Converting byte[] to String and back c#

Question

I’m trying to convert a byte[] to a string and back using Encoding.Unicode. Sometimes Encoding.Unicode is able to convert the byte[] to a string and sometimes the output is != the input. What am I doing wrong?

Thanks for your help.

public static void Main(string[] args)
{
    Random rnd = new Random();
    while(true)
    {
        Int32 random = rnd.Next(10, 20);
        Byte[] inBytes = new Byte[random];
        for(int i = 0; i < random; i++)
            inBytes[i] = (Byte)rnd.Next(0, 9);

        String inBytesString = Encoding.Unicode.GetString(inBytes, 0, inBytes.Length);
        Byte[] outBytes = Encoding.Unicode.GetBytes(inBytesString);

        if(inBytes.Length != outBytes.Length)
            throw new Exception("?");
        else
        {
            for(int i = 0; i < inBytes.Length; i++)
            {
                if(inBytes[i] != outBytes[i])
                    throw new Exception("?");
            }
        }
        Console.WriteLine("OK");
    }
}

Well, some random byte arrays won't be valid unicode, so, sometimes this won't work. If you start with a random string and round trip it via a byte array, it will work every time. — Jodrell
– Jodrell, Commented Oct 9, 2012 at 13:33
The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!) — Tim Schmelter
– Tim Schmelter, Commented Oct 9, 2012 at 13:34

Marc Gravell · Accepted Answer · 2012-10-09 13:33:01Z

6

You cannot use Encoding for that: you must use something like Convert.ToBase64String / Convert.FromBase64String.

Encoding assumes the byte[] is formatted according to specific rules, which are not the case for a random non-string byte[].

To summarise:

An Encoding turns an arbitrary string to/from a formatted byte[]

Base-64 turns an arbitrary byte[] to/from a formatted string

answered Oct 9, 2012 at 13:33

Marc Gravell

1.1m273 gold badges2.6k silver badges3k bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Ravindra Bagale · Accepted Answer · 2012-10-09 13:40:03Z

0

you cannot use encoding use base64

using base64 u can safely convert bytes to a string and back

base64 guaranteed to not to get "invalid" unicode sequences like:
first half of a surrogate pair without the second half use like this:

string base64 = Convert.ToBase64String(bytes);
byte[] bytes = Convert.FromBase64String(base64);

edited Oct 9, 2012 at 13:40

answered Oct 9, 2012 at 13:34

Ravindra Bagale

17.7k9 gold badges46 silver badges70 bronze badges

1 Comment

Jodrell Over a year ago

strictly speaking, base64 is an encoding.

Andrew Brower · Accepted Answer · 2012-10-09 17:58:13Z

0

Here is an example where I changed and image to a bit array and then converted it back to a readable string.

protected bool isImageCMYK(HttpPostedFile image, Stream fileContent)
    {
            //creating byte array
        byte[] imageToByteArray = new byte[image.ContentLength];

            //filling the byte array
        fileContent.Read(imageToByteArray, 0 , image.ContentLength);

            //convering byte array back to a readable string
        UTF8Encoding byteToString = new UTF8Encoding();
        string imageString = byteToString.GetString(imageToByteArray);

        return imageString.ToLower().Contains("cmyk");
    }

here is the edited code which results in an output of "OK"

public static void Main(string[] args)
        {
            Random rnd = new Random();
            while (true)
            {
                Int32 random = rnd.Next(10, 20);
                Byte[] inBytes = new Byte[random];
                for (int i = 0; i < random; i++)
                    inBytes[i] = (Byte)rnd.Next(0, 9);

                UTF8Encoding inBytesString = new UTF8Encoding(); 
                string byteString = inBytesString.GetString(inBytes, 0, inBytes.Length);
                //Byte[] outBytes = Encoding.Unicode.GetBytes(inBytesString);
                Byte[] outBytes = inBytesString.GetBytes(byteString);

                if (inBytes.Length != outBytes.Length)
                    throw new Exception("?");
                else
                {
                    for (int i = 0; i < inBytes.Length; i++)
                    {
                        if (inBytes[i] != outBytes[i])
                            throw new Exception("?");
                    }
                }
                Console.WriteLine("OK");
            }

edited Oct 9, 2012 at 17:58

answered Oct 9, 2012 at 13:46

Andrew Brower

1,3672 gold badges9 silver badges10 bronze badges

5 Comments

Marc Gravell Over a year ago

If the incoming data is binary data (rather than UTF8 data), this is not a valid implementation. This is only correct if the data is UTF-8 text (or ASCII text).

Andrew Brower Over a year ago

Right so if you are using integers like above you should be fine right?

Marc Gravell Over a year ago

Integers how? Strings of integer values? integers encoded into binary? If so, what encoding? 4-byte LE? 4-byte BE? "varint"? Something else?

Andrew Brower Over a year ago

read it wrong but I edited my code using the example and received identical byte arrays

Marc Gravell Over a year ago

yes, but that is coincidence because the single-byte values 0-9 happen to all be valid in UTF-8. Now try round-tripping the three bytes (hex) ff ff ff. Or just a single (hex) ff.

Collectives™ on Stack Overflow

Converting byte[] to String and back c#

3 Answers 3

Comments

1 Comment

5 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

1 Comment

5 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related