前言
最近在研究某某app的数据库，发现自己在so层的调试比较薄弱，专门找了看雪的CTF-变形金刚来学习。希望在用ida调试so方面有所突破。
利用国庆期间整理成笔记。技术不成熟或许描述的不够清晰请大伙见谅。
也拜读了几位大佬的文章。
工具准备
- ida7.0
- Transformers.apk
- idea
- jadx
- frida
- 小米4root手机（android6.0）
app分析目标
为了熟悉该app，先介绍一下大神分析后的结果。
输入错误密码截图
比如输入密码是：12345678，会提示错误信息：Transformers:error
密码错误登陆失败的截图
输入正确密码截图
输入密码(长度16个字符)是：fu0kzHp2aqtZAuY6，会提示错误信息：Transformers:flag{android4-9}
正确密码提示
静态分析-java
把Transformers.apk拖入jadx
障眼法1:该app设计作者采用了障眼法，很容易欺骗分析人员，让我们认为OnClick的回调处理逻辑在MainActivity中。
障眼法2：在MainActivity的OnCreate中欺骗代码注册OnClick回调。主要原理是利用Activity的onStart事件会晚与OnCreate事件，从而实现在OnStart中注册的OnClick回调覆盖了OnCreate中注册的回调。
MainActivity障眼法OnCreate事件
MainActivity基类也是障眼法
Activity基类的onStart才是真正注册OnClick回调的地方
Activity基类的onStart事件才是逻辑代码
Base64解码图
为了方便理解eq中涉及的魔改的Base64编码，这里先贴上一直Base64解码的草稿图
下图是Base64编码4个字符'{6*的解码草稿图，画的不是很好，主要是方便我自己理解写解码Base64的逻辑代码。
base64解码草图
它是用64个可打印字符表示二进制所有数据方法。由于2的6次方等于64，所以可以用每6个位元为一个单元，对应某个可打印字符。我们知道三个字节有24个位元，就可以刚好对应于4个Base64单元，即3个字节需要用4个Base64的可打印字符来表示。
静态分析与动态调试跟踪liboo000oo.so
这里将分3个关键步骤：
- 1：JNI_OnLoad分析动态注册eq对应的函数sub_784，稍后再详细介绍。
- 2：.fini_array部分跟踪，主要是eq函数用到的一下基础数据，比如Base64编码表、RC4加密算法用到的key初始值，之后它会通过一系列异或处理算出真正的密钥key。也稍后再详细介绍。
- 3：eq函数的算法跟踪分析。
这里先列一下分析结果
JNI_onLoad部分==========
eq是由sub_784动态注册
ini_array部分==========
#解码36长度字符串byte_4020：
byte_4020 =650f909c-7217-3647-9331-c82df8b98e98+0x00（结束符）
#解码Base64为的64+1个编码字符byte_4050=
base64Chars = byte_4050 =!:#$%&()+-*/`~_[]{}?<>,.@^abcdefghijklmnopqrstuvwxyz0123456789\';+0x00（结束符）
#app伪装java类的名称byte_40A0：
className_Sign= android/support/v7/app/AppCompiatActivity +0x00（结束符）
sub_784部分============
获取RC4的密钥算法============：
v5=算法01 = 删除(byte_4040中的"-") =650f909c721736479331c82df8b98e98
v4=算法02 =删除byte_4020中的”-“后，特殊运算又加上”-“符号=89e89b8f-d28c-1339-7463-7127c909f056
自定义算法03==========
v6=v17=自定义算法03 = 特殊算法处理后 =36f36b3c-a03e-4996-8759-8408e626c215
算法参数01：v4
算法参数02：unk_23DE(10位长度) =2409715836
算法参数03：unk_23D8（16位长度）=dbeafc2409715836
RC4魔改部分====== （已经逆向算法）
RC4第1步：v48:拷贝&unk_B4D863E8中的256个字符到v48(应视是RC4算法的魔改数据)
RC4第2步：v49:计算RC4 的临时256自己T向量，公式：iK[i]=(byte)aKey.charAt((i % aKey.length()));
RC4第3步：V48状态向量S进行置换操作V48
RC4第4步：RC4产生密钥流iOutputChar,然后return new String(iOutputChar)？？？？？？
v27 =33 ,即v6[3];
RC4第5步：产生密钥流--其他还结合魔改Base64运算
分配输入密码对应的base64字节的存储空间
Base64算法的字典
Base64魔改部分======（已经逆向算法）
Base64的字典部分-64个字符+1个结束符
每4个字符中第0个与0x6异或，第0个与0xF异或
switch (i%4){
case 0:
base64char = (char) (iAscii ^0x07);
break;
case 2:
base64char = (char) (iAscii ^0xF);
break;
default:
base64char =encodeChars[i];
}
JNI_OnLoad 代码逻辑分析RegisterNatives
1：搜索JNI_OnLoad，双击进入函数体
2：导入Jni.h文件。菜单路径：ida/File/Load file/Parse C header file
3：修改参数类型为_JavaVM
4：按g快捷键直接跳转到0x4014
JNI_OnLoad截图
右键菜单中可选择
找到eq注册对应的sub_784函数
sub_784其实就是eq的真正函数名称
当然frida的hook_art_so_register.js脚本快速定位eq对应sub_784
该脚本通过hook art.so 的register函数，打印动态注册的地址
//调用方式==测试ok
frida -U --no-pause -f package_name -l hook_art_so_register.js
跟踪.fini_array（其实这步可以跳过）
另外根据so文件的加载流程应该是先加载init_array，然后是JNI_OnLoad
ida 中ctrl+s 进入init_array，会先执行.datadiv_decode5009363700628197108，其实就是通过一些运算得到一些初始化数据。
按f5 采用伪代码方式查看datadiv_decode5009363700628197108，根据得到一些数据。
1：解码36长度字符串byte_4020：650f909c-7217-3647-9331-c82df8b98e98+0x00（结束符）
2：魔改Base64编码的table表，为的64+1个编码字符byte_4050= !:#$%&()+-*/`~_[]{}?<>,.@^abcdefghijklmnopqrstuvwxyz0123456789\';+0x00（结束符）
init_array函数解码初始数据
eq函数的分析（即sub_784函数）
