I am working on an assignment regarding RSA encryption and decryption. For this assignment we have to be able to encrypt and decrypt ANY size key.
I have written code that successfully encrypts and decrypts (given that key does not exceed the maximum key size) but now I need to modify this code to handle any size.
public partial class Form1 : Form
{
//Strings to hold public & private keys
String publicKey, privateKey;
UnicodeEncoding encoder = new UnicodeEncoding();
public Form1()
{
RSACryptoServiceProvider myRSA = new RSACryptoServiceProvider();
InitializeComponent();
privateKey = myRSA.ToXmlString(true);
publicKey = myRSA.ToXmlString(false);
}
private void btnEncrypt_Click(object sender, EventArgs e)
{
var myRSA = new RSACryptoServiceProvider();
//Set cryptoserviceprovider with the proper key
myRSA.FromXmlString(publicKey);
//Encode the data to encrypt as a byte array
var dataToEncrypt = encoder.GetBytes(txtPlain.Text);
//Encrypt the byte array
var encryptedByteArray = myRSA.Encrypt(dataToEncrypt, false).ToArray();
var length = encryptedByteArray.Count();
var item = 0;
var sb = new StringBuilder();
//Change each byte in the encrypted byte array to text
foreach(var x in encryptedByteArray)
{
item++;
sb.Append(x);
if (item < length) sb.Append(",");
}
txtCypher.Text = sb.ToString();
}
private void btnDecrypt_Click(object sender, EventArgs e)
{
var myRSA = new RSACryptoServiceProvider();
//Split data into an array
var dataArray = txtCypher.Text.Split(new char[] { ',' });
//Convert chars to bytes
byte[] dataByte = new byte[dataArray.Length];
for(int i = 0; i < dataArray.Length; i++) dataByte[i] = Convert.ToByte(dataArray[i]);
//Decrypt the byte array
myRSA.FromXmlString(privateKey);
var decryptedBytes = myRSA.Decrypt(dataByte, false);
//place into cypher text box
txtPlain.Text = encoder.GetString(decryptedBytes);
}
}
I have come up with the following code to be able to encrypt any size (which seems to work to my knowledge):
//store dataLength
int dataLength = dataToEncrypt.Length;
//Check if dataLength > 117
if (dataLength > 117)
{
//Divide dataLength by 128 to determine how many cycles will be needed
float numOfCycles = (dataLength / 117);
//round up to nearest whole number
cycles = (int)Math.Ceiling(numOfCycles);
//for however many cycles
for (int i = 0; i < cycles; i++)
{
var myByteArray = new byte[117];
for (int j = 0; j < 117; j++)
{
int currentByte = i * 117 + j;
myByteArray[j] = dataToEncrypt[currentByte];
}
var encryptedByteArray = myRSA.Encrypt(myByteArray, false).ToArray();
var length = encryptedByteArray.Count();
var item = 0;
//Change each byte in the encrypted byte array to text
foreach (var x in encryptedByteArray)
{
item++;
sb.Append(x);
if (item < length) sb.Append(",");
}
txtCypher.Text = sb.ToString();
}
}
else
{
var encryptedByteArray = myRSA.Encrypt(dataToEncrypt, false).ToArray();
var length = encryptedByteArray.Count();
var item = 0;
var sb = new StringBuilder();
//Change each byte in the encrypted byte array to text
foreach(var x in encryptedByteArray)
{
item++;
sb.Append(x);
if (item < length) sb.Append(",");
}
txtCypher.Text = sb.ToString();
}
And the following code to decrypt any size which ISN'T WORKING:
float length = dataArray.Count();
float numOfCycles = (length / 117);
int cycles = (int)Math.Ceiling(numOfCycles);
for (int i = 0; i < cycles; i++)
{
byte[] dataByte = new byte[117];
for(int j = 0; j < 117; j++)
{
//Convert chars to bytes
dataByte[j] = Convert.ToByte(dataArray[ i * 117 + j ]);
}
//Decrypt the byte array
myRSA.FromXmlString(privateKey);
var decryptedBytes = myRSA.Decrypt(dataByte, false);
txtPlain.Text += encoder.GetString(decryptedBytes);
}
The following line: var decryptedBytes = myRSA.Decrypt(dataByte, false);
throws the error: An unhandled exception of type 'System.Security.Cryptography.CryptographicException' occurred in mscorlib.dll
Additional information: Bad Data.
