From: Eric Young <eay@mincom.oz.au>
Date: Fri, 13 Oct 95 06:53:04 PDT
To: Adam Shostack <adam@homeport.org>
Subject: Re: IDEA algorithm
In-Reply-To: <199510130250.WAA08368@homeport.org>
Message-ID: <Pine.SOL.3.91.951013135056.2444A-100000@orb>
MIME-Version: 1.0
Content-Type: text/plain


On Thu, 12 Oct 1995, Adam Shostack wrote:
> | (Btw... what would be the absolute coolest thing folk would like to see
> | added to RSAREF? That is, other than code that we will need to negotiate
> | for. What new thing, speed upgrade, enhancement, etc would just tickle?)
...
> 	Another really big win would be to add more entry points to
> the code to legitimize Wei Dai's excellent Crypto++ library, which has
> to jump in at all sorts of unpublished interfaces.

How about just making all the interfaces at the different levels 
available.  For the next version of SSLeay (when I get it and the 
documentation finished) will have defined interfaces at the following 
levels for the various ciphers etc.

The actual cipher level
	eg des_set_key, des_cfb_encrypt, idea_cbc_encrypt...
	This level is wanted by things like SSL for the fastest on the fly
	encryption possible.

The 'by parts' routines
	eg EncryptBegin, EncryptUpdate, EncryptFinish (which take a CIPHER 
	pointer as an argument).
	This is sort of wanted by people implementing higher level digital 
	signature routines but they are usefull in their own right.

The 'digital' envelope routines.
	eg SealBegin, SealUpdate, SealFinish (which take a CIPHER pointer, a 	
	MD pointer and the public keys etc.
	The high level stuff for PKCS-7/S-MIME/PEM etc.

I am also documenting the interface to the big number library (anyone 
want to do DSS :-).
--
Eric Young                  | Signature removed since it was generating
AARNet: eay@mincom.oz.au    | more followups than the message contents :-)