ida定位到sub_784，修改参数类型，并做了动态调试后，梳理sub_784的逻辑。
sub_784逻辑梳理
unk_23E8字符串（RC4魔改算法的初始化256个字符S状态向量）
unk_23E8字符串Base64的table
unk_23E8字符串Base64的table 拷贝
unk_23E8字符串Base64的table 拷贝
unk_23E8=Sbox = [0xD7,0xDF,0x02,0xD4,0xFE,0x6F,0x53,0x3C,0x25,0x6C,0x99,0x97,0x06,0x56,0x8F,0xDE,0x40,0x11,0x64,0x07,0x36,0x15,0x70,0xCA,0x18,0x17,0x7D,0x6A,0xDB,0x13,0x30,0x37,0x29,0x60,0xE1,0x23,0x28,0x8A,0x50,0x8C,0xAC,0x2F,0x88,0x20,0x27,0x0F,0x7C,0x52,0xA2,0xAB,0xFC,0xA1,0xCC,0x21,0x14,0x1F,0xC2,0xB2,0x8B,0x2C,0xB0,0x3A,0x66,0x46,0x3D,0xBB,0x42,0xA5,0x0C,0x75,0x22,0xD8,0xC3,0x76,0x1E,0x83,0x74,0xF0,0xF6,0x1C,0x26,0xD1,0x4F,0x0B,0xFF,0x4C,0x4D,0xC1,0x87,0x03,0x5A,0xEE,0xA4,0x5D,0x9E,0xF4,0xC8,0x0D,0x62,0x63,0x3E,0x44,0x7B,0xA3,0x68,0x32,0x1B,0xAA,0x2D,0x05,0xF3,0xF7,0x16,0x61,0x94,0xE0,0xD0,0xD3,0x98,0x69,0x78,0xE9,0x0A,0x65,0x91,0x8E,0x35,0x85,0x7A,0x51,0x86,0x10,0x3F,0x7F,0x82,0xDD,0xB5,0x1A,0x95,0xE7,0x43,0xFD,0x9B,0x24,0x45,0xEF,0x92,0x5C,0xE4,0x96,0xA9,0x9C,0x55,0x89,0x9A,0xEA,0xF9,0x90,0x5F,0xB8,0x04,0x84,0xCF,0x67,0x93,0x00,0xA6,0x39,0xA8,0x4E,0x59,0x31,0x6B,0xAD,0x5E,0x5B,0x77,0xB1,0x54,0xDC,0x38,0x41,0xB6,0x47,0x9F,0x73,0xBA,0xF8,0xAE,0xC4,0xBE,0x34,0x01,0x4B,0x2A,0x8D,0xBD,0xC5,0xC6,0xE8,0xAF,0xC9,0xF5,0xCB,0xFB,0xCD,0x79,0xCE,0x12,0x71,0xD2,0xFA,0x09,0xD5,0xBC,0x58,0x19,0x80,0xDA,0x49,0x1D,0xE6,0x2E,0xE3,0x7E,0xB7,0x3B,0xB3,0xA0,0xB9,0xE5,0x57,0x6E,0xD9,0x08,0xEB,0xC7,0xED,0x81,0xF1,0xF2,0xBF,0xC0,0xA7,0x4A,0xD6,0x2B,0xB4,0x72,0x9D,0x0E,0x6D,0xEC,0x48,0xE2,0x33]
byte_24E8指向的内容
byte_24E8指向的内容
把sub_784业务逻辑直接采用在ida中注解的方式修订如下
int __fastcall sub_784(_JNIEnv *a1_env, int a2, void *a3_str_app_password)
{
size_t v3_4020_RC4_key_length; // r10
unsigned __int8 *v4_4020_del_handle02; // r6
_BYTE *v5_4020_del_hyphen_handle01; // r8
_BYTE *v6_byte_4020_handle3; // r11
int v7; // r0
size_t v8; // r2
char *v9; // r1
int v10; // r3
int v11_len; // r1
unsigned int v12; // r2
int v13; // r3
int v14; // r0
int v15_forJump_; // r4
unsigned __int8 v16; // r0
_BYTE *v17_4020_handle3_ptr; // r3
_BYTE *v18_tmp_byte_move; // r5
char *v19_v49_256_Rc4_T; // r4
int v20; // r5
int v21; // r1
int v22; // r0
signed int v23; // r1
int v24_S; // r2
size_t v25_app_password_length; // r0
unsigned int v26_app_password_length; // r8
unsigned int v27; // r5
_BYTE *v28_new_base64; // r0
int v29; // r3
int v30; // r10
unsigned int v31_pw_index; // r2
int v32; // r12
bool v33_isAppPW_index01; // zf
_BYTE *v34_ret; // r1
bool v35; // zf
int v36; // r3
int v37_tmp; // r1
unsigned __int8 v38_char_needTodoBase64; // r11
unsigned int v39; // lr
char v40; // r1
char *v41; // r2
int v42; // t1
unsigned int v44_new_base64_length; // [sp+4h] [bp-234h]
unsigned int v45_by_app_password_length; // [sp+8h] [bp-230h]
unsigned int v46; // [sp+10h] [bp-228h]
char *s_char_app_password; // [sp+14h] [bp-224h]
char v48_256_Rc4_S[256]; // [sp+18h] [bp-220h] RC4的S向量
char v49_256_Rc4_T[256]; // [sp+118h] [bp-120h] RC4的T向量
int v50; // [sp+218h] [bp-20h]
s_char_app_password = (char *)a1_env->functions->GetStringUTFChars(&a1_env->functions, a3_str_app_password, 0);
v3_4020_RC4_key_length = strlen(byte_4020); // 长度0x24，即36长度
v4_4020_del_handle02 = (unsigned __int8 *)malloc(v3_4020_RC4_key_length);// byte_4020被删除"-"符号，再经过特殊处理，同时有添加回新的“-”符号
// 新的结果：89e89b8f-d28c-1339-7463-7127c909f056
