You can try the f.read(size) function. If size is set to 1, it will read character by character, and then filter the data.

This is easy to usepython, but the result is the same, it’s because your posture is not right

For example, the content you filtered> twiiiiiii@2, 我模拟一下, 假如待处理的文件(raw.dat) is as follows

> t我w爱i北i京i天i安i门ii@2

Up code

with open("raw.dat", "rb") as fh: line = fh.readline() print(line) print(line.decode('ascii', 'ignore'))

Output

b'> t\xce\xd2w\xb0\xaei\xb1\xb1i\xbe\xa9i\xcc\xeci\xb0\xb2i\xc3\xc5ii@2' > twiiiiiii@2

So there’s no problem, it’s no different from what you useC