https://twitter.com/aoirint/status/1637363788784734208

decode_utf8_bytestream.py

def decode_utf8_bytestream(byte_stream):
    i = 0
    decoded_strings = []
    while i < len(byte_stream):
        byte = byte_stream[i]
        if byte <= 0x7F:
            decoded_strings.append(byte_stream[i:i+1])
            i += 1
        elif byte <= 0xDF:
            decoded_strings.append(byte_stream[i:i+2])
            i += 2
        elif byte <= 0xEF:
            decoded_strings.append(byte_stream[i:i+3])
            i += 3
        else:
            decoded_strings.append(byte_stream[i:i+4])
            i += 4
    return [bytes(s).decode('utf-8') for s in decoded_strings]


byte_str = b'\xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\x00\xe3\x81\x8a\xe3\x81\x86\xe3\x81\x8c\xe3\x81\x84\x00'
decoded_strings = decode_utf8_bytestream(byte_str)
print(decoded_strings[0]) # 'あいう'
print(decoded_strings[1]) # 'おうがい'