v5_4020_del_hyphen_handle01 = malloc(v3_4020_RC4_key_length);// byte_4020中的内容删除“-” ：650f909c721736479331c82df8b98e98
v6_byte_4020_handle3 = malloc(v3_4020_RC4_key_length);
_aeabi_memclr(v4_4020_del_handle02, v3_4020_RC4_key_length);
_aeabi_memclr(v5_4020_del_hyphen_handle01, v3_4020_RC4_key_length);
_aeabi_memclr(v6_byte_4020_handle3, v3_4020_RC4_key_length);
if ( v3_4020_RC4_key_length )
{
v7 = 0;
v8 = v3_4020_RC4_key_length;
v9 = byte_4020; // 指向36个字符串长度，在调试.fini_array已经得到：650f909c-7217-3647-9331-c82df8b98e98
do // 其实就是把数据byte_4020的asci码copy到v5中byte数组
{
v10 = (unsigned __int8)*v9++; // 取一个字符转为asci码
if ( v10 != 45 ) // "-"的asci=45,其实就是删除“-”符号，复制byte_4020（包括“-”符号）
v5_4020_del_hyphen_handle01[v7++] = v10;
--v8; // 及长度减一
}
while ( v8 ); // 循环体结束后v5=删除(byte_4040中的"-") =650f909c721736479331c82df8b98e98
// ===========================算法处理01
if ( v7 >= 1 ) // v7 =0x20;即字符串32为长度,其实就是36位byte_4020删除“-”后的长度
{
v11_len = v7 - 1;
v12 = -8; // 0xFFFFFFF8，调试>>结果发现这里应该不是-8，而是一个很大的数
v13 = 0;
v14 = 0;
do
{
if ( (v13 | (v12 >> 2)) > 3 ) // v12>>2 = -2?后第一次=0x3FFFFFFE
{
v15_forJump_ = v14;
}
else
{
v15_forJump_ = v14 + 1;
v4_4020_del_handle02[v14] = 45; // 其实就是“-”字符的asc码45
}
v16 = v5_4020_del_hyphen_handle01[v11_len--];// v5其实就是byte_4020删除了“-”符号后的：650f909c721736479331c82df8b98e98
v13 += 0x40000000;
v4_4020_del_handle02[v15_forJump_] = v16;
++v12;
v14 = v15_forJump_ + 1; // 其实也就是跳过"-"
}
while ( v11_len != -1 ); // 该循环体计算v4的结果是：89e89b8f-d28c-1339-7463-7127c909f056
// 处理逻辑是：byte_4020被删除"-"符号，再经过特殊处理，同时有添加回新的“-”符号
// ===========================算法处理02
if ( v15_forJump_ >= 0 ) // 第一次变量是0x23,即36-1
{
v17_4020_handle3_ptr = v6_byte_4020_handle3;// 初始是0x00,v17计算后指向RC4的key：36f36b3c-a03e-4996-8759-8408e626c215
while ( 1 )
{
v18_tmp_byte_move = (_BYTE *)*v4_4020_del_handle02;// ascii的hex：第一次 =0x38(对应字符"8");第2次 =0x39(对应字符"9");就是v4逐个字节取出
if ( (unsigned __int8)((_BYTE)v18_tmp_byte_move - 97) <= 5u )// ascii(97)='a';ascii(48)='0'
break;
if ( (unsigned __int8)((_BYTE)v18_tmp_byte_move - 48) <= 9u )
{
v18_tmp_byte_move = (char *)&unk_23DE + (_DWORD)v18_tmp_byte_move - 48;// unk_23DE(10位长度)指向：2409715836
goto LABEL_18;
}
LABEL_19:
*v17_4020_handle3_ptr++ = (_BYTE)v18_tmp_byte_move;// 存放到v17指向的地址
--v14; // 本if语句中初始是0x24,即36
++v4_4020_del_handle02; // 移动v4指向的指针，逐个取出
if ( !v14 ) // 如果为0退出循环体？？？
// if语句内部的循环体;但还在if语句内
goto LABEL_20;
} // end for while ( 1 )
//
v18_tmp_byte_move = (char *)&unk_23D8 + (_DWORD)v18_tmp_byte_move - 97;// unk_23D8（16位长度）好像是什么密码指向：dbeafc2409715836
LABEL_18:
LOBYTE(v18_tmp_byte_move) = *v18_tmp_byte_move;// 这个是IDA的常用宏,相当于取变量的最低byte位来赋值
goto LABEL_19;
} // end if ( v15_forJump_ >= 0 )
// ？？？？？？这个if语句里到底处理了哪些逻辑
// 本if语句结束用于计算v6
// 计算结束后
// v6 = 36f36b3c-a03e-4996-8759-8408e626c215
// v17 = v6的指针进行移动
// v18 = v4的字节逐个取出。（会与unk_23D8的内容进行运算）
// 涉及参数：
// 1：v4,即算法1-2步骤后(byte_4020)：去”-“后计算又重新加上”-“符号=89e89b8f-d28c-1339-7463-7127c909f056
// 2：unk_23DE(10位长度)指向：2409715836
// 3：unk_23D8（16位长度）指向：dbeafc2409715836
// ===========================算法处理03--获取RC4的密钥
} // end if ( v7 >= 1 )
} // end for if ( v3_RC4_key_length ) 处理后得到key*v17（即v6） 是：36f36b3c-a03e-4996-8759-8408e626c215
LABEL_20:
_aeabi_memcpy8(v48_256_Rc4_S, &unk_23E8, 256);// 用于RC4魔改算法（魔改）-256字节初始化：
// unk_23E8指向的是256个字符：-拷贝到v48
// 256字符不会重复
// bytes:0xD7 0xDF 0x02 0xD4 0xFE 0x6F 0x53 0x3C ....
// chars:.....oS<%l...V....d.6.p.
