On Oct 7, 10:04 am, "M.-A. Lemburg" <m...@egenix.com> wrote: > Mike Driscoll wrote: > > Hi, > > > I am working on a project where I need to decrypt some data that has > > been encrypted with AES. It looks like M2Crypto is the module of > > choice for these sorts of things, but I cannot figure out how to do > > this stuff from the docs. I have the following PHP code that needs to > > be translated into Python: > > > $iv_size = mcrypt_get_iv_size(MCRYPT_RIJNDAEL_256, MCRYPT_MODE_ECB); > > $iv = mcrypt_create_iv($iv_size, MCRYPT_RAND); > > return rtrim(mcrypt_decrypt(MCRYPT_RIJNDAEL_256, $key, > > $data,MCRYPT_MODE_ECB, $iv),"\0"); > > > I can't find a method in M2Crypto that gets the "initialization > > vector" size. I found the right method in the tests, which appears to > > be EVP.Cipher. So I would assume, I would need to do something like: > > > EVP.Cipher(alg="aes_256_ecb", key=SomeKey, iv=SomeIV, op=dec, > > padding=False) > > > I don't really see where I pass the data that needs the decrypting > > though. Can someone shed some light on this? > > If you just need AES, you're probably better of with pycrypto: > > http://www.amk.ca/python/code/crypto > > Still, to answer your question: AES uses blocks of 16 bytes (256 bits) > each, so the IV-size is always 16 bytes. > > BTW: I'm not sure what the PHP code is trying to do ... ECB mode > doesn't use the IV at all. It's only used for chained modes and > there you include the IV in the encrypted data (usually at the > beginning), since you need it for decryption. The PHP code > apparently generates a random IV block for decryption. This > would never work in e.g. CBC mode. > > -- > Marc-Andre Lemburg > eGenix.com >
That's good to know. I had originally started with PyCrypto by following the example here: http://www.codekoala.com/blog/2009/aes-encryption-python-using-pycrypto/ Unfortunately, no matter which base64 decoding method I use, I get a padding error or in the case of b16decode, I get "TypeError: Non- base16 digit found". AES decoding is something I've never done before, so I apologize for my greenness. I'll bug the guys on the pycrypto list as well. Thanks, Mike -- http://mail.python.org/mailman/listinfo/python-list