I have no idea why this error is being thrown, I have debugged the program and the data stored in dataByte[] does not seem to be any different from what was being stored inside when I successfully decrypt a message of size smaller than the maximum key size.
RSA, as a asymmetric cryptosystem, is really slow compared to symmetric ciphers like AES. The proper way to implement this would be to generate a symmetric AES key, encrypt the actual data with AES (e.g. with CBC/CTR/GCM/EAX) and then encrypt the single key with RSA.
There is no standard to apply RSA to multiple "blocks", so it is not recommended to implement this. If you still want to implement it, here are some problems with your code.
RSA needs padding to be secure. Since you're using a 128 byte key and split the plaintext into 117 byte blocks, this means that you're using PKCS#1 v1.5 padding. Each block will be padded and then encrypted. The ciphertext is numerically strictly smaller than the modulus, but usually pretty close to it. Most ciphertexts will be 128 byte long, but some may be smaller, because a ciphertext is a serialized number where the leading zeros don't have to be stored. That means that you cannot simply write them one after the other and expect to read them back correctly.
You either need to pad the ciphertexts with 0x00 bytes so that they are all of the same length (128 byte) or store the length of each ciphertext before the corresponding ciphertext so that you know how much you have to read.
Related
I am trying to decrypt a ciphersaber encrypted hexadecimal message using an IV mixing round of 20 with the key MyKey.
The messages is:
bad85d9e7f5aff959b6b332b44af2cc554d8a6eb
I am doing this in pure C# and it should return the message: Hola Mundo
using System;
using System.Text;
public class Program
{
public static void Main(string[] args)
{
// Hexadecimal text
string hexText = "bad85d9e7f5aff959b6b332b44af2cc554d8a6eb";
// Convert hexadecimal text to byte array
byte[] encryptedData = new byte[hexText.Length / 2];
for (int i = 0; i < encryptedData.Length; i++)
{
encryptedData[i] = Convert.ToByte(hexText.Substring(i * 2, 2), 16);
}
// IV length
int ivLength = 1;
// Key loop iterations
int keyIterations = 20;
// Encryption key
string encryptionKey = "MyKey";
// Convert encryption key to byte array
byte[] keyData = Encoding.UTF8.GetBytes(encryptionKey);
// Create an array to store the IV
byte[] ivData = new byte[ivLength];
// Copy the first `ivLength` bytes of the encrypted data to the IV array
Array.Copy(encryptedData, 0, ivData, 0, ivLength);
// Create an array to store the encrypted message
byte[] messageData = new byte[encryptedData.Length - ivLength];
// Copy the remaining bytes of the encrypted data to the message data array
Array.Copy(encryptedData, ivLength, messageData, 0, messageData.Length);
// Create an array to store the decrypted message
byte[] decryptedData = new byte[messageData.Length];
// Perform the decryption
for (int i = 0; i < messageData.Length; i++)
{
decryptedData[i] = (byte)(messageData[i] ^ keyData[i % keyData.Length]);
for (int j = 0; j < keyIterations; j++)
{
decryptedData[i] = (byte)(decryptedData[i] ^ ivData[j % ivData.Length]);
}
}
// Convert the decrypted data to a string and print it
string decryptedMessage = Encoding.UTF8.GetString(decryptedData);
Console.WriteLine("Decrypted message: " + decryptedMessage);
}
}
Now when I try it returns: �$�#���Jf=�I���
What mistake am I making in the code or am I implementing it wrong?
I tested the text with the following site to see if it was ok: https://ruletheweb.co.uk/cgi-bin/saber.cgi
CipherSaber uses as IV the first 10 bytes of the encrypted message. The rest is the actual ciphertext. The IV is appended to the key (giving the key setup input), which is used as input to the CipherSaber key setup, see CipherSaber, Technical description, 1st section.
