（二）如何确保数据一致性，单向散列函数

博主最近在做数据库加密项目，项目中需要实现数据库中数据密文与密钥异地存储，其功能已实现，但考虑到密钥在传输过程中虽然已经做了加密处理，但是第三方如果恶意替换数据包的话，还是会有安全隐患，应此需要保证二端传输接受到的数据一致性。

生活中，也有不少地方需要保证，接收端和发送端接收到的消息需要保证一致性。例如软件的发行商如何保证自己所发行的软件和用户下载到的软件一致性，而没有被恶意篡改。这里就需要运用到单向散列函数。

单向散列函数

  单向散列函数指的是有一个输入和一个输出的函数，其中输入的数据成为消息，输出称为散列值（hash value）。单向散列函数根据消息内容计算出散列值，用户根据散列值保证数据的一致性。一份合格的单向散列函数须具备以下性质

  1：输入的数据可以是任意长度的，输出的散列值最好为固定长度。

  2：其算法复杂度优异，能够快速计算出散列值，不然博主的项目性能又得不合格。

  3：需要具备强抗碰撞性，即要找到散列值相同的二条不同的消息是非常困难的。

  4：需要具备单向性，即无法通过散列值推算出消息。

  在使用单散列函数时，需要注意的是。消息传输的过程可以存在被篡改的风险，但是散列值的传输必须是安全的。

 

下面我列举几个目前未被攻破的单向散列函数：

 

