electron の pak文件解包解析

一个electron打包的项目，除了asar还有pak文件，解包后感觉类似资源文件，使用到的工具如下:

pak_mingw64

推荐pak_mingw64：https://github.com/myfreeer/chrome-pak-customizer

pak_mingw64.exe  -u *.pak  ./unpack

UnrealPakViewer

https://github.com/jashking/UnrealPakViewer/releases/tag/v1.1

umodel

PakerUnpaker

grit

参考：

https://blog.csdn.net/weixin_30894583/article/details/97602164

https://stackoverflow.com/questions/10633357/how-to-unpack-resources-pak-from-google-chrome

https://www.cnblogs.com/jeason1997/p/6853737.html

git clone https://chromium.googlesource.com/chromium/src/tools/grit

直接可使用

python data_pack.py *.pak

但解压出来就2个文件，应该跟代码中读取数量或者其他参数有关，有兴趣的可以看一下，且目前代码跟参考文件中不一样，可能有些关系

[Python] pak文件解析代码

import os
import struct

class Binary:
    def __init__(self, data: [bytearray, bytes, str]):
        self.index = 0
        if isinstance(data, bytearray):
            self.data = data
        elif isinstance(data, bytes):
            self.data = bytearray(data)
        elif isinstance(data, str):
            self.data = bytearray(data.encode(encoding='utf-8'))
        else:
            raise TypeError('data type is invalid:', data.__class__.__name__)
        self.size = len(self.data)

    def read_int8(self) -> int:
        ret = self.data[self.index]
        self.index += 1
        return ret

    def read_int16(self) -> int:
        ret, = struct.unpack('h', self.data[self.index:self.index + 2])
        self.index += 2
        return ret

    def read_int32(self) -> int:
        ret, = struct.unpack('i', self.data[self.index:self.index + 4])
        self.index += 4
        return ret

    def read_int64(self) -> int:
        ret, = struct.unpack('q', self.data[self.index:self.index + 8])
        self.index += 8
        return ret

    def read_string(self) -> str:
        pass

    def read_bytes(self, size: int) -> bytearray:
        if size > self.rest():
            ret = self.data[self.index:]
            self.index += self.rest()
            return ret
        ret = self.data[self.index:self.index + size]
        self.index += size
        return ret

    def __len__(self):
        return len(self.data)

    def rest(self) -> int:
        return len(self.data) - self.index

    @staticmethod
    def bytes2str(data: [bytearray, bytes]):
        out = ''
        if data is None or len(data) == 0:
            return 'Null'
        for b in data:
            if out:
                out += ','
            out += '0x%02X' % b
        out += '\n'
        out += data.decode('utf-8')
        return out

class PakData:
    def __init__(self, br: Binary):
        self.version = br.read_int32()
        self.count = br.read_int32()
        self.coding = br.read_int8()
        self.resources = list()
        for i in range(self.count):
            item = {
                'id': br.read_int16(),
                'offset': br.read_int32(),
                'data': bytearray(),
                'size': 0
            }
            self.resources.append(item)
        for i, item in enumerate(self.resources):
            if i + 1 < self.count:
                item['size'] = self.resources[i + 1]['offset'] - item['offset']
            else:
                item['size'] = len(br) - item['offset']
            item['data'] = br.data[item['offset']:item['offset'] + item['size']]

class PakFile:
    def __init__(self, path: str):
        self.path = path
        if os.path.exists(path):
            self.decode()
            return
        self.data = None

    def set_path(self, path):
        self.path = path

    def decode(self) -> int:
        if os.path.exists(self.path) is False:
            return -1
        with open(self.path, mode='rb') as file:
            raw = bytearray(file.read())
            file.close()
            br = Binary(raw)
            self.data = PakData(br)

def main():
    pak = PakFile(r'.\zh-CN.pak')
    if isinstance(pak.data, PakData):
        print('版本:', pak.data.version)
        print('数量:', pak.data.count)
        print('编码:', pak.data.coding)
        for i in range(pak.data.count):
            print('ID:', pak.data.resources[i]['id'])
            print('偏移:', pak.data.resources[i]['offset'])
            print('大小:', pak.data.resources[i]['size'])
            print('内容:', Binary.bytes2str(pak.data.resources[i]['data']))

if __name__ == '__main__':
    main()

代码来自52pojie，不过测试的时候报字符错，修改了gbk gbk2312都不行，先记录下：