I'm working with the Kraken API and to query the private methods I need to send a hash as a parameter. This is what their documentation says:
Public methods can use either GET or POST.
Private methods must use POST and be set up as follows:
HTTP header:
API-Key = API key
API-Sign = Message signature using HMAC-SHA512 of
(URI path + SHA256(nonce + POST data)) and base64 decoded secret API
key
POST data:
nonce = always increasing unsigned 64 bit integer
otp = two-factor
password (if two-factor enabled, otherwise not required)
Note: There
is no way to reset the nonce to a lower value so be sure to use a
nonce generation method that won't generate numbers less than the
previous nonce. A persistent counter or the current time in hundredths
of a second precision or higher is suggested.
They also have a PHP/Node.JS/Python example that creates the API-Sign hash. I am trying to port this code to C# for Windows Phone 8, but I encountered a big problem: the HMACSHA512 class isn't available for Windows Phone. I tried searching for alternatives that can create a HMAC-SHA512 hash, but couldn't find much. HashLib isn't available for Windows Phone. CryptSharp is, but I can't figure out how to add both a message and a password like PHP's hash_hmac() function allows. I also went searching for the algorithm/pseudo code for the HMAC-SHA512 algorithm to implement my own class but strangely enough I couldn't find it (does it have another name?).
Long story short, I need to convert this code in a Windows Phone 8 compatible piece of code that yields the same result:
if(!isset($request['nonce'])) {
// generate a 64 bit nonce using a timestamp at microsecond resolution
// string functions are used to avoid problems on 32 bit systems
$nonce = explode(' ', microtime());
$request['nonce'] = $nonce[1] . str_pad(substr($nonce[0], 2, 6), 6, '0');
}
// build the POST data string
$postdata = http_build_query($request, '', '&');
// set API key and sign the message
$path = '/' . $this->version . '/private/' . $method;
$sign = hash_hmac('sha512', $path . hash('sha256', $request['nonce'] . $postdata, true), base64_decode($this->secret), true);
$headers = array(
'API-Key: ' . $this->key,
'API-Sign: ' . base64_encode($sign)
);
The first part (till $sign = ...) looks pretty straightforward:
long nonce = DateTime.UtcNow.Ticks;
string postData = "nonce=" + nonce;
if (!string.IsNullOrEmpty(otp))
{
postData += "&otp=" + otp;
}
But when I get to the cryptography part, I get stuck due to the lack of libraries.
This isn't likely to even compile, but it should give you the general idea..
byte[] uriPath = GetBytes(uriPathString);
byte[] nonceAndPostData = GetBytes(nonce + postData);
byte[] keyData = Convert.FromBase64String(apiKey);
string decodedKey = Encoding.UTF8.GetString(keyData);
byte[] result;
SHA512 shaM = new SHA512Managed();
SHA256 shaN = new SHA256Managed();
result = shaN.ComputeHash(nonceAndPostData);
result = shaM.ComputeHash(result + decodedKey);
static byte[] GetBytes(string str)
{
byte[] bytes = new byte[str.Length * sizeof(char)];
System.Buffer.BlockCopy(str.ToCharArray(), 0, bytes, 0, bytes.Length);
return bytes;
}
Related
I have got an Angular + Net Core application with an (RSA + AES) encrypted connection.
All requests from client are coming via POST. (You will be given an example below.
