP in DOS, it works fine (although it still is very
> unsafe).

If you're looking for portability, then you're interested in strict
conformance to ANSI C (or draft standard C++).  In that case, you'd better
not do it.  Whether it works on one or more platforms is not a good
indication of whether something is good C (or C++) programming practice.

After all, some operating systems will let you dereference the null pointer
and won't complain, even though that's clearly a naughty thing and
something that you would hope (but unfortunately will not always happen)
the compiler will detect.

-- 
       Erik Max Francis, &tSftDotIotE / email:  max AT alcyone DOT com
                     Alcyone Systems /    web:  http://www.alcyone.com/max/
San Jose, California, United States /    icbm:  37 20 07 N  121 53 38 W
                                   \
              "I am become death, / destroyer of worlds."
                                 / J. Robert Oppenheimer (quoting legend)

- Raw text -