// R0 =v48地址=0xBEE74D38
// =================算法处理04===RC4-01初始化256个字节的状态向量
v19_v49_256_Rc4_T = v49_256_Rc4_T;
v20 = 0;
do
{
sub_D20(v20, v3_4020_RC4_key_length); // ？？？ 具体做了哪些工作需要跟进去看看
v49_256_Rc4_T[v20++] = v6_byte_4020_handle3[v21];// 根据算出的key再转换临时向量T
}
while ( v20 != 256 ); // 运算v49的值--RC4魔改的另外一个256字节向量-临时向量T
// 循环体算法是：
// for (short i= 0;i<256;i++)//初始化S和T(根据密钥mkey)
// {
// iK[i]=(byte)aKey.charAt((i % aKey.length()));//i%密钥长度（取值范围为1-256）；
// }
//
// =================参数：RC4密钥长度，即base_4020长度
// 循环体结束后===
// v48：内容没有被修改
// v49(只复制256字符) =36f36b3c-a03e-4996-8759-8408e626c21536f36b3c-a03e-4996-8759-8408e626c21536f36b3c-a03e-4996-8759-8408e626c21536f36b3c-a03e-4996-8759-8408e626c21536f36b3c-a03e-4996-8759-8408e626c21536f36b3c-a03e-4996-8759-8408e626c21536f36b3c-a03e-4996-8759-8408e626c21536f3
// =================算法处理04-===RC4-02初始化256个字节的临时向量
//
//
v22 = (unsigned __int8)(v49_256_Rc4_T[0] - 41);// #0x33（即51）-41=0xA(即10)
// 正常的RC4这里是0；也就是说这里也被魔改了****
v48_256_Rc4_S[0] = v48_256_Rc4_S[v22];
v48_256_Rc4_S[v22] = -41;
v23 = 1; // 正常的RC4这里是0；也就是说这里也被魔改了1；也就是第一个byte不处理***
do // RC4向量--初始排列S
{
v24_S = (unsigned __int8)v48_256_Rc4_S[v23];
v22 = (v22 + (unsigned __int8)v49_256_Rc4_T[v23] + v24_S) % 256;
v48_256_Rc4_S[v23++] = v48_256_Rc4_S[v22];
v48_256_Rc4_S[v22] = v24_S;
}
while ( v23 != 256 ); // =================算法处理04===RC-03===开始对状态向量S进行置换操作（用来打乱初始种子1）
// 其实就是打乱v48
// 魔改代码：
//
// j=10; //======注意=====正常这里是0，不过魔改的2019的CTF=10?????????
// iS_48[0] =iS_48[j];//======注意=====这段是魔改零添加的
// iS_48[j] = -41; //======注意=====这段是魔改零添加的
//
//
// for (int i=1;i<255;i++)//初始排列for (int i=0;i<255;i++) //2019的CTF魔改for (int i=1;i<256;i++)
// {
// j=(j+iS_48[i]+iK_49[i]) % 256;
// int temp = iS_48[i];
// iS_48[i]=iS_48[j];
// iS_48[j]=temp;
// }
v25_app_password_length = strlen(s_char_app_password);// 就是解密数组密码的字符串:"12345678".lenght =8
v26_app_password_length = v25_app_password_length;
v27 = (unsigned __int8)v6_byte_4020_handle3[3];// v27=0x33 =51
v45_by_app_password_length = 8 * (3 - -3 * (v25_app_password_length / 3));
v44_new_base64_length = v27 + v45_by_app_password_length / 6;
v28_new_base64 = malloc(v44_new_base64_length + 1);// 分配新的base64需要使用的字节。
// //==========Base64分配需要新的字节大小
if ( v26_app_password_length ) // app输入的密码长度
{
v30 = 0;
v31_pw_index = 0; // 循环体的index
v32 = 0;
v46 = v27;
do
{
v30 = (v30 + 1) % 256; // // RC4 算法的产生密钥流循环体开始
v37_tmp = (unsigned __int8)v48_256_Rc4_S[v30];
v32 = (v32 + v37_tmp) % 256;
v48_256_Rc4_S[v30] = v48_256_Rc4_S[v32];
v48_256_Rc4_S[v32] = v37_tmp; // ==========刚刚重新分析到这里，苦逼呀
v19_v49_256_Rc4_T = (char *)(unsigned __int8)v48_256_Rc4_S[v30];
v38_char_needTodoBase64 = v48_256_Rc4_S[(unsigned __int8)(v37_tmp + (_BYTE)v19_v49_256_Rc4_T)] ^ s_char_app_password[v31_pw_index];//
// ===========Base64内嵌在RC4的内部
// 先线程一个RC4加密后的字符
// 后面用这个字符进行Base64的解码操作
// Base64魔改的算法：
// 1：Base64 字典被替换
// 2：Base最后以为“=”会被替换成“;”
// 3: 对特定字符进行异或操作.每4个字符中第0个与0x6异或，第0个与0xF异或
// switch (i%4){
// case 0:
// base64char = (char) (iAscii ^0x07);
// break;
//
// case 2:
// base64char = (char) (iAscii ^0xF);
// break;
// default:
// base64char =encodeChars[i];
//
// }
if ( v31_pw_index
&& (v29 = 2863311531u * (unsigned __int64)v31_pw_index >> 32, v39 = 3 * (v31_pw_index / 3), v39 != v31_pw_index) )//
// 逗号运算符是指在C语言中，多个表达式可以用逗号分开，其中用逗号分开的表达式的值分别结算，但整个表达式的值是最后一个表达式的值。
// 影响结果：
// false:其实就是index/3(取摸)!=0
// true: index/3 =0
// v39:index可以是3的最大倍数，0，3，6，9，12...
// index=3,则v39=3;index =6,则v39=6;index=7,则v39=6
{
v33_isAppPW_index01 = v31_pw_index == 1;// 是否是appPassword[1]字符
if ( v31_pw_index != 1 )
v33_isAppPW_index01 = v39 + 1 == v31_pw_index;
if ( v33_isAppPW_index01 ) // 判断index%3=1的情况 ，即1，4，7，10
{
v34_ret = byte_4050; // v34第一次=0x33(51)；
// byte_4050：存放的是base64的table字典字符64+1个
// byte_4050（字符长度是64+1）：!:#$%&()+-*/`~_[]{}?<>,.@^abcdefghijklmnopqrstuvwxyz0123456789\';
v28_new_base64[v46 + v31_pw_index] = byte_4050[(unsigned __int8)v28_new_base64[v46 + v31_pw_index] | ((unsigned int)v38_char_needTodoBase64 >> 4)];
v19_v49_256_Rc4_T = &v28_new_base64[v46 + v31_pw_index];
v29 = 4 * v38_char_needTodoBase64 & 0x3C;// 0x3C =60 ,即111100二进制
v19_v49_256_Rc4_T[1] = v29;
if ( v31_pw_index + 1 >= v26_app_password_length )
goto LABEL_53; // 跳转到程序的倒数第2个Lable
}
else // index%3=2的情况 2，5，8，11....