  05年SHA-1已被攻破，但NIST发布的SHA-224、SHA-256、SHA-384，和SHA-512（统称SHA-2），截至18年还是安全的。

  下面列举SHA-256函数的使用：

SHA-256 算法输入消息的最大长度不超过2⁶⁴比特（基本满足应用）。输出的散列值为256比特。

#include <stdio.h>  
#include <stdlib.h>  
#define SHA256_ROTL(a,b) (((a>>(32-b))&(0x7fffffff>>(31-b)))|(a<<b))  
#define SHA256_SR(a,b) ((a>>b)&(0x7fffffff>>(b-1)))  
#define SHA256_Ch(x,y,z) ((x&y)^((~x)&z))  
#define SHA256_Maj(x,y,z) ((x&y)^(x&z)^(y&z))  
#define SHA256_E0(x) (SHA256_ROTL(x,30)^SHA256_ROTL(x,19)^SHA256_ROTL(x,10))  
#define SHA256_E1(x) (SHA256_ROTL(x,26)^SHA256_ROTL(x,21)^SHA256_ROTL(x,7))  
#define SHA256_O0(x) (SHA256_ROTL(x,25)^SHA256_ROTL(x,14)^SHA256_SR(x,3))  
#define SHA256_O1(x) (SHA256_ROTL(x,15)^SHA256_ROTL(x,13)^SHA256_SR(x,10))  
extern char* StrSHA256(const char* str, long long length, char* sha256){  
    /* 
    计算字符串SHA-256 
    参数说明： 
    str         字符串指针 
    length      字符串长度 
    sha256         用于保存SHA-256的字符串指针 
    返回值为参数sha256 
    */  
    char *pp, *ppend;  
    long l, i, W[64], T1, T2, A, B, C, D, E, F, G, H, H0, H1, H2, H3, H4, H5, H6, H7;  
    H0 = 0x6a09e667, H1 = 0xbb67ae85, H2 = 0x3c6ef372, H3 = 0xa54ff53a;  
    H4 = 0x510e527f, H5 = 0x9b05688c, H6 = 0x1f83d9ab, H7 = 0x5be0cd19;  
    long K[64] = {  
        0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5, 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,  
        0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3, 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,  
        0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc, 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,  
        0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7, 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,  
        0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13, 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,  
        0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3, 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,  
        0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5, 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,  
        0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208, 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2,  
    };  
    l = length + ((length % 64 > 56) ? (128 - length % 64) : (64 - length % 64));  
    if (!(pp = (char*)malloc((unsigned long)l))) return 0;  
    for (i = 0; i < length; pp[i + 3 - 22 * (i % 4)] = str[i], i++);  
    for (pp[i + 3 - 22 * (i % 4)] = 128, i++; i < l; pp[i + 3 - 22 * (i % 4)] = 0, i++);  
    *((long*)(pp + l - 4)) = length << 3;  
    *((long*)(pp + l - 8)) = length >> 29;  
    for (ppend = pp + l; pp < ppend; pp += 64){  
        for (i = 0; i < 16; W[i] = ((long*)pp)[i], i++);  
        for (i = 16; i < 64; W[i] = (SHA256_O1(W[i - 2]) + W[i - 7] + SHA256_O0(W[i - 15]) + W[i - 16]), i++);  
        A = H0, B = H1, C = H2, D = H3, E = H4, F = H5, G = H6, H = H7;  
        for (i = 0; i < 64; i++){  
            T1 = H + SHA256_E1(E) + SHA256_Ch(E, F, G) + K[i] + W[i];  
            T2 = SHA256_E0(A) + SHA256_Maj(A, B, C);  
            H = G, G = F, F = E, E = D + T1, D = C, C = B, B = A, A = T1 + T2;  
        }  
        H0 += A, H1 += B, H2 += C, H3 += D, H4 += E, H5 += F, H6 += G, H7 += H;  
    }  
    free(pp - l);  
    sprintf(sha256, "%08X%08X%08X%08X%08X%08X%08X%08X", H0, H1, H2, H3, H4, H5, H6, H7);  
    return sha256;  
}  
extern char* FileSHA256(const char* file, char* sha256){  
    /* 
    计算文件SHA-256 
    参数说明： 
    file        文件路径字符串指针 
    sha256         用于保存SHA-256的字符串指针 
    返回值为参数sha256 
    */  
    FILE* fh;  
    char* addlp, T[64];  
    long addlsize, j, W[64], T1, T2, A, B, C, D, E, F, G, H, H0, H1, H2, H3, H4, H5, H6, H7;  
    long long length, i, cpys;  
    voidvoid *pp, *ppend;  
    H0 = 0x6a09e667, H1 = 0xbb67ae85, H2 = 0x3c6ef372, H3 = 0xa54ff53a;  
    H4 = 0x510e527f, H5 = 0x9b05688c, H6 = 0x1f83d9ab, H7 = 0x5be0cd19;  
    long K[64] = {  
        0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5, 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,  
        0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3, 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,  
        0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc, 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,  
        0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7, 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,  
        0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13, 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,  
        0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3, 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,  
        0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5, 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,  
        0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208, 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2,  
    };  
    fh = fopen(file, "rb");  
    fseek(fh, 0, SEEK_END);  
    length = _ftelli64(fh);  
    addlsize = (56 - length % 64 > 0) ? (64) : (128);  
    if (!(addlp = (char*)malloc(addlsize))) return 0;  
    cpys = ((length - (56 - length % 64)) > 0) ? (length - length % 64) : (0);  
    j = (long)(length - cpys);  
    if (!(pp = (char*)malloc(j))) return 0;  
    fseek(fh, -j, SEEK_END);  
    fread(pp, 1, j, fh);  
    for (i = 0; i < j; addlp[i + 3 - 22 * (i % 4)] = ((char*)pp)[i], i++);  
    free(pp);  
    for (addlp[i + 3 - 22 * (i % 4)] = 128, i++; i < addlsize; addlp[i + 3 - 22 * (i % 4)] = 0, i++);  
    *((long*)(addlp + addlsize - 4)) = length << 3;  
    *((long*)(addlp + addlsize - 8)) = length >> 29;  
    for (rewind(fh); 64 == fread(W, 1, 64, fh);){  
        for (i = 0; i < 64; T[i + 3 - 22 * (i % 4)] = ((char*)W)[i], i++);  
        for (i = 0; i < 16; W[i] = ((long*)T)[i], i++);  
        for (i = 16; i < 64; W[i] = (SHA256_O1(W[i - 2]) + W[i - 7] + SHA256_O0(W[i - 15]) + W[i - 16]), i++);  
        A = H0, B = H1, C = H2, D = H3, E = H4, F = H5, G = H6, H = H7;  
        for (i = 0; i < 64; i++){  
            T1 = H + SHA256_E1(E) + SHA256_Ch(E, F, G) + K[i] + W[i];  
            T2 = SHA256_E0(A) + SHA256_Maj(A, B, C);  
            H = G, G = F, F = E, E = D + T1, D = C, C = B, B = A, A = T1 + T2;  
        }  
        H0 += A, H1 += B, H2 += C, H3 += D, H4 += E, H5 += F, H6 += G, H7 += H;  
    }  
    for (pp = addlp, ppend = addlp + addlsize; pp < ppend; (long*)pp += 16){  
        for (i = 0; i < 16; W[i] = ((long*)pp)[i], i++);  
        for (i = 16; i < 64; W[i] = (SHA256_O1(W[i - 2]) + W[i - 7] + SHA256_O0(W[i - 15]) + W[i - 16]), i++);  
        A = H0, B = H1, C = H2, D = H3, E = H4, F = H5, G = H6, H = H7;  
        for (i = 0; i < 64; i++){  
            T1 = H + SHA256_E1(E) + SHA256_Ch(E, F, G) + K[i] + W[i];  
            T2 = SHA256_E0(A) + SHA256_Maj(A, B, C);  
            H = G, G = F, F = E, E = D + T1, D = C, C = B, B = A, A = T1 + T2;  
        }  
        H0 += A, H1 += B, H2 += C, H3 += D, H4 += E, H5 += F, H6 += G, H7 += H;  
    }  
    free(addlp); fclose(fh);  
    sprintf(sha256, "%08X%08X%08X%08X%08X%08X%08X%08X", H0, H1, H2, H3, H4, H5, H6, H7);  
    return sha256;  
}  

回到开头的问题，博主的项目的安全问题便有了解决方案。假设Alice端需要给Bob端发送一组密钥，Alice将密钥的散列值算出然后附在发送包里，通过Alice和Bob之间的共享密钥进行XOR加密后发送。Bob收到数据后通过共享密钥解密，然后比较散列值与密钥是否一致。由于第三方不可能获取到共享密钥，改动发送包里的数据，Bob端就能立马知晓。

而例子中，软件开发商会事先将散列值公布在官网上，用户通过不同渠道下载到软件后，跟官网上的散列值进行比较，便可以知道软件是否被篡改。