The script provided below works quite well but throws in 5% cases an exception:
The length of the data to decrypt is not valid for the size of this key
in the line:
var decryptedAesKey = Encoding.UTF8.GetString(rsaCng.Decrypt(Convert.FromBase64String(request.k), RSAEncryptionPadding.Pkcs1));
Encryption part (Front-end)
encrypt(requestObj:any):any {
var rsaEncrypt = new JsEncryptModule.JSEncrypt();
var key = this.generateAesKey(32); //secret key
var iv = this.generateAesKey(16); //16 digit
var stringifiedRequest = CryptoJS.enc.Utf8.parse(JSON.stringify(requestObj));
var aesEncryptedRequest = CryptoJS.AES.encrypt(stringifiedRequest,
CryptoJS.enc.Utf8.parse(key),
{
keySize: 128 / 8,
iv: CryptoJS.enc.Utf8.parse(iv),
padding: CryptoJS.pad.Pkcs7,
mode: CryptoJS.mode.CBC
});
rsaEncrypt.setPrivateKey(this.publicPemKey);
var encryptedKey = rsaEncrypt.encrypt(key);
var encryptedIV = rsaEncrypt.encrypt(iv);
var encryptedRequestObj = {
k: encryptedKey,
v: encryptedIV,
r: aesEncryptedRequest.toString()
};
return encryptedRequestObj;
}
Decryption part (C# Back-end)
var decryptedAesKey = Encoding.UTF8.GetString(rsaCng.Decrypt(Convert.FromBase64String(request.k),
RSAEncryptionPadding.Pkcs1));
var decryptedAesIV = Encoding.UTF8.GetString(rsaCng.Decrypt(Convert.FromBase64String(request.v), RSAEncryptionPadding.Pkcs1));
byte[] encryptedBytes = request.r;
AesCryptoServiceProvider aes = new AesCryptoServiceProvider()
{
Mode = CipherMode.CBC,
Padding = PaddingMode.PKCS7,
Key = Encoding.UTF8.GetBytes(decryptedAesKey),
IV = Encoding.UTF8.GetBytes(decryptedAesIV)
};
ICryptoTransform crypto = aes.CreateDecryptor(aes.Key, aes.IV);
byte[] secret = crypto.TransformFinalBlock(encryptedBytes, 0, encryptedBytes.Length);
crypto.Dispose();
requestJson = Encoding.UTF8.GetString(secret);
Example, a user wants to open a page by id.
Front-end:
1) encrypts request Id using AES
2) encrypts AES' key & iv using RSA
3) sends to Back-end
Back-end:
1) decrypts AES' key & value using RSA <--- BREAKS HERE
2) decrypts request Id using AES' key & iv
3) decrypts and get id as if there was no encryption
This logic works quite well but breaks sometimes...
EXAMPLE OF FAILING request:
{ "k":"L+ikMb/JGvFJmhBpADMGTVLFlkHOe69dZUVSQ5r7yHCvWSwY2x6KMR274ByflF0lDMYdCmywo+Nfq6JUybRctDqmAp8UFHXnhwBAv49d99mF5x2yGbJr/j0cn6EZyhweNK4p97i5yMM6MQtluZTIErpsUa22Cajtj8F+xl0jJPUMXIf8cs2X+ooFr5VP/p/vlbPmnEY3K/hMCRZRdXMkEqaCWoA5EnYMTQABtRXPZWgLSQwJpr4dqEAhGCBtga1AGsKF3dQCsKO92NYyst0ngkBiKwFNfy1QDwbk4SzKAKeBckaY17SHt526NMvpEv08BGV6btBxcM+ypsmpB4o0",
"v":"LIndJOjUgKHDlXqwpg7uSmDuut3oi5z9L/GKm2KgU7P2EXmf/JIpXM0JgpTXPJL7wUTndq3F9UMlMdU70JBOV56x/4uIBRbHbyvaG2JZYxbBZblwyYgdo1ZcK1OSE4k5oesQmMEGNEk9RVu+EZO4xAme6+mlyd2/Y/709jaC90PuiOG/k/4JMTTI/2q4s7tk6IgSxLBT8ZiOtgJVGdasSaAksEBMRHyUkzAIr5tSUw1VXedwJFPfwQT2nOD5dU2cxiNJKOwtO9uAYXly0U0FDoa/nkWskca8zaU+4EiPikJ6Km7phViH9JvwZFgHhBj+8FM6Jof+AdrY3q1dcMLFlg==",
"r":"OJnA3wFoKKG+iu4FciXyJg=="
}
EXAMPLE OF CORRECT REQUEST:
{ "k":"uW8d7vIzlgkEkKTkDnHbBZeqKwdgoG+1BVZ/NUiC0pZ/LqZM9aUasQSx+qDg+X50ur30uRnEyAyIZXruYeHQb8cacx5mvr9LWLud+wueJXsOlEEdocD/4A1DfE9TDFdnTaVcMSIwhSVlLPUjO7ubJdANY9yK4S+vb0IyPbsrYpAT7ho01mDkvsH1rZsId/TmzQadmsGhThowu+mrQlz78rrdlN8nI5LnUQHXRNWMUgBvuteTpVBmyrfnIELIKoo/jI6Nj4rGPQBf7+2OOoZPs0Y1GtjXxUCTAt7madNLKSOdaPjdWjaOfGSwnymDNeEFyJQOmAwHZoOGYNd2B/UhQQ==",
