为什么这两个字符串不等于？

Question

I am sending a packet through UDP and for some reason I can't compare the string I extract from the packet and the string I create even though the values are the same when I print them (no trailing white spaces).

byte[] incoming = new byte[1000];
DatagramPacket request = new DatagramPacket(incoming, incoming.length);
serverSocket.receive(request);
String str = new String(request.getData());
String str2 = new String("message received");

if(str.equals(str2))
{
   System.out.println("equal");
}

Is there any reason for this?

Answer 1

This occurs because new String(request.getData()) does not return "message received" .

The problem is [likely] due to the fact that new String(byte[]) attempts to use all (1000 of) the bytes supplied, in the default encoding, which ends with a bunch of NUL ('\\0') characters that append to the actual string content making it not equal with the literal. Such can be easily seen a debugger, although such NUL characters are often "lost" when displaying as normal text as with println .

Trivially: "hello".equals("hello\\0") is false.

Several solutions include:

1. Frame the string, such as prefixing the sent data with the number of bytes that make up the string, and then using a String constructor that takes a limit/length or;

2. Prevent any trailing 0 from being processed, again by specifiying the limit to decode or;

3. Remove any NUL characters after decoding the data.

Since option #3 is easy ¹ (until it can be fixed to use #1/#2), consider:

String str = new String(request.getData(), "UTF-8"); // Specify an encoding!
int nul = str.indexOf('\0');
if (nul > -1) {
   str = str.substring(0, nul);
}

---

¹ While trimming is the easiest, it is not generally appropriate. The biggest problem with #3 over #2 is it first decodes all the bytes and then filters the characters. Under different encodings (although ASCII and UTF-8 should be "safe"), this may result in non-NUL garbage after the actual string content depending upon what exists in the buffer.

Also, specify an encoding manually to new String(byte[] ..) or String.getBytes(..) . Otherwise the "default encoding" will be used, which can cause problems if the different systems are using a different default.