def run():
buff = base64.b64decode(''.join([line.rstrip() for line in open('set1/e6.txt', 'r')]))
# Number of sample blocks we want to use to calculate hamming dist
avg_blocks = 4
keys = []
for key_size in range(2,41):
# Grab first X number of buffer blocks
blocks = [buff[i * key_size:(i * key_size) + key_size] for i in list(range(0, avg_blocks))]
# Calculate hamming dist between them
hamm = sum([hamming(blocks[i], blocks[j]) for i in list(range(0, avg_blocks)) for j in list(range(i, avg_blocks))])
keys.append({'size': key_size, 'hamming': hamm / key_size})
skeys = sorted(keys, key=lambda s: s['hamming'])
key_size=skeys[0]['size']
print("Key size: %d" % key_size)
# Transpose the buffer into key_size number
# eg: for ks: 4 and buff: '123456789' -> ['159', '26', '37', '48']
len_buff=len(buff)
blocks = [bytearray([buff[(key_size * j) + i] for j in range(0, math.ceil((len_buff - i) / key_size))]) for i in range(0, key_size)]
key = bytearray()
for i in list(range(0, len(blocks))):
res = find_best_xor_match(blocks[i], 1)
key += res[0].mask
print("Key: %s" % key)
print(key_encode(buff, key))
def run():
expected = b"0b3637272a2b2e63622c2e69692a23693a2a3c6324202d623d63343c2a26226324272765272a282b2f20430a652e2c652a3124333a653e2b2027630c692b20283165286326302e27282f"
orig = bytearray("Burning 'em, if you ain't quick and nimble\nI go crazy when I hear a cymbal", "ascii")
key = bytearray("ICE", "ascii")
encoded = key_encode(orig, key)
print(binascii.hexlify(encoded) == expected)