"v":"IimiJFcKv5ZHWHljJixX0LUgV4I2GWAWPbk7dWHVhwmHEhTHA/hCdih/E1wiWFS+0KaL05ZobiZInyK7gCwYPHaz0aRCSQtVeBPiFg4f7L0gwfvk1GHwJ1wZjqNJZaYf0elXJzc2l5BwN+aXNWaNJDPA7M6kfK6UPkq84IV3ohCQcTuC8zPM7aMJHxpz9IudcrMmYIkeqrj9Do88CkTLv8yg5hk3EASPk9HqsUieuQixggv/8ZlHnp00iftc62LJlIuCkGn4WR3FkMdFdqpKXf6Ebj8PU1HOmokEtKtYJiOZ5JxieZO5Pnd+ez6sO7khIbdRFDhAQ20chsxKUypezw==",
"r":"2mbUgU44JFFDlWu8As2RIw=="
}
In the case of the failed request, the Base64 decoded encrypted AES key has a length of 255 bytes. For a 2048 bit RSA key it should actually be 256 bytes, as it is for the remaining data.
For the RSA-encryption JSEncrypt is used, which has a known bug that sporadically causes too short ciphertexts and which is probably responsible for your issue, see here. This bug was opened in July 2019 and is not fixed yet.
Within JSEncrypt the too short ciphertexts are processed correctly, so that no error occurs. Cross platform however, this is often not the case, because the too short ciphertexts are strictly speaking invalid and therefore some programming languages identify them as invalid, e.g. Python, apparently C# is another one.
If the too short ciphertext is manually padded from the left to the length of the modulus with 0x00, the ciphertext should also be decryptable in the C# code.
Update:
I have successfully tested the suggested fix using your code. The ciphertext can be fixed in the JavaScript or C# code. A possible implementation for the JavaScript side is e.g. for the key:
encryptedKey = btoa(atob(encryptedKey).padStart(256, "\0"));
where encryptedKey is the Base64 encoded ciphertext as returned by JSEncrypt#encrypt. To ensure that this correction isn't applied to ciphertexts that already have the correct length, a length check is useful: A Base64 encoded ciphertext of length 4 * Math.ceil(256 / 3) doesn't need to be fixed because it corresponds to a ciphertext of the correct length of 256 bytes, see here.
You apply the method setPrivateKey in the JSEncrypt part when setting the public key for the encryption, correct would be setPublicKey, see here. However, JSEncrypt seems to fix this internally, because it works as well. Nevertheless it should be changed, because it's misleading.
As already mentioned in the comments by #kelalaka, the IV is no secret and doesn't need to be encrypted.
I am encrypting the message in .NET with RSACryptoServiceProvider with private key. (PKCS#1 v1.5)
When I try to decrypt in .NET with the following code that uses public key everything works fine:
private static string Decrypt(string key, string content)
{
byte[] rgb = Convert.FromBase64String(content);
var cryptoServiceProvider = new RSACryptoServiceProvider(new CspParameters()
{
ProviderType = 1
});
cryptoServiceProvider.ImportCspBlob(Convert.FromBase64String(key));
return Convert.ToBase64String(cryptoServiceProvider.Decrypt(rgb, false));
}
When on the other hand I try to find an algorithm to make the same decrypt method in Android, I am failing to decrypt it properly with public key. I exported the modulus and exponent from public key in .NET in order to load it properly on Android.
