Encoding problem (Umlaute) in HTTP-Headers

Question

Hi!
I have a servlet that is being called by another server via a POST request.
The request contains user data in HTTP-headers, and whenever there is a
German Umlaut in the user's name, it ends up garbled. :(
The requests character encoding is set to 'ISO-8859-1', according to the servlet
requests getCharacterEncoding() method.
I try to copy the value of a header variable into a String like this:
headerNameValue = request.getHeader("headerNameKey");Which, unfortunatly, doesn't work. For example, the value "G�nter" becomes "G����nter".
Shouldn't the API take care of the encoding, when the CharacterEncoding is set
right in the servlet call?
It looks like it interprets the ISO88591 data as Unicode?
BTW, I'm on a WebSphere 5 server, with Java 1.4.
I've been tearing my hair out on this one for days, so any help is highly appreciated.
Thanks a lot!
Andreas

Answer 1

A header variable with the key 'accept-charset' is not sent.
Only 'accept-encoding', with the value 'gzip, deflate'.
Like I wrote in my first post, the servlet request's character encoding is set to ISO-8859-1,
according to the getCharacterEncoding-method.
I have kind of solved this for now by converting the String 2times back, like this:
try {
     headerNameValue = new String((new String(headerNameValue.getBytes(),"UTF-8")).getBytes(),"UTF-8");
} catch (UnsupportedEncodingException e) {
      logger.error("Fehler beim Umwandeln der Umlaute im Benutzernamen.",e);
}                    }This code converts the '����', that my servlet receives, back to '�'.
(The first conversion changes the '����' to '��', the second the remaining '��' to '�').
While this seems to work, I don't think it is the right solution, and I still don't understand what's going on. :)
Regards,
Andreas