In the posted code, an IV length of 1 is applied instead of 10, which incorrectly determines IV (and thus key setup input) and actual ciphertext. The correct determination of IV and actual ciphertext is:
private static (byte[], byte[]) SeparateIvCiphertext(byte[] ivCiphertext)
{
int ivLen = 10;
byte[] iv = new byte[ivLen];
Buffer.BlockCopy(ivCiphertext, 0, iv, 0, iv.Length);
byte[] ciphertext = new byte[ivCiphertext.Length - iv.Length];
Buffer.BlockCopy(ivCiphertext, iv.Length, ciphertext, 0, ciphertext.Length);
return (iv, ciphertext);
}
and of the key setup input:
private static byte[] GetKeySetupInput(byte[] key, byte[] iv)
{
byte[] keySetupInput = new byte[key.Length + iv.Length];
Buffer.BlockCopy(key, 0, keySetupInput, 0, key.Length);
Buffer.BlockCopy(iv, 0, keySetupInput, key.Length, iv.Length);
return keySetupInput;
}
Furthermore, the decryption itself seems to be implemented incorrectly or at least incompletely. CipherSaber uses RC4 as its encryption/decryption algorithm, which can be divided into a key setup and the actual encryption/decryption:
The referenced website performs decryption using CipherSaber-2. Compared to the original CipherSaber (referred to as CipherSaber-1), a modified key setup is used in which the CipherSaber-1/RC4 key setup is repeated multiple times, 20 times in the case of the posted data.
A description of the CipherSaber-1/RC4 key setup can be found here, Key-scheduling algorithm (KSA), a possible implementation for CipherSaber-2 is:
private static byte[] sBox = new byte[256];
private static void KeySetup(byte[] input, int iterations)
{
for (int i = 0; i < 256; i++)
{
sBox[i] = (byte)i;
}
int j = 0;
for (int cs2loop = 0; cs2loop < iterations; cs2loop++) // CipherSaber-2 modification
{
for (int i = 0; i < 256; i++)
{
j = (j + sBox[i] + input[i % input.Length]) % 256;
Swap(ref sBox[i], ref sBox[j]);
}
}
}
private static void Swap(ref byte val1, ref byte val2)
{
if (val1 == val2) return;
val1 = (byte)(val1 ^ val2);
val2 = (byte)(val2 ^ val1);
val1 = (byte)(val1 ^ val2);
}
The loop marked CipherSaber-2 modification in the code snippet is the modification compared to CipherSaber-1/RC4!
The actual encryption/decryption is described here, Pseudo-random generation algorithm (PRGA), a possible implememtation is:
private static byte[] Process(byte[] input)
{
int i = 0, j = 0;
byte[] result = new byte[input.Length];
for (int k = 0; k < input.Length; k++)
{
i = (i + 1) % 256;
j = (j + sBox[i]) % 256;
Swap(ref sBox[i], ref sBox[j]);
result[k] = (byte)(sBox[(sBox[i] + sBox[j]) % 256] ^ input[k]);
}
return result;
}
Note that this algorithm is used for both encryption and decryption.
With this, the posted encrypted message can be decrypted as follows:
using System;
using System.Text;
...
byte[] key = Encoding.UTF8.GetBytes("MyKey");
byte[] encryptedData = Convert.FromHexString("bad85d9e7f5aff959b6b332b44af2cc554d8a6eb");
(byte[] iv, byte[] ciphertext) = SeparateIvCiphertext(encryptedData);
byte[] keySetupInput = GetKeySetupInput(key, iv);
int iterations = 20;
KeySetup(keySetupInput, iterations);
byte[] plaintext = Process(ciphertext);
Console.WriteLine(Encoding.UTF8.GetString(plaintext)); // Hola Mundo
which gives Hola Mundo as plaintext.
I read a file as binary, convert to hex string, convert back to binary, and write to a new file.
I expect a duplicate, but get a corrupted file.
I have been trying different ways to convert the binary into the hex string but can't seem to retain the entire file.
byte[] binary1 = File.ReadAllBytes(#"....Input.jpg");
string hexString = "";
int counter1 = 0;
foreach (byte b in binary1)
{
counter1++;
hexString += (Convert.ToString(b, 16));
}
List<byte> bytelist = new List<byte>();
int counter2 = 0;
for (int i = 0; i < hexString.Length/2; i++)
{
counter2++;
string ch = hexString.Substring(i*2,2);
bytelist.Add(Convert.ToByte(ch, 16));
}
byte[] binary2 = bytelist.ToArray();
File.WriteAllBytes(#"....Output.jpg", binary2);
Counter 1 and counter 2 should be the same count, but counter 2 is always about 10% smaller.
