Date: Mon, 12 Apr 1999 13:07:49 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: DJ Delorie <dj AT delorie DOT com>
cc: djgpp-workers AT delorie DOT com
Subject: Re: v2.03 release: what else has to be done?
In-Reply-To: <199904111705.NAA01422@envy.delorie.com>
Message-ID: <Pine.SUN.3.91.990412130548.3441A-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
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


On Sun, 11 Apr 1999, DJ Delorie wrote:

> Note: I tried adding this, but there's a lot of places in libc where
> we blindly call fflush() on streams (rewind is one)

I was afraid of something like that.  It seems to become complicated 
enough to raise again the question: do we really want this so badly?

I said this earlier, and I say it again: it seems to me that the real
problem is not what `fflush' does or doesn't do for input streams.
The real problem is that many PC-based programs call `fflush' on stdin
because they expect that to solve some other problem.  If we solve
that other problem, there will be no need to change anything in
`fflush'.

I'm still asking those who know what is the real problem behind using
`fflush' on stdin, to please speak up.