Path: utzoo!attcan!uunet!tank!ncar!unmvax!tut.cis.ohio-state.edu!rutgers!att!ulysses!attibr!vch
From: vch@attibr.UUCP (Vincent C. Hatem)
Newsgroups: comp.bugs.sys5
Subject: Re: possible bug in HDB uucp?
Summary: Hmm...
Message-ID: <116@attibr.UUCP>
Date: 24 Jan 89 22:52:30 GMT
References: <1025@vsi.COM>
Distribution: comp
Organization: AT&T International, Basking Ridge, NJ
Lines: 30

In article <1025@vsi.COM>, friedl@vsi.COM (Stephen J. Friedl) writes:
} Is this a bug?  I'm running Sys V Rel 3.1 on a 3B2/400 with
} (obviously) HDB uucp.  I'm in the process of connecting with a
} local site (hi Larry!), and the password I was given included a %
} character.  After a handful of failed logins, it occurred to me
} that perhaps the Systems send strings were subject to "printf-
} expansion" (is that a real term?).  I tried a password of
} 
}         %foobar
} 
} and I got a floating zero followed by "oobar".  I tried doubling
} the % to %%foobar but it still doesn't work.  By now, we've
} gotten things to work by using a different password, but this %
} business can't be right...

Very strange... the source code uses write(2) to send the string to the
remote host. It shouldn't interere. It might be mashing the % before then,
such as when it reads it from the Systems file and tries to translate
it somehow. I dodn't find anything that would do such a thing in the 
code, though.

Have you tried a backlash before the %?

-Vince

-- 
Vincent C. Hatem
e-mail: {any Action Central site || att}!attibr!vch
snail-mail: AT&T International, 1200 Mt. Kemble Ave, Basking Ridge, NJ 07920
tele: +1-201-953-7959         ------ Gruezi wohl! -------