Working with japanese and other char sets..

Question

Hey all,
I primarily use XML Pull parsing becuase of its speed and my lack of need to validate a document.
One problem we are having is saving text input by a user into an xml when it is japanese characters. We can clearly see the japanese characters in the text field.
I have tried setting the encoding to utf-8, utf-16 and ucs-2. When I set it to utf-16, the output gets all garbled and it cant be read. Actually, in fairness, I am using a basic file writer and building up a StringBuffer with the xml I want, where I put the declaration tag at the top in and write it all out. At least on the mac, when I open the xml with textedit, I see a slew of japanese characters but no valid xml. Somehow it turns the entire file into japanese and not just the values within quotes.
What's the trick in writing xml out and reading it back in so that various languages can be saved and restored?
Thank you.

Answer 1

I am using a basic file writer and building up a StringBuffer with the xml I want,That isn't the trick. The trick is to use a FileWriter that has been set up to use the same encoding you put into the XML document's prolog. Just having the XML document declare that it's encoded in UTF-8 doesn't make it be encoded in UTF-8, that is your responsibility. Like this:Writer w = new OutputStreamWriter(new FileOutputStream(...), "UTF-8"));And make sure you have a text editor that can understand UTF-8 and also render Japanese characters properly.