Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.csd.uwm.edu!cs.utexas.edu!uunet!mcsun!ukc!edcastle!aiai!richard
From: richard@aiai.ed.ac.uk (Richard Tobin)
Newsgroups: comp.lang.c
Subject: Re: UNIX-like crypt function
Keywords: crypt UNIX
Message-ID: <806@skye.ed.ac.uk>
Date: 24 Aug 89 18:56:25 GMT
References: <855@eutrc3.urc.tue.nl> <2152@netcom.UUCP> <17369@ut-emx.UUCP> <10793@smoke.BRL.MIL> <164@spam.ua.oz> <10802@smoke.BRL.MIL>
Reply-To: richard@aiai.UUCP (Richard Tobin)
Organization: AIAI, University of Edinburgh, Scotland
Lines: 28

In article <10802@smoke.BRL.MIL> gwyn@brl.arpa (Doug Gwyn) writes:
>In article <164@spam.ua.oz> wvenable@spam.oz.au (Bill Venables) writes:
>> Let me confirm (although noone seems to doubt it) that the crypt() facility
>> is not available on UNIX machines in Australia [...]
>
>It may not be distributed in commercial releases, but I guarantee that
>some UNIX system in Australia does have all the usual UNIX crypt code,
>obtained at a time when nobody was paying much attention to this matter.

All the Unix systems I've seen in the UK have the crypt() library
function.  What recent ones don't have is the crypt *program*.  It would
be very awkward if the function were removed, since it is used for
passwords.  The source for the program crypt has been posted to the net
at least once (in the crypt-breakers workbench).

As I understand it, the crypt() function implements the DES algorithm
(modified by the presence of the salt).  The crypt program uses an
enigma-type encoding, after using crypt() (indirectly through the
trivial program makekey) to generate a key from the password.

All this seems a bit strange, since I'd have thought that DES was the
thing that might not be exportable...

-- Richard
-- 
Richard Tobin,                       JANET: R.Tobin@uk.ac.ed             
AI Applications Institute,           ARPA:  R.Tobin%uk.ac.ed@nsfnet-relay.ac.uk
Edinburgh University.                UUCP:  ...!ukc!ed.ac.uk!R.Tobin