Re: Equivalent transformation spec. in Bouncy Castle API 1.4

From: Michael (michael.tran_at_ict.nl)
Date: 09/30/04 

Date: 30 Sep 2004 01:14:49 -0700

"Greg Stark" <stark2000@comcast.net> wrote in message news:<DLmdnYcUkuyXgMbcRVn-jg@comcast.com>...
> >
> > [This is why I write my own implementations. I could never figure out
> > how to make the providers work.]
> >
>
> I thought it was just me. I spent about 2 hours trying figure out how to
> enter a triple DES key from a byte [] array in jdk 1.4.2. The api is
> ridiculously abstract. I haven't looked at the bouncy castle api.
>
> ghstark AT pobox DOT com

Hi All,

Does anybody know what the 1.2 implementation of Bouncy Castle does
with "DESede//NoPadding" since the 1.4 implementation could not deal
with this and by substituting "DESede/ECB/NoPadding", we have got
cryptogram computed by the server that is not equal with the one
computed by the client.

The idea is to re-use the code that used to process
"DESede//NoPadding" successfully in 1.2 in 1.4, so that we still can
live with "DESede//NoPadding" in 1.4.

Quite a challenge huh !

Michael

Relevant Pages