Library tutorials & articles

Using Encryption in .NET

Overview

Summary

In this article I will explain the classes and code necessary to put encryption to work in your applications.  I will also explain the basic principles of operation of these classes and point out several pitfalls to avoid.

Introduction

In today's hostile, interconnected computing environment, we often find the need to protect data while it is stored on a physical medium or transmitted over a network. To achieve this goal, we turn to cryptography. While the field of cryptography entails much more, in this article we will focus on encryption; specifically the ciphers available in the .NET Framework Class Library (FCL) and how to correctly utilize them.

There exists a group of developers who approach the subject of cryptography with a cavalier attitude, who think they can wave the wand of cryptography at their applications and make them secure. This group of developers mistakenly believes that applying cryptography is fool-proof and rests secure in the fact that their applications utilize "128-bit encryption" or comply with some other such buzzword. I hope that this article will awaken this group to the many pitfalls and subtleties that can silently render your crypto system easily attackable.

While the application of encryption technologies to applications is a subject that must be approached with a modicum of care and understanding, the FCL has made excellent cryptographic technologies more accessible than ever, though there are still a number of principles that must be adhered to and pitfalls to be avoided.

Why use the FCL classes? Can't I just "roll my own"?

Although there are several reasons why you shouldn't do this, I'll just state the primary reason: You'll get it wrong. While the application of standard, public, seasoned algorithms is relatively simple and accessible to the common developer, the design of the cryptographic algorithms themselves is extraordinarily difficult and is best left to the elite of the field. While the output of your cipher may look completely secure to you, a skilled cryptanalyst can likely easily find many weaknesses, such as simple algebraic relationships between input, output and key. The tiniest weakness can reveal information about the plaintext or, worse yet, the key. Unless you happen to be an elite cryptographer, stick with standard, public, seasoned algorithms and proven implementations.

