We are installing JDA v4r5 for a client in Israel. JDA doesn't currently have a Hebrew version which means we have to handle the right-to-left fields keyed by the users - especially when it comes to storing key values. Their current system runs on a HP using an Oracle db. It appears that the flipping is either handled by the os (doubtful) or the db. We're not sure which but we've been told that it is definitely not at the application level.

Does anyone know if the iSeries and DB2 UDB can handle this for us or do we need to modify every program that uses a Hebrew field as an index?

We developed a utility that identifies Hebrew text and flips the field before displaying / storing but we'd rather not have to modify program source if it isn't necessary.

Any assistance / suggestions is greatly appreciated.