I'm not sure what "WVARCHAR" is, but you need to make sure that the encoding being used is the same everywhere, e.g. Unicode in the shape of UTF-8. Note that this is a setting that usually needs to be made when the database is created.
If the encoding being used for some reason can't be the same everywhere, you need to convert between encodings in appropriate places.
Never rely on the platform default encoding - always be specific which one should be used.
I assumed that the data we get from the server supports 0-255 range and since java can support only (-127 to +127)
Rather than assuming anything, make sure what kind of encoding the DB uses. Java can support the full range of Unicode, so the remark about +/-127 doesn't make sense in this context.
[ February 27, 2008: Message edited by: Ulf Dittmer ]