I want to get a duplicate file output that I could have transferred around via that string value.
Converting byte 10 will give a single char, and not 2 characters. Your convert back method (logically) build on 2 chars per byte.
this case works
byte[] binary1 = new byte[] { 100 }; // convert will result in "64"
and this case fails
byte[] binary1 = new byte[] { 10 }; // convert will result in "a"
I quick fixed your code, by padding with a "0" in case of a single char.
so working code:
byte[] binary1 = new byte[] { 100 };
string hexString = "";
int counter1 = 0;
foreach (byte b in binary1)
{
counter1++;
var s = (Convert.ToString(b, 16));
// new
if (s.Length < 2)
{
hexString += "0";
}
// end new
hexString += s;
}
List<byte> bytelist = new List<byte>();
int counter2 = 0;
for (int i = 0; i < hexString.Length / 2; i++)
{
counter2++;
string ch = hexString.Substring(i * 2, 2);
var item = Convert.ToByte(ch, 16);
bytelist.Add(item);
}
byte[] binary2 = bytelist.ToArray();
Please note, your code could use some refactoring, e.g. don't string concat in a loop and maybe check the Single Responsibility Principle.
Update, got it fixed, but there are better solutions here: How do you convert a byte array to a hexadecimal string, and vice versa?
I have the following function:
public void SetTagData(string _data)
{
string data = _data;
byte[] ba = Encoding.Default.GetBytes(data);
string hexString = BitConverter.ToString(ba);
hexString = hexString.Replace("-", "");
var blockStart = 0;
var bufferHexBlocks = String.Empty;
try
{
for (var i = 0; i < hexString.Length; i++)
{
var byteList = new List<byte>();
byte[] datablockKey = ConvertHelpers.ConvertHexStringToByteArray(i.ToString().PadLeft(2, '0'));
var block = hexString.Substring(blockStart, 8);
byte[] datablockValue = ConvertHelpers.ConvertHexStringToByteArray(block);
byteList.AddRange(datablockKey);
byteList.AddRange(datablockValue);
_reader.Protocol("wb", byteList.ToArray());
blockStart += 8;
}
}
catch (Exception ex)
{
console.log(ex.message);
}
}
The data coming in is a bunch of hex as a string. I need to split this hex string into batches of 8 characters, append an incrementing 0 padded hex number from 00 to 1f and send this new string as a byte array to the _reader.Protocol function, which accepts a string wb as first parameter and the block as the second.
For example incoming data is:
string data = "3930313B36313B5350542D53504C3B3830303B3B352E373B3B303B303B3B3B34353036383B4E3B4E3B"
I need to send the following to the _reader.Protocol object:
(incremented padded hex 01, 02, 03, ... , 0f) and the first 8 characters of the data string, then the next, and so on as a byte array.
[013930313B], [0236313B53], etc.
I think I'm getting close... but missing something...
My problem at the moment is that I can't figure out how to loop in blocks of 8 and if the hex string is say 82 characters instead of 80 (multiple of 8), then how would I grab the last two characters without getting a IndexOutofRange exception.
Note: This is for a Windows CE application, so no new C# features please.
This below will work fine in conjunction with this answer and the sample string data given.
public static byte[] Parse(string data)
{
var count = data.Length / 8; //Might be worth throwing exception with any remainders unless you trust the source.
var needle = 0;
List<byte> result = new List<byte>(); //Inefficient but I'm being lazy
for (int i = 0; i < count; i++)
{
char[] buffer = new char[8];
data.CopyTo(needle, buffer, 0, buffer.Length);
//To get around the odd number when adding the prefixed count byte, send the hex string to the convert method separately.
var bytes = ConvertHexStringToByteArray(new string(buffer)); //Taken From https://stackoverflow.com/a/8235530/6574422
//As the count is less than 255, seems safe to parse to single byte
result.Add(byte.Parse((i + 1).ToString()));
result.AddRange(bytes);
needle += 8;
}
return result.ToArray();
}
I'm figured it out. It might not be the most efficient solution but it works just fine. I did it using a for loop inside a for loop.
