From ngps at POST1.COM  Sun Jun  3 17:54:24 2001
From: ngps at POST1.COM (Ng Pheng Siong)
Date: Sun, 3 Jun 2001 23:54:24 +0800
Subject: [PYTHON-CRYPTO] [ANN] M2Crypto 0.06-snap6 and Win32 binaries
Message-ID: <20010603235424.A246@madcap.dyndns.org>

Hi,

Snapshot #5 claimed to be Python 2.1-compatible.

Snapshot #6 is now _even_more_ compatible! ;-)
Specifically, httpslib and m2urllib now work with Python 2.1.

Get it here:

    http://www.post1.com/home/ngps/m2


For once, Win32 binaries have been posted simultaneously. ;-)

As usual, feedback is welcome.

--
Ng Pheng Siong <ngps at post1.com> * http://www.post1.com/home/ngps

Quidquid latine dictum sit, altum viditur.




From kalath at lycos.com  Mon Jun 18 12:35:05 2001
From: kalath at lycos.com (Mark Brady)
Date: Mon, 18 Jun 2001 10:35:05 0000
Subject: [PYTHON-CRYPTO] Better Python Crypto.
Message-ID: <OCCPDMJBIMDMACAA@mailcity.com>

Hello,
      Has anyone looked int wrapping Crypto++ (http://www.eskimo.com/~weidai/cryptlib.html) to provide better crypto for python. All the current python crypto solutions are either OpenSSL based or single algorithm implementations. The OpenSSL lib is fine but it only supports a limited number of Cipers and pure cryptography is not the focus of their project so it seems unlikely that they will be supporting very many more algorithms. So has anyone any ideas, comments?

Mark.



Get 250 color business cards for FREE!
http://businesscards.lycos.com/vp/fastpath/




From bram at GAWTH.COM  Mon Jun 18 19:13:43 2001
From: bram at GAWTH.COM (Bram Cohen)
Date: Mon, 18 Jun 2001 10:13:43 -0700
Subject: [PYTHON-CRYPTO] Better Python Crypto.
In-Reply-To: <OCCPDMJBIMDMACAA@mailcity.com>
Message-ID: <Pine.LNX.4.21.0106181010250.3177-100000@ultra.gawth.com>

On Mon, 18 Jun 2001, Mark Brady wrote:

> Has anyone looked into wrapping Crypto++
> (http://www.eskimo.com/~weidai/cryptlib.html)

Yeah, I wrapped Crypto++ for a previous project, and it was a disaster -
Crypto++ itself is 'well written', meaning it uses lots of operator
overloading, and templates, and skads of C++ garbage which has the
'amusing' effect of making you have to recompile for redhat and debian.

I'd say Crypto++ is well written, but doing it in C++ was a bad idea.

-Bram Cohen

"Markets can remain irrational longer than you can remain solvent"
                                        -- John Maynard Keynes




From bram at GAWTH.COM  Fri Jun 22 23:10:23 2001
From: bram at GAWTH.COM (Bram Cohen)
Date: Fri, 22 Jun 2001 14:10:23 -0700
Subject: [PYTHON-CRYPTO] need counter mode implementation
Message-ID: <Pine.LNX.4.21.0106221404010.9644-100000@ultra.gawth.com>

I've now finished the encryption part of an application I'm writing, and
the crypto is slow (unsurprising, since it's in pure python.) I've got it
set up to try to import a C version and only use the pure python version
if there's an ImportError (I love Python).

I'm wondering if anyone could throw together a pure C version of the
following together quickly -

A module named _StreamEncrypter has a function called make_encrypter.
make_encrypter takes a 128 bit rijndael key and returns a callable which
encrypts strings it's passed in big endian counter mode. So, if you pass
it the same string twice, it will return two different values because it
xored with different encrypted blocks, and decrypting is done using the
same function as encrypting.

I prefer an implementation which is public domain. Anyone sending me one
gets their name in the credits :)

-Bram Cohen

"Markets can remain irrational longer than you can remain solvent"
                                        -- John Maynard Keynes




From bram at GAWTH.COM  Thu Jun 28 13:09:33 2001
From: bram at GAWTH.COM (Bram Cohen)
Date: Thu, 28 Jun 2001 04:09:33 -0700
Subject: [PYTHON-CRYPTO] A full-blown application
Message-ID: <Pine.LNX.4.21.0106280407290.8975-100000@ultra.gawth.com>

The app I've been working on is now at release 1.0, you can get it here -

http://bitconjurer.org/BitTorrent/

It could still use a C implementation of it's crypto - basically the only
crypto call is StreamEncrypter.make_encrypter, which returns big-endian
rijndael counter mode.

-Bram Cohen

"Markets can remain irrational longer than you can remain solvent"
                                        -- John Maynard Keynes




From bram at GAWTH.COM  Sat Jun 30 00:07:14 2001
From: bram at GAWTH.COM (Bram Cohen)
Date: Fri, 29 Jun 2001 15:07:14 -0700
Subject: [PYTHON-CRYPTO] New release of BitTorrent out
Message-ID: <Pine.LNX.4.21.0106291503400.8975-100000@ultra.gawth.com>

I've put up a new release, you can get it here -

http://bitconjurer.org/BitTorrent/BitTorrent-01-00-01.tar.gz

It's also now the one linked off the BitTorrent pages.

This is a bug fix release - it no longer hoses the CPU after connection
fails, and doesn't throw an exception when an upload connection hasn't
gotten it's status set yet.

The next version will include copying of comments from blobs_want to
blobs_have, instructions when you execute with inappropriate parameters,
and (hopefully) crypto in C.

That last one I could use some help with - if anyone throws together a
public domain distutils-ified C implementation of StreamEncrypter.py, I'll
include it immediately (should be fairly straightforward - it's just
rijndael in counter mode). Otherwise it's dependant on me successfully
bugging my younger brother to write one.

-Bram Cohen

"Markets can remain irrational longer than you can remain solvent"
                                        -- John Maynard Keynes