can my idea somehow be built utilizing the mechanisms of those [upgraded Library] processes?
But come to think of it: if Library back then was using a temporary file - then why cannot there be such temporary file created when a search is done? And then used quickly for an immediate reverse search?