This article was originally published at DotNetDevs.com

Comments

  1. 19 Jan 2009 at 18:40
    Good the article. I have used the base and added support for: - automatically Combining Message + IV - uuEncode/Decode support to write as String - Direct String accessors to quick support for Strings - Direct Object accessors to quick support for Objects. As my thank you, i have attached the entire class below: class Cipher { RijndaelManaged _cipher = null; #region Construction public Cipher() { CreateCipher(); _cipher.GenerateKey(); } public Cipher(byte[] key) { CreateCipher(); Key = key; } public Cipher(String keyPhrase) { CreateCipher(); KeyPhrase = keyPhrase; } private void CreateCipher() { _cipher = new RijndaelManaged(); _cipher.KeySize = 256; _cipher.BlockSize = 256; _cipher.Mode = CipherMode.CBC; _cipher.Padding = PaddingMode.ISO10126; } #endregion Construction #region Key Support public byte[] Key { get { return _cipher.Key; } set { _cipher.Key = value; } } // String Key Support public String KeyPhrase { get { return Encoding.ASCII.GetString(_cipher.Key); } set { // Algo supports only 128, 192, or 256 bits // i.e. 16, 24 or 32 bytes // Ensure we have a 32byte key for 256bit encryption String sKey = value; if (sKey.Length > 32) sKey = sKey.Substring(0, 32); else { while (sKey.Length < 32) { if ((sKey.Length % 2) == 0) sKey = "R" + sKey; else sKey = sKey + "j"; } } _cipher.Key = Encoding.ASCII.GetBytes(sKey); } } #endregion Key Support #region Encrypt/Decrypt Message+IV public byte[] EncryptMessage(byte[] plainText, out byte[] iv) { _cipher.GenerateIV(); iv = _cipher.IV; ICryptoTransform transform = _cipher.CreateEncryptor(); byte[] cipherText = transform.TransformFinalBlock(plainText, 0, plainText.Length); return cipherText; } public byte[] DecryptMessage(byte[] cipherText, byte[] iv) { _cipher.IV = iv; ICryptoTransform transform = _cipher.CreateDecryptor(); byte[] plainText = transform.TransformFinalBlock(cipherText, 0, cipherText.Length); return plainText; } #endregion Encrypt/Decrypt Message+IV #region Encrypt/Decrypt wrapper to combine IV + Message void BlockCopy(ref byte[] buf, ref byte[] obj, ref int nOffset, bool bWrite) { if (bWrite) { // Pack Buffer.BlockCopy(obj, 0, buf, nOffset, obj.Length); } else { // Unpack Buffer.BlockCopy(buf, nOffset, obj, 0, obj.Length); } nOffset += obj.Length; } // Enc Wrapper to combine IV + Message public byte[] EncryptMessage(byte[] plainText) { byte[] iv; byte[] encr; encr = EncryptMessage(plainText, out iv); // Pack format: short cIVLen = (short)iv.Length; byte[] data = new byte[iv.Length + encr.Length + sizeof(short)]; byte[] byIV = BitConverter.GetBytes(cIVLen); int nOffset = 0; BlockCopy(ref data, ref byIV, ref nOffset, true); BlockCopy(ref data, ref iv, ref nOffset, true); BlockCopy(ref data, ref encr, ref nOffset, true); return data; } // Enc Wrapper to combine IV + Message public byte[] DecryptMessage(byte[] data) { // Pack format: // Unpack to get IV+encrText short cIVLen = BitConverter.ToInt16(data, 0); int nOffset = sizeof(short); byte[] iv = new byte[cIVLen]; byte[] encr = new byte[data.Length - sizeof(short) - cIVLen]; // Read iv and encr message BlockCopy(ref data, ref iv, ref nOffset, false); BlockCopy(ref data, ref encr, ref nOffset, false); // Decrypt and return return DecryptMessage(encr, iv); } #endregion Encrypt/Decrypt wrapper to combine IV + Message #region UUEncode for String support public String uuEncrypt(byte[] byPlain) { byte[] byEncr = EncryptMessage(byPlain); String sUUEncoded = UUCode(byEncr); return sUUEncoded; } public byte[] uuDecrypt(String sUUEncoded) { byte[] byEncr = UUCode(sUUEncoded); byte[] byPlain = DecryptMessage(byEncr); return byPlain; } public String Encrypt(String s) // Works with string too { return uuEncrypt(Encoding.ASCII.GetBytes(s)); } public String Decrypt(String sEncr) { return Encoding.ASCII.GetString(uuDecrypt(sEncr)); } String UUCode(byte[] byAry) { return Convert.ToBase64String(byAry); } byte[] UUCode(String sUU) { return Convert.FromBase64String(sUU); } #endregion UUEncode for String support #region Object Support public String EncryptObject(Object o) { return uuEncrypt(Obj2Bytes(o)); } public Object DecryptObject(String sEncr) { return Bytes2Obj(uuDecrypt(sEncr)); } // Support for Object Encryption/Decryption directly BinaryFormatter _bf; BinaryFormatter bf { get { if (_bf == null) _bf = new BinaryFormatter(); return _bf; } } public byte[] Obj2Bytes(Object o) { MemoryStream ms = new MemoryStream(); bf.Serialize(ms, o); int nObjSize = (int) ms.Length; byte[] byAry = new byte[nObjSize]; Buffer.BlockCopy(ms.GetBuffer(), 0, byAry, 0, nObjSize); return byAry; } public Object Bytes2Obj(byte[] byAry) { int nObjSize = byAry.Length; MemoryStream ms = new MemoryStream(byAry); Object o = bf.Deserialize(ms); return o; } #endregion Object Support }
  2. 21 May 2005 at 19:53

    I totally agree that varbinary could have been used.  My only problem with that solution is that it seems difficult to work with binary data in ADO.NET and SQL Server.  I may be wrong.  I haven't tried it, but I seem to remember seeing some nasty code dealing with storing images and other binary data in SQL Server.  The conversion to and from Base64 strings is 1 line of code in each case and then I can deal with regular char and varchar fields.

  3. 19 May 2005 at 17:26

    Couldn't you also store this in a Varbinary field instead of trying to convert it to a string for a varchar field?

