How to specify charset when converting byte to string

jmusther

I have a working method to import a string from a UDP packet:

void receive( byte[] data ) { // handler String message = new String(data); println(message); }

While this works, I'd like to make it more robust by specifying the charset of the incoming data (which is ASCII).

Thanks in advance

GoToLoop

http://docs.oracle.com/javase/8/docs/api/java/lang/String.html#String-byte:A-java.nio.charset.Charset-
http://docs.oracle.com/javase/8/docs/api/java/nio/charset/Charset.html

jmusther

So shouldn't it be:

void receive( byte[] data ) { // handler String message = new String(data, Charset US-ASCII); println(message); }

But I get a syntax error...

GoToLoop

http://docs.Oracle.com/javase/8/docs/api/java/nio/charset/Charset.html#forName-java.lang.String-

jmusther

I'm afraid I still can't figure it out. Most of what I can find online suggests I don't need the 'Charset' there and instaed should have:

`String message = new String(data, US-ASCII);'

But this gives me an UnsupportedEncodingException

GoToLoop

Since I've never even thought about such rare usage, I haven't checked all of the overloaded constructors available for class String.
Actually there are 15 overloaded constructors for it from the link I've passed to ya:
http://docs.Oracle.com/javase/8/docs/api/java/lang/String.html#constructor.summary

And indeed 2 among them accept a String argument for the name of the chosen charset:

1. http://docs.Oracle.com/javase/8/docs/api/java/lang/String.html#String-byte:A-java.lang.String-
2. http://docs.Oracle.com/javase/8/docs/api/java/lang/String.html#String-byte:A-int-int-java.lang.String-

However, that approach demands a try...catch () block in order to deal w/ UnsupportedEncodingException:
https://Processing.org/reference/try.html
http://docs.oracle.com/javase/8/docs/api/java/io/UnsupportedEncodingException.html

There's no need for such if we use the Charset overload approach though! :-\"

In short, what you had was an "Unhandled exception type".
Which is how the Java compiler "hints" us that it "demands" a try...catch () block there: [-(

static final String receive(byte... data) {
  try {
    return new String(data, "US-ASCII");
  }
  catch (java.io.UnsupportedEncodingException e) {
    throw new RuntimeException(e);
  }
}

The same util function if you had used my original tip for Charset + forName(): ;;)

static final String receiveAscii(byte... data) {
  return new String(data, java.nio.charset.Charset.forName("US-ASCII"));
}

PhiLho

Surprising, you should have a syntax error instead, since US-ASCII isn't a valid Java identifier...

"Most of what I can find online"
They you find mostly outdated information... The constructor with Charset was added with Java 1.6. It is safer to use than the version with String, precisely to avoid this "UnsupportedEncodingException": you can put garbage in a string, but Charset offsets something more type-safe.