In case anyone is interested here is the final code:
public void SetTagData(string _data)
{
string data = _data;
byte[] ba = Encoding.Default.GetBytes(data);
string hexString = BitConverter.ToString(ba);
hexString = hexString.Replace("-", "");
var blockStart = 0;
try
{
_reader.Protocol("s");
for(var count = 0; count < 16; count++)
{
var byteList = new List<byte>();
byte[] datablockKey = ConvertHelpers.ConvertHexStringToByteArray(count.ToString("X2"));
byteList.AddRange(datablockKey);
for (var innerCount = 0; innerCount < 4; innerCount++)
{
var block = String.Empty;
if (!String.IsNullOrEmpty(hexString.Substring(blockStart, 2)))
{
block = hexString.Substring(blockStart, 2);
}
else
{
block = "20";
}
byte[] datablockValue = ConvertHelpers.ConvertHexStringToByteArray(block);
byteList.AddRange(datablockValue);
blockStart += 2;
}
_reader.Protocol("wb", byteList.ToArray());
}
}
catch (Exception)
{
}
}
I am trying to convert c++ source to c# which encrypt and decrypt file using Rinjdael cryptography.
But c++ source has got a little bit difference from the normal en/decryptions.
And I am not really good at c++, so I am getting confused.
One of my customers' application is written in VC++, and to convert it into c# is part of my job.
And the previous c++ developer used open source code from http://www.codeproject.com/Articles/10657/A-Simple-Portable-Rinjdael-AES-Based-Stream-Cipher to manipulate en/decryption.
Here is c++ source codes.
int DCipher::DecryptFile(LPCTSTR szSrcFile, LPCTSTR szDestFile, const char* pwd, int head[19])
{
if(CheckMemSize() != 0)
return INSUFFICIENT_MEMORY;
FileSize=CurPosition=0;
_tcscpy(SrcFile, szSrcFile);
_tcscpy(OutFile, szDestFile);
//_tcscpy(OutFile, _T(".enc"));
strcpy(password, pwd);
for(int i=0; i<19; i++)
{
header[i] = head[i];
}
FILE *r, *w;
GetFileLength();
int nCheck = CheckIfEncrypted();
if(nCheck != ENCRYPTED_FILE )
return nCheck; //either NORMAL_FILE or BAD_SIGNATURE
if((r = _tfopen(SrcFile, _T("rb"))) == NULL)
return ERROR_SRC_FILE;
if((w = _tfopen(OutFile, _T("wb"))) == NULL)
{
fclose(r);
return ERROR_DST_FILE;
}
char zzz[26]; //fixed invalid pointer - DKeesler
fread(zzz, 25, 1, r); // Skip first 25 bytes of the file.
int pad = header[19];
pad *= 10;
pad += header[20];
// convert password to Rijndael key
strcpy((char*)key, (const char*)CalcMD5FromString((const char*)password));
/***************************************
Decryption algorithm
***************************************/
int rval = NO_ERRORS_DONE;
FileSize -= 25;
unsigned int BUFF_SIZE = liChunkSize;
unsigned int WRITE_SIZE = liChunkSize;
int nRound = FileSize / liChunkSize;
unsigned int LAST_BLOCK = FileSize % liChunkSize;
if(LAST_BLOCK >= 1)
nRound++;
const unsigned char* intext;
unsigned char* output;
intext = (const unsigned char*)malloc(BUFF_SIZE);
output = (unsigned char*)malloc(BUFF_SIZE+16);
if(intext == NULL || output == NULL)
{
fclose(r);
fclose(w);
return ALLOC_ERROR;
}
Rijndael rj;
rj.init(Rijndael::CBC, Rijndael::Decrypt, key, Rijndael::Key32Bytes);
for(int loop=1; loop <= nRound; loop++)
{
if(loop == nRound && LAST_BLOCK >= 1)
{
BUFF_SIZE = LAST_BLOCK;
WRITE_SIZE = LAST_BLOCK - pad;
}
fread((void*)intext, sizeof(char), BUFF_SIZE, r); // read plaintext into intext[] buffer
int bsize = BUFF_SIZE*8;
int len = rj.blockDecrypt((const UINT8*)intext, bsize, (UINT8*)output);
if(len >= 0)
{
fwrite((const void*)output, sizeof(char), WRITE_SIZE, w);
}
else
{
rval = READ_WRITE_ERROR;
break;
}
}
fclose(r); //close input file
fclose(w); //close output file
free((void*)intext);
free((void*)output);
//change these two lines if you want to leave backups or unencrypted copies...
