Okay, so this was a dumb question ;-)

I guess it's all in the IEEE math representation and how printf deals with
this on any given platform. The compiler does the math correctly, it just
represents a quiet NaN in a different way to gcc. Changed the code that used
it (cricket) to recognise -1.#INF as nan and problem went away.

----- Original Message ----- 
From: "Eastbarn" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Monday, July 19, 2004 5:08 PM
Subject: [rrd-users] NaN representation on win32


All,

hope someone can help me with a win32 newbie question. Alternatively send me
across to the developer mailing list if I'm in the wrong place ;-)

I've compiled a snapshot version of 1.1 on win32 using VC6 (w2k, activestate
perl 5.8) . The representation that I'm seeing for NaN is rather different
to my previous experience (Solaris/Linux). I'm seeing NaN represented as
"-1.#IND" (the float representation for quiet NaN), rather than the usual
NaN by the perl shared bindings. This is causing me a few problems further
down the line.

Can anyone point me in the right direction - math never was my best subject.

Thanks in advance .... Jonathan



________________________________________________
Message sent using UebiMiau 2.7.2

--
Unsubscribe mailto:[EMAIL PROTECTED]
Help        mailto:[EMAIL PROTECTED]
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi


--
Unsubscribe mailto:[EMAIL PROTECTED]
Help        mailto:[EMAIL PROTECTED]
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi

Reply via email to