  4. 18 May 2005 at 17:12

    Dear Steve,
    I was wondering if you had some insight as to how to best to file encrypting with the rijndael encryption.  Most examples I've found only give info. about how to encrypt messages.  I know you said that the CryptoStream object was beyond your scope, but that's where I'm needing to go with this and I like to use a FileStream instead of a MemoryStream.  I would also like to save the encrypted file in place of the original (which I may just have to rename them differently and then delete the original).  I also know you said that you don't have to write the encryption byte for byte, but how do you do this with a FileStream?  And if you do have to do it byte for byte with a FileStream, then how many bytes do I/should I set it to?  Also, all examples I've seen only demonstrate how to Encrypt a file and not Decrypt it again.  On the decryption, though, I only need to display it and not keep or save it anywhere.  Maybe loading it into a MemoryStream might be better?  Keep in mind that this files could be big (most are images).  Is there any advice you could give or maybe another discussion you have?


    Thanks!

  5. 24 Apr 2005 at 00:32

    Exactly.  If you don't use the same IV for encryption and decryption, the message can't be properly decoded.

  6. 23 Apr 2005 at 10:29

    Problem seems to be fixed now - I'd uncommented the initialisation vector code and it looks like it's mandetory.


    Also - replaced the hex encoding with base64 strings (many thanks to Ben Mills on the other thread about storing values in SQL server for pointing me in this direction).

  7. 23 Apr 2005 at 10:26

    Many thanks Ben - I've got a working solution now!


    I think the problem was the fact I'd commented out the initialisation vector code.. I didn't think this was essential.


    Anyhow - I've also removed my conversion routines to and from hex as the conversion to base64 strings is a lot more elegant (and the strings generated shorter).


    Thanks again.

  8. 22 Apr 2005 at 19:32

    I actually ended up cracking the problem in a different way.  Here's what I do to encode a .NET string:


    1.  Turn the string into a byte array using the ASCII encoder.
    2.  Run the byte array through the Rijndael encryptor (some bytes cannot now be represented as ASCII chars).
    3.  Convert the resulting array of bytes to a Base64 string using Convert.ToBase64String().
    4.  The result is an ASCII string which can be strored in regular (non-unicode) database columns.


    To decrypt the string I do the reverse:


    1.  Convert the string from the database into a byte array using Convert.FromBase64String().
    2.  Run the byte array through the Rijndael decryptor.
    3.  Convert the resulting array back into a string using the ASCII encoder.


    The trick with all of this is that the an 8 bit unicode string can be stored as a 7 bit ASCII string by converting it into a Base64 string.  Base 64 strings use 4 bytes to represent every block of 3 byte unicode characters.  See http://email.about.com/cs/standards/a/base64_encoding.htm for a good intro on base64 strings.  The one thing you have to think about is how long to make your database columns.  Just remember that every block of 3 unencrypted characters results in 4 encrypted characters in the database.


    I hope this helps,
    Ben Mills


  9. 22 Apr 2005 at 18:49

    Hi all,


    Excellent article - made a lot of sense.


    I've created a little test app which converts the byte arrays to hex strings (for storing in DB) - and then does the reverse. This all seems to work, but for some strange reason the first 32 characters of the decrypted string are garbled (everything after this point is fine!!).


    I'm guessing it's something to do with padding, but not sure - maybe I'm missing something really obvious!


    Any help would be hugely appreciated. I've posted the code from my test form below (hope you're okay about this and hope it benefits others)..


    cheers.



    using System;
    using System.Drawing;
    using System.Collections;
    using System.ComponentModel;
    using System.Windows.Forms;
    using System.Data;


    using System.Security.Cryptography;
    using System.Text;
    using System.IO;
    using System.Globalization;        // for NumberStyles


    namespace EncryptionTool
    {
       /// <summary>
       /// Summary description for Form1.
       /// </summary>
       public class Form1 : System.Windows.Forms.Form
       {
           private System.Windows.Forms.TextBox txtToEncrypt;
           private System.Windows.Forms.Label label1;
           private System.Windows.Forms.Button btnDecrypt;
           private System.Windows.Forms.Label label2;
           private System.Windows.Forms.TextBox txtKey;
           private System.Windows.Forms.Label label3;
           private System.Windows.Forms.Button btnEncrypt;
           private System.Windows.Forms.TextBox txtEncrypted;
           private System.Windows.Forms.TextBox txtDecrypted;
           private System.Windows.Forms.Label label4;
           /// <summary>
           /// Required designer variable.
           /// </summary>
           private System.ComponentModel.Container components = null;


           public Form1()
           {
               //
               // Required for Windows Form Designer support
               //
               InitializeComponent();


