This is IBIS case 70131512.
Apparently in customer's lab testing everything went well, but in the field
they have found some keyCodes that don't match to unique raw codes.
Below is a snapshot of the results of pressing three different keys on
the system.
Windows :
KeyCode = 255 , ScanCode = 112
KeyCode = 255 , ScanCode = 125 (F14)
KeyCode = 255 , ScanCode = 114
Sun Java
KeyCode = 0 , Rawcode = 255
KeyCode = 0 , RawCode = 255
KeyCode = 0 , RawCode = 255
As we can see based on the above there is no way to distinguish the
three keys in the SunJVM.
The keys won't be found on a regular US 101 keyboard.
These are keys found on customer's 160 key Access keyboard.
However customer's point to Sun when they designed this rawcode workaround was
that regardless of keycode as there may be several that may be undefined
based on layout as long as there is a unique scancode passed by the OS it shouldbe uniquely represented in the JVM.
escalated to sustaining IBIS task 80137307