New Python block cipher API, comments wanted

[Andrew Bennetts]

On Thu, Jan 30, 2003 at 12:47:07AM +1300, Andrew McGregor wrote:
> 
> 
> --On Tuesday, January 28, 2003 21:48:28 -0800 Paul Rubin 
> <phr-n2003b at NOSPAMnightsong.com> wrote:
> 
> >In fact, I don't think any module implementing PEP 272 has ever been
> >written, and the interface it describes isn't that great, so I
> >wouldn't call the PEP process a big success where block ciphers are
> >concerned.  There was a bunch of subsequent discussion on
> >Python-crypto that led up to the current proposal, and that seems to
> >have done a better job.
> 
> Um, there is such a module, it's called PyCrypto and is at at least 
> revision 1.9.  I've used it to write a significant application, which 
> interoperates with others using different libraries.  Nice C or even ix86 
> assembler ciphers, very, very fast.  It has an API feature I'd call 
> essential that are not part of PEP 272 but is included in your proposal, 
> that is the ability to change IV after creating a cipher, between uses of 
> it.

PyCrypto certainly seems solid -- Twisted has an SSH2 implementation,
Twisted Conch, that uses it, and it seems to work very nicely (as does conch
:).

I haven't looked at any of the implementations or APIs or anything, but the
I take the fact that Conch "Just Works" to be a good sign :)

-Andrew.