               //
               // TODO: Add any constructor code after InitializeComponent call
               //
           }


           /// <summary>
           /// Clean up any resources being used.
           /// </summary>
           protected override void Dispose( bool disposing )
           {
               if( disposing )
               {
                   if (components != null)
                   {
                       components.Dispose();
                   }
               }
               base.Dispose( disposing );
           }


           #region Windows Form Designer generated code
           /// <summary>
           /// Required method for Designer support - do not modify
           /// the contents of this method with the code editor.
           /// </summary>
           private void InitializeComponent()
           {
               this.txtToEncrypt = new System.Windows.Forms.TextBox();
               this.label1 = new System.Windows.Forms.Label();
               this.txtEncrypted = new System.Windows.Forms.TextBox();
               this.btnDecrypt = new System.Windows.Forms.Button();
               this.label2 = new System.Windows.Forms.Label();
               this.txtKey = new System.Windows.Forms.TextBox();
               this.label3 = new System.Windows.Forms.Label();
               this.btnEncrypt = new System.Windows.Forms.Button();
               this.txtDecrypted = new System.Windows.Forms.TextBox();
               this.label4 = new System.Windows.Forms.Label();
               this.SuspendLayout();
               //
               // txtToEncrypt
               //
               this.txtToEncrypt.Location = new System.Drawing.Point(8, 72);
               this.txtToEncrypt.Multiline = true;
               this.txtToEncrypt.Name = "txtToEncrypt";
               this.txtToEncrypt.Size = new System.Drawing.Size(448, 80);
               this.txtToEncrypt.TabIndex = 1;
               this.txtToEncrypt.Text = "";
               //
               // label1
               //
               this.label1.Location = new System.Drawing.Point(8, 56);
               this.label1.Name = "label1";
               this.label1.TabIndex = 2;
               this.label1.Text = "Text to encrypt";
               //
               // txtEncrypted
               //
               this.txtEncrypted.Location = new System.Drawing.Point(8, 176);
               this.txtEncrypted.Multiline = true;
               this.txtEncrypted.Name = "txtEncrypted";
               this.txtEncrypted.Size = new System.Drawing.Size(448, 80);
               this.txtEncrypted.TabIndex = 3;
               this.txtEncrypted.Text = "";
               //
               // btnDecrypt
               //
               this.btnDecrypt.Location = new System.Drawing.Point(464, 184);
               this.btnDecrypt.Name = "btnDecrypt";
               this.btnDecrypt.TabIndex = 4;
               this.btnDecrypt.Text = "&Decrypt";
               this.btnDecrypt.Click += new System.EventHandler(this.btnDecrypt_Click);
               // <

  10. 22 Apr 2005 at 18:42

    Hi,


    I'm attempting to do the same thing. I've gone down the route of converting the cipher's byte array to a Hex string using the following: -


       BitConverter.ToString(bytEncryptedMessage);


    This creates a hyphon delimitted string (e.g. 0A-10-FA etc...) which can be stored in the database.



    When I want to decrypt I'm first calling the following function: -


    // Assumes the hex string values are "-" delimitted. e.g. 0A-FA-BD
           public byte[] HexStringToBytes(string sHexString)
           {
               string[]        sStringArray;
               byte[]            bytByteArray;
               int                iCount = 0;


               sStringArray = sHexString.Split('-');
               bytByteArray = new byte[sStringArray.Length];
       
               foreach (string sHexValue in sStringArray)
               {
                   bytByteArray[iCount] = byte.Parse(sHexValue, NumberStyles.HexNumber);
                   iCount ++;
               }


               return bytByteArray;
           }


    which re-creates the byte array.




    I thought I'd cracked it, but something really weird is happening!! When decrypted - the first 32 characters are garbage - everything after is fine!!!


