原文:字符串混淆技术应用 设计一个字符串混淆程序 可混淆.NET程序集中的字符串
关于字符串的研究,目前已经有两篇。
原理篇:字符串混淆技术在.NET程序保护中的应用及如何解密被混淆的字符串
实践篇:字符串反混淆实战 Dotfuscator 4.9 字符串加密技术应对策略
今天来讲第三篇,如何应用上面所学内容,设计一个字符串混淆程序。
先设计一个控制台程序,它是将要被我混淆的程序集文件:
public static void Main() { try { RunSnippet(); } catch (Exception e) { string error = string.Format("---\nThe following error occurred while executing the snippet:\n{0}\n---", e.ToString()); Console.WriteLine(error); } finally { Console.Write("Press any key to continue..."); Console.ReadKey(); } }
代码是Snippet Compiler 的标准模板,在控制台上打印一段字符串。这里,有二个字符串常量,是我需要对它进行加密的地方。
不能直接改源代码,而且要以程序的方式来操作程序集。要能修改.net程序集,现在能找到的方法是Mono.Cecil,最新的版本是0.9.5.0。
先设计混淆算法,是个很简单的Base64代码转换,这一步可以深入挖掘,做更复杂的混淆算法。
public static string StringDecode(string text1) { return Encoding.UTF8.GetString(Convert.FromBase64String(text1)); }
.csharpcode, .csharpcode pre
{
font-size: small;
color: black;
font-family: consolas, "Courier New", courier, monospace;
background-color: #ffffff;
/*white-space: pre;*/
}
.csharpcode pre { margin: 0em; }
.csharpcode .rem { color: #008000; }
.csharpcode .kwrd { color: #0000ff; }
.csharpcode .str { color: #006080; }
.csharpcode .op { color: #0000c0; }
.csharpcode .preproc { color: #cc6633; }
.csharpcode .asp { background-color: #ffff00; }
.csharpcode .html { color: #800000; }
.csharpcode .attr { color: #ff0000; }
.csharpcode .alt
{
background-color: #f4f4f4;
width: 100%;
margin: 0em;
}
.csharpcode .lnum { color: #606060; }
.csharpcode, .csharpcode pre
{
font-size: small;
color: black;
font-family: consolas, "Courier New", courier, monospace;
background-color: #ffffff;
/*white-space: pre;*/
}
.csharpcode pre { margin: 0em; }
.csharpcode .rem { color: #008000; }
.csharpcode .kwrd { color: #0000ff; }
.csharpcode .str { color: #006080; }
.csharpcode .op { color: #0000c0; }
.csharpcode .preproc { color: #cc6633; }
.csharpcode .asp { background-color: #ffff00; }
.csharpcode .html { color: #800000; }
.csharpcode .attr { color: #ff0000; }
.csharpcode .alt
{
background-color: #f4f4f4;
width: 100%;
margin: 0em;
}
.csharpcode .lnum { color: #606060; }
要把这段代码转化为IL代码,用Mono.Cecil来注入到制定的程序集中,先来看看翻译成IL之后的代码
.method public hidebysig static string StringDecode(string) cil managed { .maxstack 8 L_0000: call class [mscorlib]System.Text.Encoding [mscorlib]System.Text.Encoding::get_UTF8() L_0005: ldarg.0 L_0006: call uint8[] [mscorlib]System.Convert::FromBase64String(string) L_000b: callvirt instance string [mscorlib]System.Text.Encoding::GetString(uint8[]) L_0010: ret }
再对比Mono.Cecil的语法例子,把上面的IL代码,翻译成C#代码
MethodDefinition new_method = new MethodDefinition("StringDecode", attr, asm.MainModule.Import(typeof(string))); ParameterDefinition para = new ParameterDefinition(asm.MainModule.Import(typeof(string))); new_method.Parameters.Add(para); tp.Methods.Add(new_method); new_method.Body.MaxStackSize = 8; MethodReference mr; ILProcessor worker = new_method.Body.GetILProcessor (); mr = asm.MainModule.Import(typeof(Encoding).GetMethod("get_UTF8")); worker.Append(worker.Create(OpCodes.Call, mr)); worker.Append(worker.Create(OpCodes.Ldarg_0)); mr = asm.MainModule.Import(typeof(Convert).GetMethod("FromBase64String")); worker.Append(worker.Create(OpCodes.Call, mr)); mr = asm.MainModule.Import(typeof(Encoding).GetMethod("GetString", new Type[] { typeof(Byte[]) })); worker.Append(worker.Create(OpCodes.Callvirt, mr)); worker.Append(worker.Create(OpCodes.Ret));
再次,我样要搜索目标程序集中的所有字符串,把它转化成Base64的字符串编码,于是遍历IL指令,进行转化
List<Instruction> actionInsert=new List<Instruction> (); foreach (Instruction ins in entry_point.Body.Instructions) { if (ins.OpCode.Name == "ldstr") { Console.WriteLine("Find target instruction, start modify.."); byte[] bytes = System.Text.Encoding.UTF8.GetBytes (Convert.ToString (ins.Operand)); ins.Operand = Convert.ToBase64String (bytes); actionInsert.Add(ins); }
