counter i is reset to 0 for each round
Opened this issue · 2 comments
Line 26 in 1e7f289
In reading the accompanying doc on this: https://medium.com/asecuritysite-when-bob-met-alice/ffx-schemes-for-format-preserving-encryption-a2a7aa4f1377 I noticed that the counter is reset each round. Works fine removing the initialization and allows the value to be carried.
I agree that reusing the variable like that is ugly. The round number i
is still used to salt the hash though: it's packed into key
before i
is reset.
If we wouldn't include the round number directly and instead just used it as an offset for i
all rounds would end up using the same sequence, just shifted by the round number.
Thanks for bringing the post to my attention, I hadn't seen that before!
In my testing it is not incremented in the generator and is set to 0 each round. The way I see the code, i is set in the encrypt function using the range of self.rounds:
a, b = self.split(v)
for i in range(self.rounds):
c = self.add(radix, a, self.round(radix, i, b))
a, b = b, c
return a + b
In this way, i is incrementing already and there doesn't seem to be a reason to attempt to increment again.
To test, I simply put a print(i) in the round function:
key = struct.pack('I%sI' % len(s), i, *s)
chars_per_hash = int(self.digest_size * math.log(256, radix))
i = 0
while True:
print(i)
h = hmac.new(self.key, key + struct.pack('I', i), self.digestmod)
d = int(h.hexdigest(), 16)
for _ in range(chars_per_hash):
d, r = divmod(d, radix)
yield r
key = h.digest()
i += 1
However, if you want to keep the increment inside round, you just have to move the i += 1 to inside the for loop (this brings up that the assignment of key should also be moved if that is indeed needed):
key = struct.pack('I%sI' % len(s), i, *s)
chars_per_hash = int(self.digest_size * math.log(256, radix))
i = 0
while True:
print(i)
h = hmac.new(self.key, key + struct.pack('I', i), self.digestmod)
d = int(h.hexdigest(), 16)
for _ in range(chars_per_hash):
d, r = divmod(d, radix)
yield r
i += 1
key = h.digest()