Unicode, UTF-8 and java servlet woes

Question

Hi,
I'm writing a content management system for a website about russian music.
One problem I'm having is trying to get a java servlet to talk Unicode to the Content mangament client.
The client makes a request for a band, the server then sends the XML to the client.
The XML reading works fine and the client displays the unicode fine from an XML file read locally (so the XMLReader class works fine).
The servlet unmarshals the request perfectly (its just a filename).
I then find the correct class, and pass it through the XML writer. that returns the XML as string, that I simply put into the output stream.
out.write(XMLWrite(selectedBand));I have set correct header property
response.setContentType("text/xml; charset=UTF-8");And to read it I
         //Make our URL
         URL url = new URL(pageURL);
         HttpURLConnection conn = (HttpURLConnection)url.openConnection();
         conn.setRequestMethod("POST");
         conn.setDoOutput(true); // want to send
         conn.setRequestProperty( "Content-type", "application/x-www-form-urlencoded" );
         conn.setRequestProperty( "Content-length", Integer.toString(request.length()));
         conn.setRequestProperty("Content-Language", "en-US"); 
         //Add our paramaters
         OutputStream ost = conn.getOutputStream();
         PrintWriter pw = new PrintWriter(ost);
         pw.print("myRequest=" + URLEncoder.encode(request, "UTF-8")); // here we "send" our body!
         pw.flush();
         pw.close();
         //Get the input stream
         InputStream ois = conn.getInputStream();
            InputStreamReader read = new InputStreamReader(ois);
         //Read
         int i;
         String s="";
         Log.Debug("XMLServerConnection", "Responce follows:");
         while((i = read.read()) != -1 ){
          System.out.print((char)i);
          s += (char)i;
         return s;now when I print
read.getEncoding()It claims:
ISO8859_1Somethings wrong there, so if I force it to accept UTF-8:
InputStreamReader read = new InputStreamReader(ois,"UTF-8");It now claims its
UTF8However all of the data has lost its unicode, any unicode character is replaced with a question mark character! This happens even when I don't force the input stream to be UTF-8
More so if I view the page in my browser, it does the same thing.
I've had a look around and I can't see a solution to this. Have I set something up wrong?
I've set, "-encoding utf8" as a compiler flag, but I don't think this would affect it.

Answer 1

I don't know what your problem is but I do have a couple of comments -
1) In conn.setRequestProperty( "Content-length", Integer.toString(request.length())); the length of your content is not request.length(). It is the length of th URL encoded data.
2) Why do you need to send URL encoded data? Why not just send the bytes.
3) If you send bytes then you can write straight to the OutputStream and you won't need to convert to characters to write to PrintWriter.
4) Since you are reading from the connection you need to setDoInput() to true.
5) You need to get the character encoding from the response so that you can specify the encoding in           InputStreamReader read = new InputStreamReader(ois, characterEncoding);
6) Reading a single char at a time from an InputStream is very inefficient.