The method in Android is here:
public String Decrypt(String input) {
try {
KeyFactory keyFactory = KeyFactory.getInstance("RSA");
String modulusString = "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";
String exponentString = "AQAB";
byte[] modulusBytes = Base64.decode(modulusString.getBytes("UTF-8"), Base64.DEFAULT);
byte[] dBytes = Base64.decode(exponentString.getBytes("UTF-8"), Base64.DEFAULT);
BigInteger modulus = new BigInteger(1, modulusBytes);
BigInteger d = new BigInteger(1, dBytes);
RSAPublicKeySpec keySpec = new RSAPublicKeySpec(modulus, d);
PublicKey key = keyFactory.generatePublic(keySpec);
//at one point I read somewhere that .net reverses the byte array so that it needs to be reversed for java, but who knows any more
/*byte[] inputArrayReversed = Base64.decode(input.getBytes("UTF-8"), Base64.DEFAULT);
for (int i = 0; i < inputArrayReversed.length / 2; i++) {
byte temp = inputArrayReversed[i];
inputArrayReversed[i] = inputArrayReversed[inputArrayReversed.length - 1];
inputArrayReversed[inputArrayReversed.length - 1] = temp;
}*/
byte[] decryptedText = null;
Cipher cipher = Cipher.getInstance("RSA/ECB/PKCS1Padding");
cipher.init(Cipher.DECRYPT_MODE, key);
decryptedText = cipher.doFinal(Base64.decode(input.getBytes("UTF-8"), Base64.DEFAULT));
return Base64.encodeToString(decryptedText, Base64.NO_WRAP);
//return new String(decryptedText, "UTF-8");
} catch (Exception e) {
e.printStackTrace();
}
return "";
}
Actually I tried also with different algorithms specified in Cypher class, also tried many other combinations, tried using SpongyCastle instead of built in Android RSA providers, but nothing worked. If anybody has any clue to point me in right direction, I would be absolutely grateful.
First hint is that decrypted string from .NET comes as around 25 characters long, and when I get Android to return decrypted string without exceptions it is usually much longer, around 500 bytes.
Second hint deleted
Third hint I also tried spongycastle, but it didn't help that much
Anyways, thank you in advance for any help!!!
UPDATE 1
Second hint is deleted because was wrong, disregard it. Now I have one question if the following can prove that the public key is loaded correctly, just to rule that problem out.
BigInteger modulus and exponent in the upper Android code and the following BigIntegers in .NET show equal integer values.
var parameters = csp.ExportParameters(false);
var modulusInteger = new BigInteger(parameters.Modulus.Reverse().Concat(new byte[] { 0 }).ToArray());
var exponentInteger = new BigInteger(parameters.Exponent.Reverse().Concat(new byte[] { 0 }).ToArray());
UPDATE 2
This and This SO answers provide some interesting clues
Heeh, the mistake was one of the basics, we had an architecture where we were doing encryption with public key and decryption with private key. The problem was in the architecture itself because as we initially set it up, we were sending private keys to all our client apps, which is big security flaw.
My mistake was that I assumed that on the client we have public key and actually from private key all the time I was trying to load the public key and then do decrypt.
If I knew the PKI in depth and communicated a bit better with my colleague, I could have noticed few things:
Decrypt can be done with private key only, while one the other hand verify can be done with public key, so when I saw Decrypt being used on client in .NET, I should have assumed that on the client we have private key (which is a security flaw in the end in the way we want to use PKI)
Few things that I already knew or learnt and want to share with others:
Private key should be kept secret, whether you want to have it on server or preferably only on one client because public key can easily be guessed from private key and then someone can easily repeat your whole encryption process easily and breach your security
PKI works for two scenarios:
First scenario is when you want to Encrypt something and that only specific person/computer can Decrypt it. In first scenario as you see, many stakeholders can have someone's Public key and send messages to him and that only he can read them with his Private key. Second scenario is when you want to be sure that the message that came to you was not altered and was sent by specific person/computer. In that case you Sign data with Private key and Verify it on the other end with Public key. The only process that is suitable for us is Sign <-> Verify because we send plain text license with signature in it, and thus on the client we want to be sure that nobody tampered with the plain text license and that it came from us.
