Bug #68238 [Opn->Nab]: mcrypt_encode tests are broken

Edit report at https://bugs.php.net/bug.php?id=68238&edit=1 ID: 68238 Updated by: nikic@php.net Reported by: gm dot outside+php at gmail dot com Summary: mcrypt_encode tests are broken -Status: Open +Status: Not a bug Type: Bug Package: Testing related Operating System: Linux PHP Version: 5.6.1 Block user comment: N Private report: N New Comment: This is a bug in your mcrypt library, which does not properly support Cast-128 with keys <= 80 bits. Padding the keys with NUL bytes is not the same, because Cast-128 uses only 12 rounds (instead of 16) if the key is <= 80 bits. Updating your libmcrypt version will probably fix this. Previous Comments: ------------------------------------------------------------------------ [2014-10-15 17:32:24] gm dot outside+php at gmail dot com I did more testing, and I was a bit wrong about the strlen() part. Manually padding the keys with '\0' actually works, but the result does not match the ciphertext provided in RFC-2144 B.1 anymore. Additionally to that I was wrong re: the keysize requirement for that cipher should be 16 bytes (128-bit) as cipher's name 'cast-128' suggests. Once the keys are properly padded with '\0' to be 128-bit the test returns the following differences: 002+ 80-bit: 753de29f5d167d03 003+ 40-bit: f00b0530833d7444 002- 80-bit: eb6a711a2c02271b 003- 40-bit: 7ac816d16e9b302e So, something else was also changed that the mcrypt extension no longer conforms to RFC-2144 B.1. ------------------------------------------------------------------------ [2014-10-15 17:16:27] gm dot outside+php at gmail dot com Description: ------------ There was a recent commit (http://git.php.net/?p=php-src.git;a=commit;h=a861a3a93d89a50ce58e1ab1abef1eb501f97483) that changed behaviour of the mcrypt_encode() function. After that commit the key is required to be at least the expected key length long, otherwise a warning message is issued and the mcrypt_encode() routine returns a failure. The corresponding test in ext/mcrypt/tests/bug62102_rfc2144.php supplies 10 bytes key instead of 16 for cast-128 80-bit encryption and 5 bytes key instead of 10 for cast-128 40-bit encryption. A quick fix to the test would be to pad the keys with '\0' manually (RFC-2144 B.1), e.g. mcrypt_encrypt('cast-128', "\x01\x23\x45\x67\x12\x34\x56\x78\x23\x45\0\0\0\0\0\0", $plaintext, 'ecb') but unfortunately due to the way changed code treats key data (as a null terminated string) and due to calculating the key size as strlen() of that string there is no way to satisfy the RFC-2144 B.1 since all trailing '\0' will be ignored. Expected result: ---------------- That the RFC-2144 test would be passed with the explicitly specified vector and that mcrypt_encrypt() would honour the key argument as a binary string that can include '\0' anywhere in the string. Actual result: -------------- All trailing '\0' in the key argument are ignored, therefore it's impossible to pass RFC-2144 test to match section B.1. ------------------------------------------------------------------------ -- Edit this bug report at https://bugs.php.net/bug.php?id=68238&edit=1