//that would sort of defeat the purpose of encryption in my mind, but it's your
// app so write it like you want it.
if(DECRYPTION_CANCEL == rval) {
_tremove(OutFile);
}
else {
//_tremove(SrcFile); //remove input file
//_trename(OutFile, SrcFile); //rename output file to input filename
}
return rval; //ZERO .. see defines for description of error codes.
}
And c# source code is from https://msdn.microsoft.com/en-us/library/system.security.cryptography.rijndael(v=vs.110).aspx.
And I changed a little bit of codes.
Here is c# codes.
public int DecryptFile(string SourceFilePath, string DestFilePath, string Password, string Signature)
{
try
{
FileSize = CurPosition = 0;
FileInfo _fi = new FileInfo(SourceFilePath);
FileSize = _fi.Length;
// copy the signature to _header
for(int i = 0; i < 19; i++)
{
_header[i] = (byte)Signature[i];
}
/*
* check if the file is valid encrypted file.
*/
int nCheck = this.CheckIfEncrypted(SourceFilePath);
switch (nCheck)
{
case ENCRYPTED_FILE:
// The file is an encrypted file.
break;
case NORMAL_FILE:
throw new ArgumentException("The file is a normal file.");
case BAD_SIGNATURE:
throw new ArgumentException("User signature doesn't match.");
}
int pad = _header[19];
pad *= 10;
pad += _header[20];
// Rijndael session key
byte[] session_key = this.CalcMD5FromString(Password);
byte[] _restFileBytes = new byte[_fi.Length - 25];
using (FileStream _fs = new FileStream(SourceFilePath, FileMode.Open, FileAccess.Read))
{
_fs.Read(_restFileBytes, 0, _restFileBytes.Length);
}
int rval = NO_ERRORS_DONE;
FileSize -= 25;
int BUFF_SIZE = liChunkSize;
int WRITE_SIZE = liChunkSize;
int nRound = (int)FileSize / liChunkSize;
int LAST_BLOCK = (int)FileSize % liChunkSize;
if(LAST_BLOCK >= 1)
nRound++;
byte[] intext = new byte[BUFF_SIZE];
byte[] output = new byte[BUFF_SIZE + 16];
if (intext.Length == 0 || output.Length == 0)
{
return ALLOC_ERROR;
}
for (int loop = 1; loop <= nRound; loop++)
{
if (loop == nRound && LAST_BLOCK >= 1)
{
BUFF_SIZE = LAST_BLOCK;
WRITE_SIZE = LAST_BLOCK - pad;
}
intext = new byte[BUFF_SIZE];
System.Buffer.BlockCopy(_restFileBytes, (loop - 1) * this.liChunkSize, intext, 0, BUFF_SIZE);
int bsize = BUFF_SIZE * 8; // -> I still couldn't figure out what this bsize does on Rijndael decryption.
using (RijndaelManaged myRijndael = new RijndaelManaged())
{
myRijndael.Key = session_key;
//myRijndael.BlockSize = bsize;
//myRijndael.Padding = PaddingMode.None;
myRijndael.GenerateIV();
using (Rijndael rijAlg = Rijndael.Create())
{
rijAlg.Key = myRijndael.Key;
rijAlg.IV = myRijndael.IV;
// Create a decrytor to perform the stream transform.
ICryptoTransform decryptor = rijAlg.CreateDecryptor(rijAlg.Key, rijAlg.IV);
// Create the streams used for decryption.
using (MemoryStream msDecrypt = new MemoryStream(intext))
{
using (CryptoStream csDecrypt = new CryptoStream(msDecrypt, decryptor, CryptoStreamMode.Read))
{
//using (StreamReader srDecrypt = new StreamReader(csDecrypt))
//{
// // Read the decrypted bytes from the decrypting stream and place them in a string.
