On 5/14/05, Leon Rosenberg <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I have a small encoding problem, which drives me crazy...
>
> Our complete site is in ISO-8859-1 (which is java-default, as I understand
> it). I mean, the charset of the page is ISO, and meta-tags in HTML are
> telling the
> browser that the page is ISO too.
> Now the problem, that I have, is that I have to transmit some XML data to
> another system (payment provider) which expects it in UTF-8.
> The problem is that customer name can contain Umlauts (german characters:
> äöü), and they come truncated on the other side:
>
> Like I'm sending "Ümlaut" and the other side gets �mlaut.
>
> I tried each conversion method I could think of sofar:
> reinitializing the String as new String with reencoding: name = new
> String(name.getBytes("ISO-8859-1"), "UTF-8") (in all combinations)
> Using URLDecoder to decode parameters.
> Using charsetencoded Writer (OutputStreamWriter writer = new
> OutputStreamWriter(outStream, "UTF-8")).
> and so on...
>
> Can anyone give me a hint?
>
> This problem is slowly driving me crazy....
>
> regards
> and thanx in advance
>
> Leon
>
>
I recently needed to send some text encoded as UTF-8 over a TCP/IP
socket. I did it like this and it worked:
import java.nio.*; //for CharBuffer (and ByteBuffer)
import java.nio.charset.*; //For Charset and CharsetEncoder
OutputStream os = .. /* initialized somhow*/
String s = "some text to send"; //s can be anything that extends CharSequence
CharsetEncoder ce = Charset.forName("UTF-8").newEncoder();
byte[] b = ce.encode(CharBuffer.wrap(s)).array();
os.write(b, 0, b.length);
In another (production) project I used similar code - with
SocketChannels because they have send methods for sending CharBuffers.
Java Strings and chars are Unicode (size of type char is 2 bytes, 0 - 65535).
--
Why?
Because YES!
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
