Path: utzoo!utgpu!water!watmath!clyde!rutgers!cmcl2!husc6!mit-eddie!apollo!arnold
From: arnold@apollo.uucp (Ken Arnold)
Newsgroups: comp.lang.c
Subject: Re: "logical" xor
Message-ID: <39a83409.ae48@apollo.uucp>
Date: 13 Jan 88 22:27:00 GMT
References: <2946@zeus.TEK.COM> <170@illusion.UUCP>
Reply-To: arnold@apollo.UUCP (Ken Arnold)
Organization: Apollo Computer, Chelmsford, MA
Lines: 15

In article <170@illusion.UUCP> Marcus Hall writes:
>((a != 0) ^ (b != 0)) gives (1) ^ (1) == 0, which is false.

Not even true yet.  All (a != 0) is guaranteed to generate (if a is
not 0) is a non-zero value, *any* non-zero value, which just *usually*
is 1.  It could be 7.  So you *could* have ((a != 0) ^ (b != 0)) giving
((7) ^ (5)) or something.  There is no guarantee, therefore, that ^
will even solve the problem this way.  To be *sure*, you have to write

	((a != 0 ? 1 : 0) ^ (b != 0 ? 1 : 0))

Do you wonder why some of us have always wanted a ^^ operator?  Not
often, mind you, but when I need one, I *really* need it, 

		Ken Arnold