    I'm guessing this is something to do with the padding, but not sure.


    Hope this points you in the right direction and would also really appreciate any assistance with the problem I'm having!


  11. 30 Mar 2005 at 14:50

    First of all, I'd like to say that this was a fantastic article.  I was struggling through the MSDN info and other online articles and your article finally made everything clear.


    I want to use this technique to encrypt credit card data before I store it in SQL Server.  I take the ASCII credit card number, convert it to an array of bytes using the ASCII encoder and then encrypt the bytes.  The trouble is that the bytes are now out of the ASCII range (0 to 127), so I can't convert the bytes back to an ASCII string to store in a VarChar column of a SQL Server table.


    It seems like a solution is to convert the encrypted bytes to a unicode string using the unicode encoder and store that value in a NVarChar column.  The only problem with this is that I usually avoid unicode SQL Server columns.  Is this the only solution or do you have any other ideas?

  12. 07 Jan 2005 at 08:39

    Regarding my comment that "there is "no way" to reverse this process without the key."


    I want to be clear that I mean that there is no trivial way to do this.  What I'm trying to say is that the process is as secure as the key and the underlying cipher.


    There is, of course, no such thing as "no way" in cryptography.  We just try to use the best technology we can to make it as difficult and expensive as possible for an attacker to break the system.



    Steve Johnson

  13. 07 Jan 2005 at 08:23

    Using CBC as an example, encryption proceeds as follows:
    1 - XOR Plain Text Block #1 with IV
    2 - Encrypt the block resulting from step 1 with the key
    3 - The Cipher Text block resulting from step 3 is used as Cipher Text block #1
    4 - XOR Plain Text block #2 with Cipher Text block #1
    5 - Encrypt the block resulting from step 4 with the key
    6 - The Cipher Text block resulting from step 5 is used as Cipher Text block #2
    7 - etc, etc...


    There's no way to reverse this process without the key.  The IV is not being used to seed a PRNG; it's simply used as random data, with which to mix the first block of plain text on encryption and to retrieve the initial plain text block on decryption.  As you can see, there is no intermediate "non-randomized" state that can be determined using the cipher text and the IV.



    Steve Johnson

  14. 07 Jan 2005 at 08:05

    Wouldn't this mean that for eavesdroppers using the same random generator (which can not be that big a problem) and the transmitted seed it is possible to retrieve the 'un-randomized' (if thats a word) still encrypted content?


    Wilfred Kuijpers
    dbHost


  15. 07 Jan 2005 at 07:47

    You transmit the IV in the clear (unencrypted).  Just add the IV to your transmitted message so the receiving computer can decrypt the message.  There's no need to encrypt the IV because it is used simply to randomize the cipher text to prevent repetition.



    Steve Johnson

  16. 05 Jan 2005 at 17:08

    Hi,


    Great article,


    Just one question though, how do you transmit your IV's? I mean, decryction usually happens at another computer and for encryption and decryption you suggest to use the iv?


    regards,


    Wilfred Kuijpers
    dbhost

  17. 01 Jan 1999 at 00:00

    This thread is for discussions of Using Encryption in .NET.

Leave a comment

Sign in or Join us (it's free).

Steve Johnson Steve is a senior developer and consultant for 3t Systems, Inc. in Denver, CO, where he spends his days doing architecture, training, and "in-the-trenches" development. Steve began programming Wind...
AddThis

Related discussion

Related podcasts

  • More jQuery in ASP.NET

    In this episode Chris Brandsma, Rick Strahl, Dave Ward, Bertrand Le Roy, and Scott Koon conclude their discussion of Microsoft's jQuery in ASP.NET announcement1.This episode of the Alt.NET Podcast is brought to you by LLBLGen Pro, the most mature O/R mapper and code generator out there.Are ...

Events coming up

  • Nov 18

    15 Minutes of Fame

    Dresher, United States

    This is a yearly tradition. We select 10 of the favorite speakers from monthly meetings, code camps, and hands on labs. Each one does a 15 minute talk on their favorite .NET technology. This is our 10th anniversary so we plan a gala event with special prizes and refreshments.

We'd love to hear what you think! Submit ideas or give us feedback