{
v35 = v31_pw_index == 2;
if ( v31_pw_index != 2 )
v35 = v39 + 2 == v31_pw_index;
if ( v35 )
{
v19_v49_256_Rc4_T = (char *)(v38_char_needTodoBase64 & 0xC0);// 0xC0 =192即11000000二进制，保留最高2bit
v36 = v46++ + v31_pw_index;
v28_new_base64[v36] = byte_4050[(unsigned __int8)v28_new_base64[v36] | ((unsigned int)v19_v49_256_Rc4_T >> 6)] ^ 0xF;// 0xf=15 ,即二进制1111。
// 这里异或好像是对Base64模拟的多余操作
v29 = (int)&v28_new_base64[v36];
*(_BYTE *)(v29 + 1) = byte_4050[v38_char_needTodoBase64 & 0x3F];// 0x3f=63 即111111二进制
}
}
}
else // index/%3 =0 即0,3,6,9...
{
v28_new_base64[v46 + v31_pw_index] = byte_4050[(unsigned int)v38_char_needTodoBase64 >> 2] ^ 7;// 7的二进制四111
// 这里对base64进行了魔改操作
v19_v49_256_Rc4_T = &v28_new_base64[v46 + v31_pw_index];
v29 = 16 * v38_char_needTodoBase64 & 0x30;// 0x30=48,即二进制110000bit
v19_v49_256_Rc4_T[1] = v29;
if ( v31_pw_index + 1 >= v26_app_password_length )// 最后一位
{
v40 = byte_4050[v29];
*((_WORD *)v19_v49_256_Rc4_T + 1) = 15163;
goto LABEL_43;
}
}
++v31_pw_index;
} // end for if ( v26_app_password_length ) 内部的do
while ( v31_pw_index < v26_app_password_length );
} // end if ( v26_strText_length ) 输入密码lenght>0
// ========这是时RC4-04部分的算法“产生密钥流”
// v28_new_base64：同时结合if部分逐个对RC4生成的结果做魔改的Base64运算
//
//
//
while ( 1 )
{
if ( v45_by_app_password_length )
{ // v45 是根据输入密码长度计算，比如密码：12345678；v45=0x48u(即78)
v34_ret = (_BYTE *)(&dword_0 + 1); // v34 =1
v19_v49_256_Rc4_T = (char *)v44_new_base64_length;
v41 = &byte_24E8; // v41指向的char* = 0x20(即空格)+"{9*8ga*l!Tn?@#fj'j$\g;;"
do
{
v29 = (unsigned __int8)v28_new_base64[v27++];
v42 = (unsigned __int8)*v41++;
if ( v42 != v29 )
v34_ret = 0;
}
while ( v27 < v44_new_base64_length ); // v44 =0x3Fu
}
else
{
v34_ret = (_BYTE *)(&dword_0 + 1);
}
v28_new_base64 = (_BYTE *)(_stack_chk_guard - v50);
if ( _stack_chk_guard == v50 )
break;
LABEL_53:
v40 = v34_ret[v29];
v19_v49_256_Rc4_T[2] = 52; // 输入密码：12345678，这里是：wk4.
LABEL_43:
v19_v49_256_Rc4_T[1] = v40;
}
return (unsigned __int8)v34_ret; // sub_784 方法返回只有这么一个位置。我们侧重分析如何让该方法返回真即可
}
RC4Util（魔改RC4算法）
package com.younghare.utils;
/**
* 用 Java 实现的 Rc4 加密算法
* https://blog.csdn.net/shuxing520/article/details/79096032?utm_source=app
* RC4加密算法的原理及实现 :https://blog.csdn.net/lc_910927/article/details/37599161
*/
public class RC4Util {
/**
* 加密和解密都用这一个方法。也就是说参数String aInput 可以传一个明文，也可以传一个加密后的字符串，程序会自动的识别。然后执行加解密的响应操作。
* @param aInput
* @param aKey_RC4
* @return
*/
public static String HloveyRC4(String aInput,String aKey_RC4)
{
int[] iS_48 = new int[256];//状态向量S：长度为256，S[0],S[1].....S[255]。每个单元都是一个字节，算法运行的任何时候，S都包括0-255的8比特数的排列组合，只不过值的位置发生了变换；
byte[] iK_49 = new byte[256];//临时向量T(K)：长度也为256，每个单元也是一个字节。如果密钥的长度是256字节，就直接把密钥的值赋给T，否则，轮转地将密钥的每个字节赋给T；
for (int i=0;i<256;i++)//初始化S，默认初始化是0-255
iS_48[i]=i;//这里经常被用来魔改 ，
//魔改的情况，即让上面的iS初始化失效
iS_48 = new int[] {0xD7,0xDF,0x02,0xD4,0xFE,0x6F,0x53,0x3C,0x25,0x6C,0x99,0x97,0x06,0x56,0x8F,0xDE,0x40,0x11,0x64,0x07,0x36,0x15,0x70,0xCA,0x18,0x17,0x7D,0x6A,0xDB,0x13,0x30,0x37,0x29,0x60,0xE1,0x23,0x28,0x8A,0x50,0x8C,0xAC,0x2F,0x88,0x20,0x27,0x0F,0x7C,0x52,0xA2,0xAB,0xFC,0xA1,0xCC,0x21,0x14,0x1F,0xC2,0xB2,0x8B,0x2C,0xB0,0x3A,0x66,0x46,0x3D,0xBB,0x42,0xA5,0x0C,0x75,0x22,0xD8,0xC3,0x76,0x1E,0x83,0x74,0xF0,0xF6,0x1C,0x26,0xD1,0x4F,0x0B,0xFF,0x4C,0x4D,0xC1,0x87,0x03,0x5A,0xEE,0xA4,0x5D,0x9E,0xF4,0xC8,0x0D,0x62,0x63,0x3E,0x44,0x7B,0xA3,0x68,0x32,0x1B,0xAA,0x2D,0x05,0xF3,0xF7,0x16,0x61,0x94,0xE0,0xD0,0xD3,0x98,0x69,0x78,0xE9,0x0A,0x65,0x91,0x8E,0x35,0x85,0x7A,0x51,0x86,0x10,0x3F,0x7F,0x82,0xDD,0xB5,0x1A,0x95,0xE7,0x43,0xFD,0x9B,0x24,0