Our initial implementation of the PAS caching will cache files on an individual basis. This puts the onus on the customer to determine which files they want to cache.
It seems likely that a customer's real world usage scenario would be "Here's a folder full of documents I want to cache" – in which case we can save them significant effort if we write this code on their behalf.
Otherwise customers will be forced to write significant code in order to achieve the same result.
Mark Fears
