2

Ignore the reasons why someone would want to do this.... :)

I want to be able to take some bytes, covert them to a string, then later back to the same byte array. Same length and everything.

I've tried using the ASCIIEncoder class (works for only text files) and Unicode Encoder class (only works so far for arrays 1024*n big. I assume this is because of the equal length of each character) with no success.

Is there any easy way to do this? I assume I should probably write my own functions to do so huh?

Improve this question

4 Answers 4

Reset to default
5

Use Base64 encoding. It's not space-efficient though.

string s = Convert.ToBase64String(byteArray);
byte[] decoded = Convert.FromBase64String(s);
Improve this answer
1
  • Thank you! Thank you! :D The solution I was trying to write myself was extremely slow...
    – bobber205
    Commented Dec 31, 2009 at 22:06
2

Base64 is great at problems like this.

string str = Convert.ToBase64String(inArray);
...
byte[] ourArray = Convert.FromBase64String(str);

Note that the resulting string will be larger (of course) than your original array.

Improve this answer
1

If you don't mind the doubling of space required to store each 8-bit byte into a 16-bit char, and you don't need the string to be composed of printable characters, you could do it the direct way:

string BytesToString(byte[] b)
{
    StringBuilder  s;    

    s = new StringBuilder(b.Length);
    for (int i = 0;  i < b.Length;  i++)
        s[i] = (char) b[i];    // Cast not really needed
    return s.ToString();
}    

byte[] StringToBytes(string s)
{
    byte[]  b;    

    b = new byte[s.Length];
    for (int i = 0;  i < b.Length;  i++)
        b[i] = (byte) s[i];
    return b;
}

You could write more elaborate code to pack two 8-bit bytes into each 16-bit Unicode char, but then you'd have to handle the special cases of odd byte array lengths (which could be done using an extra sentinel byte in the last character).

Improve this answer
2
  • I tried this method using just regular strings and the + operator. Was mad slow. Is that because I didn't use a StringBuilder (like I usually do)?
    – bobber205
    Commented Dec 31, 2009 at 23:30
  • Yes. Typically, each + operator involves the overhad of creating a new temporary StringBuilder object, initializing it with the partially built string, appending the next character, which typically causes the internal buffer array in StringBuilder to be reallocated and copied (again). And this happens for each loop iteration. So, yeah, it's generally a very slow way to do it. Using an explicit StringBuilder having a sufficiently large buffer allocated to it eliminates all of that overhead.
    – David R Tribble
    Commented Jan 1, 2010 at 6:56
1

If you know the encoding of your string (I would guess it is UTF8) then simply use the encoding class:

Encoding.UTF8.GetBytes
Encoding.UTF8.GetString

Encoding class can be found in the System.Text namespace

Y.

Improve this answer
1
  • The byte[] array is being read from a file so I don't know its encoding.
    – bobber205
    Commented Dec 31, 2009 at 22:03

Not the answer you're looking for? Browse other questions tagged or ask your own question.