解析酷狗二进制编码的krc歌词文件

解开明文算法

import zlib

# 实际krc是用zlib将歌词压缩之后与enc_key执行了简单的异或加密
# 解析的过程就是解密后解压
# 但是解开明文之后歌词的字符编码通常是UTF-8-BOM | UTF-8-SIG
def decompress_krc(krc_binary_bytes):
    enc_key = bytearray([64, 71, 97, 119, 94, 50, 116, 71, 81, 54, 49, 45, 206, 210, 110, 105])

    content_bytes = krc_binary_bytes[4:]
    compressed_bytes = []

    for i in range(0, len(content_bytes)):
        compressed_bytes.append(content_bytes[i] ^ enc_key[i % 16])

    return zlib.decompress(bytearray(compressed_bytes))

一个把krc解析出明文文件并打印出来的例子

import zlib
import sys
import os

def decompress_krc(krc_bytes):
    enc_key = bytearray([64, 71, 97, 119, 94, 50, 116, 71, 81, 54, 49, 45, 206, 210, 110, 105])

    content_bytes = krc_bytes[4:]
    compressed_bytes = []

    for i in range(0, len(content_bytes)):
        compressed_bytes.append(content_bytes[i] ^ enc_key[i % 16])

    return zlib.decompress(bytearray(compressed_bytes))

if len(sys.argv) != 2:
    print('usage : parse_krc <krc path>')
    sys.exit(0)

filename = sys.argv[1]
target_filename = os.path.splitext(filename)[0] + '_plaintext.krc'

with open(filename, 'rb') as f:
    plaintext_bytes = decompress_krc(bytearray(f.read()))

    fp = open(target_filename, 'wb')
    fp.write(plaintext_bytes)
    fp.close()

    plaintext = plaintext_bytes.decode('utf-8')
    print(plaintext)

识别二进制或纯文本krc例子

def _is_utf8_char(buffer, idx):
    buffer_len = len(buffer)
    if buffer_len == 0:
        return False

    def _scan_remain(_s):
        for i in range(idx + 1, min(buffer_len, idx + _s)):
            if (buffer[i] & 0xC0) != 0x80:
                return 0
        return _s

    if (buffer[idx] & 0x80) == 0x00:
        return 1
    elif (buffer[idx] & 0xE0) == 0xC0:
        return _scan_remain(2)
    elif (buffer[idx] & 0xF0) == 0xE0:
        return _scan_remain(3)
    elif (buffer[idx] & 0xF8) == 0xF0:
        return _scan_remain(4)
    else:
        return 0
            
def _scan_utf8_buffer(buffer):
    i = 0
    length = 0
    max_length = 20

    while i < len(buffer):
        flag = _is_utf8_char(buffer, i)
        if flag == 0:
            return False

        length += 1
        if length >= max_length:
            break

        i += flag

    return True

with open('song.krc', 'rb') as f:
    if f.peek()[:3] == b'krc' and not _scan_utf8_buffer(f.peek()[3:]):
        print('binary')
        # 解密 & 解压...
        # raw_bytes = ...      
    else:
        print('text')
        raw_bytes = f.read()
    
    # plaintext = raw_bytes.decode('utf-8')