j2me midlet chinese character display message garbled - utf-8

Question

My J2ME midlet could retrieves message in Chinese character from a PHP server but it's garbled. The server basically returns the response string and by detecting the first 2 characters. AA = good, anything else indicates error of which the message is to be passed to the calling function for display
InputStream is = null;
StringBuffer sb = null;
String str = "";
HttpConnection http = null;
DataOutputStream dos = null;
try
{
URL = login.getURL();
URL += ctlFunction + "/" + uriStr;
URL = EncodeURL(URL);
//System.out.println(URL);
if(!ctlFunction.equals("login"))
{
msg += "&user=" + login.getUsername();
msg += "&token=" + login.getToken();
}
msg += "&lang=" + System.getProperty("microedition.locale");
// establish the connection
http = (HttpConnection) Connector.open(URL);
http.setRequestMethod(HttpConnection.POST);
http.setRequestProperty("Content-Type", "application/x-www-form-urlencoded");
http.setRequestProperty("Content-length", ""+EncodeURL(msg).getBytes().length);
dos = http.openDataOutputStream();
byte[] request_body = EncodeURL(msg).getBytes();
for (int i = 0; i < request_body.length; i++)
{
dos.writeByte(request_body[i]);
}
// server response
if (http.getResponseCode() == HttpConnection.HTTP_OK)
{
is = http.openDataInputStream();
int length = (int) http.getLength();
if (length != -1)
{
// Read data in one chunk
byte serverData[] = new byte[length];
is.read(serverData);
str = new String(serverData);
}
else // Length not available...
{
ByteArrayOutputStream bStrm = new ByteArrayOutputStream();
int ch;
while ((ch = is.read()) != -1)
bStrm.write(ch);
str = new String(bStrm.toByteArray());
bStrm.close();
}
}
else
{
networkError();
}
}
catch (Exception e)
{
System.err.println("Error3: " + e.toString());
networkError(e.toString());
}
finally
{
if (is != null)
is.close();
if (!str.equals(""))
post = str;
else
networkError();
if (http != null)
http.close();
}
if (post != null)
{
String fate = post.substring(0, 2);
if(fate.equals("AA"))
{
if(ctlFunction.equals("login"))
{
String rawPost = post.substring(2);
Vector v = new Vector();
int index = 0;
//find the first occurrence of the SPLITTER
int endIndex = rawPost.indexOf(SPLITTER, index);
String item = "";
//extract the items until the end of the last SPLITTER found in the rawPost string
while(endIndex != -1)
{
item = rawPost.substring(index, endIndex);
index = endIndex + 1;
endIndex = rawPost.indexOf(SPLITTER, index);
v.addElement(item);
}
//extract the rest of the rawPost (the text item)
item = rawPost.substring(index);
v.addElement(item);
String[] ret = new String[v.size()];
v.copyInto(ret);
login.setToken(ret[0]);
login.setToday(ret[1]);
login.setNextDrawDay(ret[2]);
}
midlet.returnResults(post.substring(2), getCurrentDisplay(), ctlFunction);
}
else
{
String errmessage = post.substring(2);
System.out.println(post);
midlet.showInfo(post, getCurrentDisplay());
}
}
else
{
networkError();
}
On the PHP server, I have set the header to UTF-8 encoding
<?php header("Content-Type:text/plain; charset=utf-8"); ?>
What could possibly be wrong?

Answer 1

I found that this user has the same problem and it's been answered
Reading UTF8 strings from a server through http using MIDP. Kudos to the answer.
I basically edited my MIDP code from
// is = http.openDataInputStream();
// int length = (int) http.getLength();
// if (length != -1)
// {
// // Read data in one chunk
// byte serverData[] = new byte[length];
// is.read(serverData);
// str = new String(serverData);
// }
// else // Length not available...
// {
// ByteArrayOutputStream bStrm = new ByteArrayOutputStream();
// int ch;
// while ((ch = is.read()) != -1)
// bStrm.write(ch);
//
// str = new String(bStrm.toByteArray());
// bStrm.close();
// }
to
Reader r = new InputStreamReader(http.openInputStream(), "UTF-8");
int ch;
while((ch = r.read()) != -1)
str = str + (char)ch;
just wondering though why does reading bytes messes up the UTF-8 characters?