x45,0xEF,0x92,0x5C,0xE4,0x96,0xA9,0x9C,0x55,0x89,0x9A,0xEA,0xF9,0x90,0x5F,0xB8,0x04,0x84,0xCF,0x67,0x93,0x00,0xA6,0x39,0xA8,0x4E,0x59,0x31,0x6B,0xAD,0x5E,0x5B,0x77,0xB1,0x54,0xDC,0x38,0x41,0xB6,0x47,0x9F,0x73,0xBA,0xF8,0xAE,0xC4,0xBE,0x34,0x01,0x4B,0x2A,0x8D,0xBD,0xC5,0xC6,0xE8,0xAF,0xC9,0xF5,0xCB,0xFB,0xCD,0x79,0xCE,0x12,0x71,0xD2,0xFA,0x09,0xD5,0xBC,0x58,0x19,0x80,0xDA,0x49,0x1D,0xE6,0x2E,0xE3,0x7E,0xB7,0x3B,0xB3,0xA0,0xB9,0xE5,0x57,0x6E,0xD9,0x08,0xEB,0xC7,0xED,0x81,0xF1,0xF2,0xBF,0xC0,0xA7,0x4A,0xD6,0x2B,0xB4,0x72,0x9D,0x0E,0x6D,0xEC,0x48,0xE2,0x33};
//iK_49 = new int[] {0x33,0x36,0x66,0x33,0x36,0x62,0x33,0x63,0x2D,0x61,0x30,0x33,0x65,0x2D,0x34,0x39,0x39,0x36,0x2D,0x38,0x37,0x35,0x39,0x2D,0x38,0x34,0x30,0x38,0x65,0x36,0x32,0x36,0x63,0x32,0x31,0x35,0x33,0x36,0x66,0x33,0x36,0x62,0x33,0x63,0x2D,0x61,0x30,0x33,0x65,0x2D,0x34,0x39,0x39,0x36,0x2D,0x38,0x37,0x35,0x39,0x2D,0x38,0x34,0x30,0x38,0x65,0x36,0x32,0x36,0x63,0x32,0x31,0x35,0x33,0x36,0x66,0x33,0x36,0x62,0x33,0x63,0x2D,0x61,0x30,0x33,0x65,0x2D,0x34,0x39,0x39,0x36,0x2D,0x38,0x37,0x35,0x39,0x2D,0x38,0x34,0x30,0x38,0x65,0x36,0x32,0x36,0x63,0x32,0x31,0x35,0x33,0x36,0x66,0x33,0x36,0x62,0x33,0x63,0x2D,0x61,0x30,0x33,0x65,0x2D,0x34,0x39,0x39,0x36,0x2D,0x38,0x37,0x35,0x39,0x2D,0x38,0x34,0x30,0x38,0x65,0x36,0x32,0x36,0x63,0x32,0x31,0x35,0x33,0x36,0x66,0x33,0x36,0x62,0x33,0x63,0x2D,0x61,0x30,0x33,0x65,0x2D,0x34,0x39,0x39,0x36,0x2D,0x38,0x37,0x35,0x39,0x2D,0x38,0x34,0x30,0x38,0x65,0x36,0x32,0x36,0x63,0x32,0x31,0x35,0x33,0x36,0x66,0x33,0x36,0x62,0x33,0x63,0x2D,0x61,0x30,0x33,0x65,0x2D,0x34,0x39,0x39,0x36,0x2D,0x38,0x37,0x35,0x39,0x2D,0x38,0x34,0x30,0x38,0x65,0x36,0x32,0x36,0x63,0x32,0x31,0x35,0x33,0x36,0x66,0x33,0x36,0x62,0x33,0x63,0x2D,0x61,0x30,0x33,0x65,0x2D,0x34,0x39,0x39,0x36,0x2D,0x38,0x37,0x35,0x39,0x2D,0x38,0x34,0x30,0x38,0x65,0x36,0x32,0x36,0x63,0x32,0x31,0x35,0x33,0x36,0x66,0x33};
int j_v32 = 1;
for (short i= 0;i<256;i++)//初始化S和T(根据密钥mkey)
{
iK_49[i]=(byte)aKey_RC4.charAt((i % aKey_RC4.length()));//i%密钥长度（取值范围为1-256）；
}
j_v32=10; //======注意=====正常这里是0，不过魔改的2019的CTF=10?????????
iS_48[0] =iS_48[j_v32];//======注意=====这段是魔改零添加的
iS_48[j_v32] = -41; //======注意=====这段是魔改零添加的
for (int i=1;i<255;i++)//初始排列for (int i=0;i<255;i++) //2019的CTF魔改for (int i=1;i<256;i++)
{
j_v32=(j_v32+iS_48[i]+iK_49[i]) % 256;
int temp = iS_48[i];
iS_48[i]=iS_48[j_v32];
iS_48[j_v32]=temp;
}
//产生密钥流
int i_v30=0;
j_v32=0;
char[] iInputChar = aInput.toCharArray();
char[] iOutputChar = new char[iInputChar.length];
for(short index = 0;index<iInputChar.length;index++)
{
i_v30 = (i_v30+1) % 256;
j_v32 = (j_v32+iS_48[i_v30]) % 256;
int temp_v37 = iS_48[i_v30];
iS_48[i_v30]=iS_48[j_v32];
iS_48[j_v32]=temp_v37;
int t = (iS_48[i_v30]+(iS_48[j_v32] % 256)) % 256;
int iY = iS_48[t];
char iCY = (char)iY;
int int_v38 = iInputChar[index] ^ iCY;//应该是得到逐个后进行base64转化计算。。。。求考证
iOutputChar[index] = (char) int_v38;
//iOutputChar[index] =(char)( iInputChar[index] ^ iCY) ;
}
return new String(iOutputChar);
}
static byte[] ints2bytes_foriS_48(int[] iS_48){
byte[] bytes= new byte[iS_48.length];
for (int i =0;i< iS_48.length;i++){
bytes[i]= (byte) iS_48[i];
}
return bytes;
}
public static void main(String[] args) {//测试RC4算法 main_ok
String inputStr ="做个好男人!!!";
String key = "abcdefg";
System.out.println("加密前："+inputStr);
String str = HloveyRC4(inputStr,key);
//打印加密后的字符串
System.out.println("加密后："+str);
//打印解密后的字符串
System.out.println("解密后："+HloveyRC4(str,key));
}
}
Base64_ctf
/**
* Utility to base64 encode and decode a string.
