Question

I have a byte array that’s been initialized with 0xFF in each byte:

for (int i = 0; i < buffer.Length; i++)
{
    buffer[i] = 0xFF;
}

Once this byte array has been filled with valid data, I need to extract an ASCII string that’s stored at offset 192 and may be up to 32 characters in length. I’m doing this like so:

ASCIIEncoding enc = new ASCIIEncoding();
stringToRead = enc.GetString(buffer, 192, 32);

This works but I need to strip off the trailing bytes that contain 0xFF to avoid the string looking something like “John Smith??????????????????????”. Is there a function in .NET that provides this ability? Something like the String.TrimEnd() function perhaps or am I looking at a regex to do this?

Answer 1

I would suggest just finding out how long the string will really be:

int firstFF = Array.IndexOf(buffer, (byte) 0xff, 192);
if (firstFF == -1)
{
    firstFF = buffer.Length;
}
stringToRead = Encoding.ASCII(buffer, 192, firstFF - 192);

I would not try to give Encoding.ASCII bytes which aren’t valid ASCII-encoded text. I don’t know offhand what it would do with them – I suspect it would convert them to ? to show the error (as suggested by your existing output), but then you wouldn’t be able to tell the difference between that and real question marks. For example:

byte[] data = { 0x41, 0x42, 0x43, 0xff, 0xff };
string text = Encoding.ASCII.GetString(data);
Console.WriteLine(text.Contains((char) 0xff)); // False
Console.WriteLine(text.TrimEnd((char) 0xff).Length); // Still 5...

Now you could create an encoding which used some non-ASCII replacement character… but that’s a lot of hassle when you can just find where the binary data stops being valid.