Question

以下是在C上使用cbc和pkcs7填充（和密码）加密的aes256的代码（Windows和C ++（使用libcrypto ++的Ubuntu）。加密结果不一样。为什么？

C＃：

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.IO;
using System.Security.Cryptography;


public static class AESEncryption
{
    public static string Encrypt(byte[] PlainTextBytes, byte[] KeyBytes, string InitialVector)
        {
            try
            {
                byte[] InitialVectorBytes = Encoding.UTF8.GetBytes(InitialVector);
                RijndaelManaged SymmetricKey = new RijndaelManaged();
                SymmetricKey.Mode = CipherMode.CBC;
               // SymmetricKey.Padding = PaddingMode.PKCS7;
                ICryptoTransform Encryptor = SymmetricKey.CreateEncryptor(KeyBytes, InitialVectorBytes);
                MemoryStream MemStream = new MemoryStream();
                CryptoStream CryptoStream = new CryptoStream(MemStream, Encryptor, CryptoStreamMode.Write);
                CryptoStream.Write(PlainTextBytes, 0, PlainTextBytes.Length);
                CryptoStream.FlushFinalBlock();
                byte[] CipherTextBytes = MemStream.ToArray();
                MemStream.Close();
                CryptoStream.Close();
                //return ByteToHexConversion(CipherTextBytes);

                return Convert.ToBase64String(CipherTextBytes);
            }
            catch (Exception a)
            {
                throw a;
            }
        }
    }
namespace aes
{ class Program
    {

        static void Main(string[] args)
        {

            string FinalValue = AESEncryption.Encrypt( Encoding.ASCII.GetBytes("My Text"),  Encoding.ASCII.GetBytes("My Password"), "0000000000000000");

            Console.WriteLine(FinalValue);

        }
}

}

C ++：

#include <iostream>
#include <iomanip>
#include <string>
#include <vector>
#include <cassert>
#include <stdlib.h>
#include <openssl/evp.h>
#include <sstream>
#include "base64.h"

int main()
{


std::string result;
std::stringstream out;

    // ctx holds the state of the encryption algorithm so that it doesn't
    // reset back to its initial state while encrypting more than 1 block.
    EVP_CIPHER_CTX ctx;
    EVP_CIPHER_CTX_init(&ctx);


    std::string keyy="My Password";// in char key[] My Password is written in bytes
    unsigned char key[] = {0x00, 0x00, 0x00, 0x00, 0x00,
                   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
                   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,0x4d,0x79, 0x20, 0x50, 0x61, 0x73, 0x73, 0x77,
                   0x6f, 0x72, 0x64};
    unsigned char iv[] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
   assert(sizeof(key) == 32);  // AES256 key size
    assert(sizeof(iv) ==  16);   // IV is always the AES block size

    // If data isn't a multiple of 16, the default behavior is to pad with
    // n bytes of value n, where n is the number of padding bytes required
    // to make data a multiple of the block size.  This is PKCS7 padding.
    // The output then will be a multiple of the block size.
    std::string plain("My Text");
    std::vector<unsigned char> encrypted;
    size_t max_output_len  = plain.length() + (plain.length() % 16) + 16;
    encrypted.resize(max_output_len);

    // Enc is 1 to encrypt, 0 to decrypt, or -1 (see documentation).
    EVP_CipherInit_ex(&ctx, EVP_aes_256_cbc(), NULL, key, iv, 1);

    // EVP_CipherUpdate can encrypt all your data at once, or you can do
    // small chunks at a time.
    int actual_size = 0;
    EVP_CipherUpdate(&ctx,
             &encrypted[0], &actual_size,
             reinterpret_cast<unsigned char *>(&plain[0]), plain.size());

    // EVP_CipherFinal_ex is what applies the padding.  If your data is
    // a multiple of the block size, you'll get an extra AES block filled
    // with nothing but padding.
    int final_size;
    EVP_CipherFinal_ex(&ctx, &encrypted[actual_size], &final_size);
    actual_size += final_size;

    encrypted.resize(actual_size);

    for( size_t index = 0; index < encrypted.size(); ++index )
    {
        std::cout << std::hex << std::setw(2) << std::setfill('0') <<
            static_cast<unsigned int>(encrypted[index]);
         //std:: cout<< "val: "<< static_cast<unsigned int>(encrypted[index]) << std::endl;

        out<< std::hex << std::setw(2) << std::setfill('0') << static_cast<unsigned int>(encrypted[index]);
    }
    result = out.str();
    std::cout <<"\n"<< result<< "\n";

    EVP_CIPHER_CTX_cleanup(&ctx);


    //
    std::cout<<"decript..\n";


    return 0;
}

Answer 1

你在c＃中的IV是一个包含'0'而不是'\ 0'的字符串，你在c ++中的IV确实包含'\ 0'，ascii值为'0'和'\ 0'是不同的。

替换以下行

string FinalValue = AESEncryption.Encrypt( Encoding.ASCII.GetBytes("My Text"),  Encoding.ASCII.GetBytes("My Password"), "0000000000000000");

与

string FinalValue = AESEncryption.Encrypt( Encoding.ASCII.GetBytes("My Text"),  Encoding.ASCII.GetBytes("My Password"), "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0");

这应该是我认为的伎俩。

20110111

尝试用Encoding.ASCII.GetBytes("My Password")替换new byte[]{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x4d,0x79, 0x20, 0x50, 0x61, 0x73, 0x73, 0x77,0x6f, 0x72, 0x64} 在你的c＃代码中应该大喊不同的结果

Answer 2

扩展dvhh的回答：

无论如何，你可能不应该使用零IV。 IV可以用于避免相同的明文（或具有相同前缀的明文）与攻击者可识别地相同。 IV本身可以是明文;所以你可以随机生成IV并将其添加到输出中以允许解密。

要在.NET中获得安全的随机IV，请执行以下操作：

byte[] initialVectorBytes = new byte[16];
using(var rng = new RNGCryptoServiceProvider())
    rng.GetBytes(initialVectorBytes);
//...
using(var memStream = new MemoryStream()) {
    memStream.Write(IV,0,16); //to permit decryption later. 
    //...
}

顺便说一句，C＃中通常的做法是在变量或参数名称的开头使用小写字母 - 如果你效仿，你将使你的代码对其他人更具可读性。

Answer 3

当我在PHP和.NET Web服务之间进行加密时，我遇到了同样的问题。

我在github上创建了一个示例项目，以显示PHP和NET之间Rijndael加密的工作示例： https://github.com/dchymko/.NET--PHP-encryption

Answer 4

如果要将IV用作值为0x00的16字节数组，则更改C＃代码

byte[] InitialVectorBytes = Encoding.UTF8.GetBytes(InitialVector);

到

byte[] InitialVectorBytes = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};

您也可以更改功能以省略第3个参数

aes256结果在C＃（Windows）和C ++（Ubuntu）实现方面有所不同

4 个答案: