Bill Frantz: Re: DES.java

[Jonathan S. Shapiro]

[I am forwarding this to the list as well.]

------- Forwarded Message

Return-Path: frantz@communities.com
Delivery-Date: Tue Sep 29 16:16:08 1998
Received: from halifax.syncomas.com (halifax.syncomas.com [192.168.0.1])
	by snocrash.cis.upenn.edu (8.8.7/8.8.7) with ESMTP id QAA21729
	for <shap@halifax.syncomas.com>; Tue, 29 Sep 1998 16:16:07 -0400
Received: from mail.earthlink.net
	by halifax.syncomas.com (fetchmail-4.4.4 POP3)
	for <shap/halifax.syncomas.com> (single-drop); Tue, 29 Sep 1998 16:16:07 EDT
Received: from eros.cis.upenn.edu (EROS.CIS.UPENN.EDU [158.130.6.119])
	by tern.prod.itd.earthlink.net (8.8.7/8.8.5) with ESMTP id QAA03411
	for <jsshapiro@earthlink.net>; Tue, 29 Sep 1998 16:07:26 -0700 (PDT)
Received: (from shap@localhost)
	by eros.cis.upenn.edu (8.8.5/8.8.5) id SAA03183
	for jsshapiro@earthlink.net; Tue, 29 Sep 1998 18:53:07 -0400
Received: from homer.communities.com (homer.communities.com [205.162.51.9])
	by eros.cis.upenn.edu (8.8.5/8.8.5) with ESMTP id SAA03180
	for <shap@eros.cis.upenn.edu>; Tue, 29 Sep 1998 18:53:06 -0400
Received: from deep-thought.communities.com (deep-thought.communities.com [205.162.51.137])
	by homer.communities.com (8.9.1/8.9.1) with SMTP id QAA14190;
	Tue, 29 Sep 1998 16:10:38 -0700
Message-Id: <3.0.32.19980929160201.00b7dc70@homer>
X-Sender: frantz@homer
X-Mailer: Windows Eudora Pro Version 3.0 (32)
Date: Tue, 29 Sep 1998 16:03:02 -0700
To: shap@eros.cis.upenn.edu
From: Bill Frantz <frantz@communities.com>
Subject: Re: DES.java 
Cc: "Jonathan Shapiro" <jsshapiro@earthlink.net>, markm@caplet.com,
        chip@homer.communities.com, sidney@homer.communities.com
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-UIDL: 6367d879a212c8496cd658b0783b3740

I believe the changes to use Cryptix instead of java.security should be
somewhere between trivial and easy.  (It depends on how similar the Cryptix
API is to the java.security API.)  Cryptix 3.0 tries to follow the Java
Cryptography Extension API, so perhaps the other pieces will "just drop in".

As I told MarkM at lunch, I am reluctant to make the change for EC's
version at this point in the product cycle.  I have no problem making it in
the erights version and then folding it back to the EC version after ECID
ships.

Your idea of a stub is a good temporary bridge.  I will look at what is
involved.

(We use MD5, SHA(1), DSS, SecureRandom, and SeedGenerator from
java.security.  MD5 and SHA(1) appear to depend on MessageDigest and
NoSuchAlgorithmException.  DSA adds KeyPair, KeyParGenerator, PublicKey,
and PrivateKey to the collection.)


At 03:38 PM 9/29/98 -0400, shap@eros.cis.upenn.edu wrote:
>The other day I discovered that the kaffe java library does not
>support the java.security classes.
>
>Actually, it didn't support BigInteger either.  I talked to Tim Wilke
>today, and he noted that their stated support was for the ``Personal
>Java'' subset, which does not include BigInteger.  The latest snapshot
>includes BigInteger support, so don't get too worried about that.
>
>My guess is that the cryptix library contains support to actually do
>everything difficult that is in the java.security classes.
>
>How hard would it be to implement a ``stub'' version of the
>java.security classes that used the cryptix library to do all of the
>hard work?  It would sure be nice if this stuff ran on kaffe!
>
>
>shap
>
>


------- End of Forwarded Message