.NET WinRT和iOS Obj-C之間的AES加密不同

Question

我很難將我的iOS AES加密/解密與WinRT上的相同。

我無法在WinRT方面更改實現，因為它已在已發布的應用程序中使用。

這是我用零鍵和零iv制作的兩個樣本，輸出是不同的。

C＃示例代碼：

using (MemoryStream saveDataMemoryStreamCrypto = new MemoryStream())
{
    var saveDataKeyProvider = Windows.Security.Cryptography.Core.SymmetricKeyAlgorithmProvider.OpenAlgorithm(Windows.Security.Cryptography.Core.SymmetricAlgorithmNames.AesCbcPkcs7);
    var saveDataKeyBuffer = Windows.Security.Cryptography.CryptographicBuffer.CreateFromByteArray(new byte[32]);
    var saveDataKey = saveDataKeyProvider.CreateSymmetricKey(saveDataKeyBuffer);
    var saveDataSaltBuffer = Windows.Security.Cryptography.CryptographicBuffer.CreateFromByteArray(new byte[32]);
    var saveDataDataBuffer =  Windows.Security.Cryptography.CryptographicBuffer.ConvertStringToBinary("ABCDEFGH", Windows.Security.Cryptography.BinaryStringEncoding.Utf16BE);
    var saveDataOutBuffer = Windows.Security.Cryptography.Core.CryptographicEngine.Encrypt(saveDataKey, saveDataDataBuffer, saveDataSaltBuffer);
    var saveDataOutBytes = saveDataOutBuffer.ToArray();
}

C＃字節輸出：

80 87 109 195 133 40 205 81 117 91 17 132 229 3 119 251 205 8 246 64 13 57 210 142 11 153 121 39 122 196 63 10

Obj-C示例代碼：

Byte keyPtr[32];
bzero(keyPtr, sizeof(keyPtr));
Byte ivPtr[32];
bzero(ivPtr, sizeof(ivPtr));

NSString *text = @"ABCDEFGH";
NSUInteger dataLength;
void * buffer = malloc([text length]);
[text getBytes:buffer maxLength:[text length] usedLength:&dataLength encoding:NSUTF16BigEndianStringEncoding options:0 range:NSMakeRange(0, dataLength) remainingRange:nil];

size_t bufferSize = dataLength * kCCBlockSizeAES128;
void * bufferOut = malloc(bufferSize);

size_t numBytesEncrypted = 0;

CCCryptorStatus cryptStatus = CCCrypt(kCCEncrypt, kCCAlgorithmAES128, kCCOptionPKCS7Padding, keyPtr, kCCKeySizeAES256, ivPtr, buffer, dataLength, bufferOut, bufferSize, &numBytesEncrypted);

Obj-C字節輸出：

23 144 186 234 149 182 123 79 155 234 250 54 52 38 151 87 179 62 176 1 203 115 59 1 35 54 176 1 44 213 120 1

有人知道我在做什么錯嗎？

謝謝，格雷格

Answer 1

這是工作代碼，主要區別在於UTF16的數據長度為每個字符兩個字節：

u_int8_t  keyPtr[32];
bzero(keyPtr, sizeof(keyPtr));
u_int8_t  ivPtr[32];
bzero(ivPtr, sizeof(ivPtr));

NSString *text = @"ABCDEFGH";
NSUInteger dataLength = [text length] * 2; // Allow for utf16
void * buffer = malloc(dataLength);
[text getBytes:buffer maxLength:dataLength usedLength:nil encoding:NSUTF16BigEndianStringEncoding options:0 range:NSMakeRange(0, dataLength) remainingRange:nil];

size_t bufferSize = dataLength * kCCBlockSizeAES128;
u_int8_t * bufferOut = malloc(bufferSize);

size_t numBytesEncrypted = 0;

CCCryptorStatus cryptStatus = CCCrypt(kCCEncrypt, kCCAlgorithmAES128, kCCOptionPKCS7Padding, keyPtr, kCCKeySizeAES256, ivPtr, buffer, dataLength, bufferOut, bufferSize, &numBytesEncrypted);

printf("encoded text in decimal: ");
for (int i=0; i<numBytesEncrypted; i++) {
    printf("%d ", bufferOut[i]);
}
printf("\n");

Printf輸出：

十進制編碼的文本：80 87 109 195 133 40 205 81 117 91 17 132 229 3 119 251 205 8 246 64 13 57 210 142 11 153 121 39 122 196 63 10

是的，這真是糟糕的代碼，我做了必要的最小改動。 我猜正在進入一個新時代，數據轉儲以十進制表示，十六進制已死。