Question

I need just a little bit of help translating C to python. I have not used c++ in about 4 years now, and when I used it, I only new the basics. the context of this snippet is reading a file and creating a checksum

int ff7_checksum( void* qw )
{
   int i = 0, t, d;
   long r = 0xFFFF, len = 4336;
   long pbit = 0x8000;
   char* b=(char*)qw;

   while( len-- ) {
      t = b[i++];
      r ^= t << 8;
      for(d=0;d<8;d++) {
         if( r & pbit )
            r = ( r << 1 ) ^ 0x1021;
         else
            r <<= 1;
      }
      r &= ( 1 << 16 ) - 1;
   }
   return (r^0xFFFF)&0xFFFF;
}

Answer 1

I assume that qw points at a buffer with a known size (4336) and that does not really represent a string (i.e. it might contain \0 characters that should still be processed).

However, in Python (2.x), we still generally model this as a string, since the string may contain embedded \0 bytes, and knows its own length. There is thus also no reason to hard-code the length.

More idiomatically, we get something like:

def ff7_checksum(data): # data used to be 'qw'
  all_bits = 0xFFFF # a 16-bit value with all bits set.
  result = all_bits # result used to be 'r'
  pbit = 0x8000 # the highest-order bit in a 16-bit value.

  for byte in data: # byte used to be 't'
    result ^= byte << 8
    for i in range(8):
      result = (result << 1) ^ (0x1021 if result & pbit else 0)
    result &= all_bits
  return result ^ all_bits # the original &-mask is not necessary