* @author Stephen Uhler
* @version 1.9, 02/07/24
* 代码来源：http://www.herongyang.com/Encoding/Base64-Sun-Java-Implementation.html
*/
public class Base64_ctf {
static byte[] encodeData;
//static String charSet = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
static String charSet = "!:#$%&()+-*/`~_[]{}?<>,.@^abcdefghijklmnopqrstuvwxyz0123456789\\';";
static char equalChar = ';';
static String equalChar2 = ";;";
static {
encodeData = new byte[64];
for (int i = 0; i<64; i++) {
byte c = (byte) charSet.charAt(i);
encodeData[i] = c;
}
}
private Base64_ctf() {}
/**
* base-64 encode a string
* @param s The ascii string to encode
* @returns The base64 encoded result
*/
public static String
encode(String s) {
return encode(s.getBytes());
}
/**
* base-64 encode a byte array
* @param src The byte array to encode
* @returns The base64 encoded result
*/
public static String
encode(byte[] src) {
return encode(src, 0, src.length);
}
/**
* 编码原理:将3个字节转换成4个字节( (3 X 8) = 24 = (4 X 6) )
* 先读入3个字节,每读一个字节,左移8位,再右移四次,每次6位,这样就有4个字节了
* base-64 encode a byte array
* @param src The byte array to encode
* @param start The starting index
* @param len The number of bytes
* @returns The base64 encoded result
*/
public static String encode(byte[] src, int start, int length) {
byte[] dst = new byte[(length+2)/3 * 4 + length/72];
int x = 0;
int dstIndex = 0;
int state = 0; // which char in pattern
int old = 0; // previous byte
int len = 0; // length decoded so far
int max = length + start;
for (int srcIndex = start; srcIndex<max; srcIndex++) {
x = src[srcIndex];
switch (++state) {//0,3,6
case 1:
dst[dstIndex++] = encodeData[(x>>2) & 0x3f]; //0x3f = 63 = 111111二进制，8bit右移动2bit，然后只需剩下的6bit，的值来当作新的编码
break;
case 2://1,4,7
dst[dstIndex++] = encodeData[((old<<4)&0x30) //0x30 =48 =110000二进制，就是上一个8bit左移4比他，清零最前面的bit。即得到的上一个字符的最后2bit
| ((x>>4)&0xf)]; //移去4bit，剩下高位的4bit 。
break;
case 3://2,5,8
dst[dstIndex++] = encodeData[((old<<2)&0x3C) //0x3c =60= 111100, 上一个字符的低位4bit
| ((x>>6)&0x3)];//下一个自己的最后高为2bit
dst[dstIndex++] = encodeData[x&0x3F];
state = 0;
break;
}
old = x;
// if (++len >= 72) {//奇怪这里添加回车换行后，base64解码就会 出问题呀
// dst[dstIndex++] = (byte) '\n'; //LF 即换行
// len = 0;
// }
}
/*
* now clean up the end bytes
*/
switch (state) {
case 1: dst[dstIndex++] = encodeData[(old<<4) & 0x30];//0x30 =48 =110000
dst[dstIndex++] = (byte) equalChar; //原来：'='
dst[dstIndex++] = (byte) equalChar;//原来：'='
break;
case 2: dst[dstIndex++] = encodeData[(old<<2) & 0x3c]; //0x3c =60= 111100
dst[dstIndex++] = (byte) equalChar;//原来：'='
break;
}
return new String(dst);
}
/**
* 解码原理:将4个字节转换成3个字节. 先读入4个6位(用或运算),每次左移6位,再右移3次,每次8位.
* A Base64 decoder. This implementation is slow, and
* doesn't handle wrapped lines.
* The output is undefined if there are errors in the input.
* @param s a Base64 encoded string
* @returns The byte array eith the decoded result
*/
public static byte[] decode(String s) {
int end = 0; // end state
if (s.endsWith(equalChar+"")) { //原来"="
end++;
}
if (s.endsWith(equalChar2)) {//原来"=="
end++;
}
int len = (s.length() + 3)/4 * 3 - end;
byte[] result = new byte[len];
int dst = 0;
try {
for(int src = 0; src< s.length(); src++) {
int code = charSet.indexOf(s.charAt(src));
if (code == -1) {
break;
}
switch (src%4) {
case 0:
result[dst] = (byte) (code<<2); //先产生高6bit
break;
case 1:
result[dst++] |= (byte) ((code>>4) & 0x3);//0x3=11二进制，再生成2bit，要生成下一字节，所有dst++
result[dst] = (byte) (code<<4); //产生高4bit
break;
case 2:
result[dst++] |= (byte) ((code>>2) & 0xf);//
result[dst] = (byte) (code<<6);
break;
case 3:
result[dst++] |= (byte) (code & 0x3f); //0x3f =63 =111111二进制
break;
}
}
} catch (ArrayIndexOutOfBoundsException e) {}
return result;
}
/**
* 加盐异或操作--是不是可以理解为很多加密算法的加盐都是这么操作？？？
* @param encodeStr
* @return
*/
public static String xor_do_salt_cft(String encodeStr) {
char[] encodeChars = encodeStr.toCharArray();
String base64Str_undoSalt = "";
char base64char;
int end = 0; // end state
if (encodeStr.endsWith(equalChar+"")) { //原来"="
end++;
}
if (encodeStr.endsWith(equalChar2)) {//原来"=="
end++;
}
for (int i=0;i<encodeChars.length-end;i++){
int iAscii = encodeChars[i];
//System.out.println(iAscii);//打印asci码
//System.out.println((char) iAscii);//字符
switch (i%4){
case 0:
base64char = (char) (iAscii ^0x07);
break;
case 2:
base64char = (char) (iAscii ^0xF);
break;
default:
base64char =encodeChars[i];
}
base64Str_undoSalt = base64Str_undoSalt+base64char;
}
for (int i=0;i<end;i++){
base64Str_undoSalt = base64Str_undoSalt+equalChar;
}
return base64Str_undoSalt;
}
/**
* Test the decoder and encoder.