// //string s = srDecrypt.ReadToEnd();
//}
byte[] rettt = msDecrypt.ToArray();
} // --> Padding is invalid and cannot be removed error occurs here and msDecrypt byte array is just same as intext. So, it's not decrypted at all.
}
}
}
}
return rval;
}
catch
{
throw;
}
}
According to Keesler(who is the writer of c++ source codes from codeproject.com), first 25 bytes are filled with user data(signature, padding and file status). So, I skipped first 25 bytes and save the rest bytes to _restFileBytes varialbes(byte array).
And Keesler has a variable called chunk size, which splits file bytes into chunk size(as long as I understand).
Anyway, I think I almost converted to c# but I still get this error message "Padding is invalid and cannot be removed" when CryptoStream disposing in c#.
Can anyone give me some guide to fix this error?
None should be used as padding mode. It seems like your colleague and the author of the original article made up their own padding scheme.
Furthermore, all of the ciphertext should be streamed from the file (making sure you read all the bytes). Currently you are restarting encryption with the IV for each chunk, which is not good, the IV should only be used at the start of the ciphertext.
Print out the key in hex for both C++ and C# and compare before you start.
Note that the Read method differs slightly from the fread method in C++.
I have to write a Vigenere encryption / decryption function that operates on full bytes (to encrypt and send files over tcp and then decrypt on the other side).
My encrypting function seems to be working (more or less, can't really test it without decrypting function).
This is the code of the encrypting function:
public static Byte[] encryptByteVigenere(Byte[] plaintext, string key)
{
Byte[] result= new Byte[plaintext.Length];
key = key.Trim().ToUpper();
int keyIndex = 0;
int keylength = key.Length;
for (int i = 0; i < plaintext.Length; i++)
{
keyIndex = keyIndex % keylength;
int shift = (int)key[keyIndex] - 65;
result[i] = (byte)(((int)plaintext[i] + shift) % 256);
keyIndex++;
}
return result;
}
However, the decrypting function, even though wrote in pretty much the same way, causes an error.
"Attempted to divide by zero."
The code of the decrypting function:
public static Byte[] decryptByteVigenere(Byte[] ciphertext, string key)
{
Byte[] result = new Byte[ciphertext.Length];
key = key.Trim().ToUpper();
int keyIndex = 0;
int keylength = key.Length;
for (int i = 0; i < ciphertext.Length; i++)
{
keyIndex = keyIndex % keylength;
int shift = (int)key[keyIndex] - 65;
result[i]= (byte)(((int)ciphertext[i] + 256 - shift) % 256);
keyIndex++;
}
return result;
}
The error points at the line
keyIndex = keyIndex % keylength;
But what wonders me is that the code is pretty much the same in the first function and it doesn't seem to cause any trouble. I'm testing it on the received fild, which arrives correctly without encryption. Could anyone help me with that?
EDIT:
The method / thread that is using the decryption function code:
public void fileListenThread()
{
try
{
fileServer.Start();
String receivedFileName = "test.dat";
String key = (textKlucz.Text).ToUpper();
while (true)
{
fileClient = fileServer.AcceptTcpClient();
NetworkStream streamFileServer = fileClient.GetStream();
int thisRead = 0;
int blockSize = 1024;
Byte[] dataByte = new Byte[blockSize];
Byte[] dataByteDecrypted = new Byte[blockSize];
FileStream fileStream = new FileStream(receivedFileName, FileMode.Create);
while (true)
{
thisRead = streamFileServer.Read(dataByte, 0, blockSize);
dataByteDecrypted = Program.decryptByteVigenere(dataByte, key);
fileStream.Write(dataByteDecrypted, 0, thisRead);
if (thisRead == 0)
break;
}
fileStream.Close();
}
}
catch (SocketException e)
{
MessageBox.Show("SocketException: " + e, "Wystąpił wyjątek", MessageBoxButtons.OK, MessageBoxIcon.Error);
}
}
Ok the problem was indeed the sending / receiving method, not the function itself. I still don't really know what caused the problem, but rewriting the functions helped. Thanks for your input!
I'm leaving it here in case someone needed such function in the future... even though it's rather trivial thing.
Cheers.