将Python 2字节校验和计算器转换为Python 3

Question

I'm trying to translate this Python 2 code to Python 3.

def calculate_checksum(packet):
  total = 0
  for char in packet:
    total += struct.unpack('B', char)[0]
  return (256 - (total % 256)) & 0xff

In Python 3 it causes a TypeError:

    total += struct.unpack('B', char)[0]
TypeError: a bytes-like object is required, not 'int'

I have been trying to research the changes in strings and bytes but it is a bit overwhelming.

Answer 1

The code basically translates individual characters in a bytestring to their integer equivalent; the character \\x42 becomes 0x42 (or decimal 66), for example:

>>> # Python 2
...
>>> struct.unpack('B', '\x42')[0]
66

As an aside, you could do the same more simply with the ord() function :

>>> ord('\x42')
66

In Python 3, you already get integers when you iterate over a bytes object, which is why you get your error:

>>> # Python 3
...
>>> b'\x42'[0]
66

The whole struct.unpack() call can simply be dropped:

for char in packet:
    total += char

or simply use sum() to calculate the total in one step:

total = sum(packet)

making the complete version:

def calculate_checksum_ord(packet):
    total = sum(ord(c) for c in packet)
    return (256 - (total % 256)) & 0xff

Note that the Python 2 code could also use sum() , with the ord() function rather than use struct : total = sum(ord(c) for c in packed) .