* Call as <code>Base64 [string]</code>.
*/
public static void main__ok_for_salt(String[] args) { //main__ok_for_salt
String input = "I love you, huhx 雪梅!I love you, huhx 雪梅!2I love you, huhx 雪梅!3I love you, huhx 雪梅!4I love you, huhx 雪梅!5I love you, huhx 雪梅!6I love you, huhx 雪梅!7";
//String input = "123456mm9";
System.out.println("编码前="+input);
String encodeStr = encode(input);
System.out.println("base64编码后encode= " + encodeStr);
System.out.println("base64编码后的长度="+encodeStr.length());
System.out.println("=======================");
byte[] byteDecode = decode(encodeStr);
String decodeStr = new String(byteDecode);
System.out.println("再解码回来decode:===" + decodeStr );
System.out.println("=======================");
System.out.println("再解码回来decode2:===" + new String(decode("}}:sb3^l+)lvd}wga)>oe#!g6^uq5q*&+<kgb(92^}:5b3<s+(h1a)gg+_mbquaih}%y}}:sb3^l+)lvd}wga)>oe#!g6^uq" )));
System.out.println("=======================");
String ctfPassword = " {9*8ga*l!Tn?@#fj'j$\\g;;";//ctfPassword.toCharArray()
String base64_undo_salt = xor_do_salt_cft(ctfPassword);
System.out.println("undo 盐操作结果："+base64_undo_salt);
System.out.println("=======================");
//System.out.println("todo 盐操作结果："+xor_do_salt_cft(base64_undo_salt));
byte[] byteDecode_CtfPasswprd =decode(base64_undo_salt);
System.out.println("ctf的变形金刚Base64解密: " + base64_undo_salt + " -> ("
+ new String(byteDecode_CtfPasswprd) + ")+长度="+byteDecode_CtfPasswprd.length);
}
public static void main(String[] args){
//String base64Str = "'{6*?gn*k![n8@,fm'e$[g4;"; //"'{6*-?gn*-k![n-8@,fm'e$[g4;"
String base64Str = "'{6*?gn*k![n8@,fm'e$[g;;"; //"'{6*===?gn*===k![n===8@,f===m'e$[g;;"
int end = 0; // end state
if (base64Str.endsWith(equalChar+"")) { //原来"="
end++;
}
if (base64Str.endsWith(equalChar2)) {//原来"=="
end++;
}
int len = (base64Str.length() + 3)/4 * 3 - end;
//byte[] result = new byte[len];
int[] result = new int[len];//这里是为了打印看看char的数组-主要是部分不可见字符
String resultStr = "";
int dst = 0;
try {//如果不做try ..catch (ArrayIndexOutOfBoundsException e) {} 这需要len+3.//最后3个没有作用
for(int i=0;i<base64Str.length();i=i+4){
int asci00,ascc01,ascc02,ascc03; //对应00-03的ascii码
char[] b3 = base64Str.substring(i,i+4).toCharArray();
System.out.println(b3);
//第一个字节
asci00 =charSet.indexOf(b3[0]);
ascc01 = charSet.indexOf(b3[1]);
int byte00 = asci00<<2|ascc01>>4;
result[dst++] = byte00;
System.out.println("asci00<<2|ascc01>>4=" + byte00);
//==========第2个字节
ascc01 = charSet.indexOf(b3[1]);
ascc02 = charSet.indexOf(b3[2]);
int byte01 = (ascc01&0xf)<<4 | ascc02>>2 ;
result[dst++] = byte01;
System.out.println("(ascc01&0xf)<<4 | ascc02>>2="+byte01);
//==========第2个字节
ascc02 = charSet.indexOf(b3[2]);
ascc03=charSet.indexOf(b3[3]);
int byte02 = (ascc02&0x03)<<6 | ascc03;
result[dst++] = byte02;
System.out.println("(ascc02&0x03)<<6 | ascc03="+byte02);
System.out.println("===============");
resultStr = resultStr+(char)byte00+(char)byte01+(char)byte02;
}
}catch (ArrayIndexOutOfBoundsException e) {}
System.out.println("解码后： result="+new Gson().toJson(result));//解码后好像最后2为是多余的：result=[253,30,138,78,9,202,144,3,231,241,133,159,155,247,131,62,14]
System.out.println("解码周：resultStr="+resultStr); //"解码周：resultStr=ý N Êçñ
÷"
String decodeRC4Str =RC4Util_ctf01ok.HloveyRC4(resultStr,"36f36b3c-a03e-4996-8759-8408e626c215");
System.out.println("decodeRC4="+decodeRC4Str);
System.out.println("decodeRC4="+new Gson().toJson(decodeRC4Str.toCharArray()));
}
}
查阅资料
syang大神的看雪CTF-变形金刚： http://blog.syang.xyz/2019/04/kanxue-transformer/
HHHso大神[原创] KCTF 2019 Q1 第二题 有的放矢： https://bbs.pediy.com/thread-250348-1.htm
经典对称加密RC4分析 ：http://ronpa.top/2018/11/27/%E7%BB%8F%E5%85%B8%E5%AF%B9%E7%A7%B0%E5%8A%A0%E5%AF%86RC4%E5%88%86%E6%9E%90/
用 Java 实现的 Rc4 加密算法（C++）：https://www.cnblogs.com/java-stack/archive/2010/10/25/11952676.html
详解Java中的Base64原理跟用法：https://blog.csdn.net/qq_36522306/article/details/80753099
在线工具
ASCII码对照表：https://tool.oschina.net/commons?type=4
在线进制转换器：https://tool.oschina.net/hexconvert/
感受
CTF变形金刚题目设计非常巧妙，通过本例的跟踪分析学到不到逆向与反逆向技术；提升自己对加密算法的理解。