}
最后,我们把原来的指令替换成字符串混淆算法调用
for (int i = 0; i < actionInsert.Count; i++) { mr = asm.MainModule.Import(new_method); worker = entry_point.Body.GetILProcessor(); worker.InsertAfter(actionInsert[i], worker.Create(OpCodes.Call, mr)); }
最后保存程序集,用.net Reflector 载入程序集,如下图所示,字符串常量已经变成了方法调用:
.csharpcode, .csharpcode pre
{
font-size: small;
color: black;
font-family: consolas, "Courier New", courier, monospace;
background-color: #ffffff;
/*white-space: pre;*/
}
.csharpcode pre { margin: 0em; }
.csharpcode .rem { color: #008000; }
.csharpcode .kwrd { color: #0000ff; }
.csharpcode .str { color: #006080; }
.csharpcode .op { color: #0000c0; }
.csharpcode .preproc { color: #cc6633; }
.csharpcode .asp { background-color: #ffff00; }
.csharpcode .html { color: #800000; }
.csharpcode .attr { color: #ff0000; }
.csharpcode .alt
{
background-color: #f4f4f4;
width: 100%;
margin: 0em;
}
.csharpcode .lnum { color: #606060; }
这样,增加了代码反编译的难度,字符串的含义完全被替换成一堆无意义的字符串。
Mono.Cecil最新的版本中,API有变化,本篇程序代码中应用到的读取程序集和写入程序集
string path = @"C:\Users\Administrator\Desktop\CPP\Default.exe"; AssemblyDefinition asm = AssemblyDefinition.ReadAssembly(path); MethodDefinition entry_point = asm.EntryPoint; path = @"C:\Users\Administrator\Desktop\CPP\DefaultSecury.exe"; asm.MainModule.Write(path);
.csharpcode, .csharpcode pre
{
font-size: small;
color: black;
font-family: consolas, "Courier New", courier, monospace;
background-color: #ffffff;
/*white-space: pre;*/
}
.csharpcode pre { margin: 0em; }
.csharpcode .rem { color: #008000; }
.csharpcode .kwrd { color: #0000ff; }
.csharpcode .str { color: #006080; }
.csharpcode .op { color: #0000c0; }
.csharpcode .preproc { color: #cc6633; }
.csharpcode .asp { background-color: #ffff00; }
.csharpcode .html { color: #800000; }
.csharpcode .attr { color: #ff0000; }
.csharpcode .alt
{
background-color: #f4f4f4;
width: 100%;
margin: 0em;
}
.csharpcode .lnum { color: #606060; }
关于字符串混淆算法,下面列举几个我找到的混淆算法,加密强度会高一些:
static string stringEncrypt(string string_0) { char[] chArray; char[] chArray1 = chArray = string_0.ToCharArray(); while (true) { int num; int length = chArray1.Length; if (length <= 0) { break; } chArray1[num = length + -1] = (char) (chArray[num] - ‘?‘); } return string.Intern(new string(chArray)); }
下面是Dotfuscator的混淆算法,还加了盐,强度提升不少。
static string GetString(string source, int salt) { int index = 0; char[] data = source.ToCharArray(); salt += 0xe74d6d7; // This const data generated by dotfuscator while (index < data.Length) { char key = data[index]; byte low = (byte)((key & ‘\x00ff‘) ^ salt++); byte high = (byte)((key >> 8) ^ salt++); data[index] = (char)((low << 8 | high)); index++; } return string.Intern(new string(data)); }
经过混淆后的字符串,完全看不出原文的含义。比如下面的代码片段,都有些怀疑它使用的字符集,有点像中东国家的语言
.csharpcode, .csharpcode pre
{
font-size: small;
color: black;
font-family: consolas, "Courier New", courier, monospace;
background-color: #ffffff;
/*white-space: pre;*/
}
.csharpcode pre { margin: 0em; }
.csharpcode .rem { color: #008000; }
.csharpcode .kwrd { color: #0000ff; }
.csharpcode .str { color: #006080; }
.csharpcode .op { color: #0000c0; }
.csharpcode .preproc { color: #cc6633; }
.csharpcode .asp { background-color: #ffff00; }
.csharpcode .html { color: #800000; }
.csharpcode .attr { color: #ff0000; }
.csharpcode .alt
{
background-color: #f4f4f4;
width: 100%;
margin: 0em;
}
.csharpcode .lnum { color: #606060; }
再来看一个代码中有中文的例子,这是一段用户登陆代码,它加密后的字符看起来更不可理解。
如果要给字符串混淆加盐,只需要简单的修改上面的代码,添加一个临时变量,再增加到调用的混淆算法中。
全文代码以NUnit测试方法写成,单元测试配合Resharper真是好用,可以节省大量的代码,一个方法即可作为入口程序启动运行。
给开发和测试带多很多方便。