In your code, if Decrypt or Verify functions throw exceptions in 50% of the time it is because of loading the incorrect key or incorrectly loading the correct key and in the other 50% it is because you are using the incorrect algorithm or because algorithm parameters are incorrectly set or because the algorithm implementations between platforms are incompatible (the last one is very rare)
.NET server code
public string Sign(string privateKey, string data)
{
_rsaProvider.ImportCspBlob(Convert.FromBase64String(privateKey));
//// Write the message to a byte array using UTF8 as the encoding.
var encoder = new UTF8Encoding();
byte[] byteData = encoder.GetBytes(data);
//// Sign the data, using SHA512 as the hashing algorithm
byte[] encryptedBytes = _rsaProvider.SignData(byteData, new SHA1CryptoServiceProvider());
return Convert.ToBase64String(encryptedBytes);
}
.NET client code (Win Mobile)
private bool Verify(string key, string signature, string data)
{
CspParameters cspParams = new CspParameters { ProviderType = 1 };
RSACryptoServiceProvider rsaProvider = new RSACryptoServiceProvider(cspParams);
rsaProvider.ImportCspBlob(Convert.FromBase64String(key));
byte[] signatureBytes = Convert.FromBase64String(signature);
var encoder = new UTF8Encoding();
byte[] dataBytes = encoder.GetBytes(data);
return rsaProvider.VerifyData(dataBytes, new SHA1CryptoServiceProvider(), signatureBytes);
}
Android client code:
public boolean Verify(RSAPublicKey key, String signature, String data)
{
try
{
Signature sign = Signature.getInstance("SHA1withRSA");
sign.initVerify(key);
sign.update(data.getBytes("UTF-8"));
return sign.verify(Base64.decode(signature.getBytes("UTF-8"), Base64.NO_WRAP));
}
catch (Exception e)
{
e.printStackTrace();
}
return false;
}
in .NET public key is exported in xml format with following code:
public string ExportPublicToXML(string publicKey)
{
RSACryptoServiceProvider csp = new RSACryptoServiceProvider(new CspParameters()
{
ProviderType = 1
});
csp.ImportCspBlob(Convert.FromBase64String(publicKey));
return csp.ToXmlString(false);
}
and then modulus and exponent are used in Android to load public key:
private RSAPublicKey GetPublicKey(String keyXmlString) throws InvalidKeySpecException, UnsupportedEncodingException, NoSuchAlgorithmException
{
KeyFactory keyFactory = KeyFactory.getInstance("RSA");
String modulusString = keyXmlString.substring(keyXmlString.indexOf("<Modulus>"), keyXmlString.indexOf("</Modulus>")).replace("<Modulus>", "");
String exponentString = keyXmlString.substring(keyXmlString.indexOf("<Exponent>"), keyXmlString.indexOf("</Exponent>")).replace("<Exponent>", "");
byte[] modulusBytes = Base64.decode(modulusString.getBytes("UTF-8"), Base64.DEFAULT);
byte[] dBytes = Base64.decode(exponentString.getBytes("UTF-8"), Base64.DEFAULT);
BigInteger modulus = new BigInteger(1, modulusBytes);
BigInteger d = new BigInteger(1, dBytes);
RSAPublicKeySpec keySpec = new RSAPublicKeySpec(modulus, d);
return (RSAPublicKey) keyFactory.generatePublic(keySpec);
}
I'm having trouble with WS-Security, and creating a nonce and password digest that is correct.
I am successfully using SoapUI to send data to an Oracle system. So I'm able to intercept SoapUI's call (change proxy to 127.0.0.1 port 8888 to use Fiddler where it fails because it's over SSL) - intercepting is important because these values can only be used once. I can then grab the nonce, created timestamp and password digest put them into my code (I've only got 30 seconds to do this as the values don't last!) and I get a success.
So I know it's nothing else - just the Password Digest.
