From: Alain Magloire <alainm AT rcsm DOT ece DOT mcgill DOT ca>
Message-Id: <199904070259.WAA10720@spock2.ECE.McGill.CA>
Subject: Re: v2.03 release: what else has to be done?
To: djgpp-workers AT delorie DOT com
Date: Tue, 6 Apr 1999 22:59:55 -0400 (EDT)
In-Reply-To: <7aa6585.243c053e@aol.com> from "Kbwms@aol.com" at Apr 6, 99 08:47:58 pm
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

Bonjour

> > It's the same as doing i = i++; // undefined behavior.
> 
> Your example is specious at best.
> 

It is right on target, it describes an undefined behavior
that no STDs make promises on.
They are both compiler/library/system/whatever specific.

> >
> > IMO, it is ill-advise to allow fflush() on an input stream even
> > if at first it may look like a usefull extension.
> >
> 
> And I disagree.  Certainly no change there.
> 
> Suppose we define fflush(FILE *stream) as
> 
>      flush the buffer associated with the stream
> 
> If it is decided that fflush() on an input stream should be defined,
> how will that affect your work?

It will not, I try to avoid non-portable code.
Usually when people do fflush(stdin), they want to discard the
input keyboard. A portable way maybe

tcflush(STDIN_FILENO, TCIFLUSH);
or just plain
int c;
do {
	c = getchar();
} while (c != EOF && c != '\n');

But if you want to do implement it;

int fflush (FILE * stream)
{
.....
	if (stream)
	{
		int fd = fileno(stream);
		/* discard the allocated library buffer */
		....
		/* discard whatever the system/device had */
		if (isatty(fd))
		{
			tcflush(fd, TCIFLUSH);
			tcflush(fd, TCOFLUSH);
		}
		.....
	}
 ...
}

But this not as trivial as you may think.

-- 
au revoir, alain
----
Aussi haut que l'on soit assis, on est toujours assis que sur son cul !!!