Houston, we have a problem: There is a TCK test in ContextTest which checks for this spec part: > 9.1. The Context interface > At a particular point in the execution of the program a scope may > be inactive with respect to the current thread. When a > scope is inactive, any invocation of the get() from the current > thread upon the Context object for that scope results in a > ContextNotActiveException.
But we currently use a completely different logic for handling the Context as RI does. So we do 'activate' the Context 'on invocation' if I read this correctly. see ContextTest#testGetContextWithNoActiveContextsFails() LieGrue, strub