The values I use are the following:
Nonce: UIYifr1SPoNlrmmKGSVOug==
Created Timestamp: 2009-12-03T16:14:49Z
Password: test8
Required Password Digest: yf2yatQzoaNaC8BflCMatVch/B8=
I know the algorithm for creating the Digest is:
Password_Digest = Base64 ( SHA-1 ( nonce + created + password ) )
using the following code (from Rick Strahl's post)
protected string GetSHA1String(string phrase)
{
SHA1CryptoServiceProvider sha1Hasher = new SHA1CryptoServiceProvider();
byte[] hashedDataBytes = sha1Hasher.ComputeHash(Encoding.UTF8.GetBytes(phrase));
return Convert.ToBase64String(hashedDataBytes);
}
I get:
GetSHA1String("UIYifr1SPoNlrmmKGSVOug==" + "2009-12-03T16:14:49Z" + "test8") = "YoQKI3ERlMDGEXHlztIelsgL50M="
I have tried various SHA1 methods, all return the same results (which is a good thing I guess!):
SHA1 sha1 = SHA1.Create();
SHA1 sha1 = SHA1Managed.Create();
// Bouncy Castle:
protected string GetSHA1usingBouncyCastle(string phrase)
{
IDigest digest = new Sha1Digest();
byte[] resBuf = new byte[digest.GetDigestSize()];
byte[] bytes = Encoding.UTF8.GetBytes(phrase);
digest.BlockUpdate(bytes, 0, bytes.Length);
digest.DoFinal(resBuf, 0);
return Convert.ToBase64String(resBuf);
}
Any ideas on how to get the correct hash?
The problem was the nonce.
I was trying to use a nonce that had already been Base64 encoded. If you want to use a Nonce that is in the form "UIYifr1SPoNlrmmKGSVOug==" then you need to decode it.
Convert.FromBase64String("UIYifr1SPoNlrmmKGSVOug==")
which is a byte array.
So we need a new method:
public string CreatePasswordDigest(byte[] nonce, string createdTime, string password)
{
// combine three byte arrays into one
byte[] time = Encoding.UTF8.GetBytes(createdTime);
byte[] pwd = Encoding.UTF8.GetBytes(password);
byte[] operand = new byte[nonce.Length + time.Length + pwd.Length];
Array.Copy(nonce, operand, nonce.Length);
Array.Copy(time, 0, operand, nonce.Length, time.Length);
Array.Copy(pwd, 0, operand, nonce.Length + time.Length, pwd.Length);
// create the hash
var sha1Hasher = new SHA1CryptoServiceProvider();
byte[] hashedDataBytes = sha1Hasher.ComputeHash(operand);
return Convert.ToBase64String(hashedDataBytes);
}
CreatePasswordDigest(Convert.FromBase64String("UIYifr1SPoNlrmmKGSVOug=="), "2009-12-03T16:14:49Z", "test8")
which returns yf2yatQzoaNaC8BflCMatVch/B8= as we want.
Remember to use the same createdTime in the digest as you put in the XML, this might sound obvious, but some people include milliseconds on their timestamps and some don't - it doesn't matter, it just needs to be consistent.
Also the Id field in the UsernameToken XML doesn't matter - it doesn't need to change.
Here's a method to create a Nonce like the one above, if you don't want to use GUIDs like Rick uses:
private byte[] CreateNonce()
{
var Rand = new RNGCryptoServiceProvider();
//make random octets
byte[] buf = new byte[0x10];
Rand.GetBytes(buf);
return buf;
}
I hope that helps someone - it took me lots of frustration, trial and error, searching web pages, and general head/wall banging.
I have a problem recreating a password that's hashed in C#. In an online project when a user do the registration process his password is save after passing for this function:
private static string ToMD5Hash(string inputString)
{
using (MD5 md5 = MD5.Create())
{
byte[] data = Encoding.Unicode.GetBytes(inputString);
byte[] hash = md5.ComputeHash(data);
return Convert.ToBase64String(hash);
}
}
I'm working on an offline version that at some point will do a sync with the online version and I can't reproduce the same results on AS3 (Adobe Air). For example the password "1234" after passing to the C# code will be "DwN1hMmef9T0+MWVUPj1Bw==".
Can someone help me out?
