Date:	Wed, 14 Jun 2000 12:17:57 +0300 (IDT)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	Alain Magloire <alain AT qnx DOT com>
cc:	djgpp-workers AT delorie DOT com
Subject:	Re: tmpfile in DJGPP
In-Reply-To:	<200006132003.QAA28055@qnx.com>
Message-ID:	<Pine.SUN.3.91.1000614121726.29091J@is>
MIME-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Tue, 13 Jun 2000, Alain Magloire wrote:

> When I write code that will be run on different platforms,
> I write it with "portability" in mind, so doing fflush(stdin)
> is not "portable".

Even if a certain behavior is defined by a standard, it still makes
sense to refrain from using some features which are known to be broken
on some platforms.  The feature we are discussing is IMHO one such
case.

In other words, the standard is not the issue here; the issue is what
standard features should better not be used because not every platform
can support it, no matter how hard does it try to be Posix-compliant.

After all, standards are supposed to help portability, and
portability is NOT defined to apply only to platforms which are 100%
compliant to the standard.  If an important platfrom does not support
some feature required by a standard, then portability dictates that
this feature be avoided.

I believe that many GNU programs already include special provisions
for known deficiencies of otherwise Posix-compliant platforms.  I'm
arguing that this is one more case.

- Raw text -