Every Artifact (binary or text based) should be stored uniquely on federated repository (even that artifact is spread from across several paths).

With this uniquely restriction based on artifact content (checked by SHA1) we also need to record the paths (from data source and virtual federated) of each artifact - its necessary 'cos the same artifact can be (and usually is) spread thru data sources.

Importante Note: Although these artifacts are uniquely stored they need to be processed based on path, witch means that it can be processed more than once.