My AS3 code is like this:
private function encode():void
{
var ba:ByteArray = new ByteArray();
ba.writeMultiByte("1234","unicode");
var str:String = MD5.hash(ba.toString());
var ba2:ByteArray = new ByteArray();
ba2.writeMultiByte(str.toString(),"unicode");
var encoder:Base64Encoder = new Base64Encoder();
encoder.encodeUTFBytes(ba2.toString());
trace(encoder.toString());
}
When I do the ba.writeMultiByte("1234","unicode"); I get exactly the same ByteArray as in the C# but when I do the MD5.hash(ba.toString()); the new ByteArray is different.
So it looks like it may be a bug in as3corelib's implementation of writing the bits to the digest ByteArray.
It seemingly writes them in big endian format, rather than little endian. Or more specifically, it writes the bits as a collection of 4 integers rather than a collection of bytes, and in the process it mangles the byte order of the bits (which is why you are seeing different Base64 results -- the bytes are in a different order).
You can address this bug by adding on line 184 of MD5.as, in as3corelib, insert the following one line of code:
digest.endian = Endian.LITTLE_ENDIAN;
Also make sure you add an import at the top of the file for import flash.utils.Endian. I've created a public gist of the changes available here
Then it should generate the same byte order as c#, and then it should Base64 encode the same way. I verified it using the following as3 function:
private function encode():void
{
var ba:ByteArray = new ByteArray();
ba.endian = Endian.BIG_ENDIAN
ba.writeMultiByte("1234","unicode");
var str:String = MD5.hashBytes(ba);
var encoder:Base64Encoder = new Base64Encoder();
encoder.encodeBytes(MD5.digest);
trace(encoder.toString()); // DwN1hMmef9T0+MWVUPj1Bw==
}
Look at this.
Be aware that the original download mentioned in that site has a few bugs, so that you have to use the corrected vesion that you can find in the same post.
If you're using AS3CoreLib, do it this way:
Different MD5 with as3corelib.
I am trying to encrypt data with a password and store it inside a ASN.1 encoded CMS message (using C# and BouncyCastle 1.4)
The code I have seems to have two problems:
the data does not seem to be signed with a HMAC, so when I tamper with the encodedData (by enabling the commented out line), the decryption still succeeds.
when I decrypt the data I have tampered with, I get beck corrupted plain text. However only a two blocks of plaintext data are corrupted. This seems to suggest that the encryption does not actually use CBC mode.
(edit: disregard the second point, this is exactly how CBC is supposed to work)
This is what I am testing with:
public void TestMethod1()
{
byte[] data = new byte[1024]; // plaintext: a list of zeroes
CmsEnvelopedDataGenerator generator = new CmsEnvelopedDataGenerator();
CmsPbeKey encryptionKey = new Pkcs5Scheme2PbeKey("foo", new byte[] { 1, 2, 3 }, 2048);
generator.AddPasswordRecipient(encryptionKey, CmsEnvelopedDataGenerator.Aes256Cbc);
CmsProcessableByteArray cmsByteArray = new CmsProcessableByteArray(data);
CmsEnvelopedData envelopeData = generator.Generate(cmsByteArray, CmsEnvelopedDataGenerator.Aes256Cbc);
byte[] encodedData = envelopeData.GetEncoded();
// encodedData[500] = 10; // tamper with the data
RecipientID recipientID = new RecipientID();
CmsEnvelopedData decodedEnvelopeData = new CmsEnvelopedData(encodedData);
RecipientInformation recipient = decodedEnvelopeData.GetRecipientInfos().GetFirstRecipient(recipientID);
byte[] data2 = recipient.GetContent(encryptionKey);
CollectionAssert.AreEqual(data, data2);
}
What am I doing wrong? What would be the correct way to write this?
To add an HMAC to a CMS message, you would have to use a AuthenticatedData-structure.
I am not especially familiar with Bouncy Castle, but from a cursory look at the API, I would say that it does not support AuthenticatedData. In fact, it looks like it only supports SignedData for authentication.
So your options seems to be:
Use another library (or write your own code) to handle the AuthenticatedData-structure.
Calculate the HMAC and provide it in a non-standard way (in a proprietary Attribute or out-of-band).
Use SignedData with an RSA key pair instead.