Hi Max,
There are definitely things weird in that patch :
- You double declare the same function - Why replace #defined constant by hardcoded numbers ? - Speaking about readability :
kc[6] = 4 * (kc[6] >> 2);
isn't the most obvious way to do kc[6] &= 0xfc;
Cheers,
Sylvain
On Mon, Nov 18, 2013 at 6:47 PM, ☎ Max.Suraev@fairwaves.ru wrote:
Hello.
I've got rid of unnecessary cycle and to made difference between v2 and v3 more visible: v2 is basically just v3 with last bits of Kc zeroed. Also - small readability improvements.
Also I've added test suite with test vectors from original python implementation.
-- best regards, Max, http://fairwaves.ru