From: ams AT ludd DOT luth DOT se
Message-Id: <200003210709.IAA03083@father.ludd.luth.se>
Subject: Re: Unnormals???
In-Reply-To: <Pine.SUN.3.91.1000320191858.27773C-100000@is> from Eli Zaretskii at "Mar 20, 2000 07:19:27 pm"
To: djgpp-workers AT delorie DOT com
Date: Tue, 21 Mar 2000 08:09:50 +0100 (MET)
X-Mailer: ELM [version 2.4ME+ PL54 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: dj-admin AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

According to Eli Zaretskii:
> 
> On Mon, 20 Mar 2000, Martin Stromberg wrote:
> 
> > Perhaps the n-char-sequence shall be used to show the bitpattern of
> > the NaN? Then we would see the sign if we knew what bit it is...
> 
> This is also a possibility I won't object.

I've just discovered chapter 7.12.11.1, copysign():
"double copysign(double x, double y);
...

The copysign functions produce a value with the magnitude of x and the
sign of y. They produce a NaN (with the sign of y) if x is a NaN."

This means the standard do think that NaNs have a sign (however
misguided that is), hence we really do need to print the "-" of a
negative NaN. Period.


Right,

							MartinS