spark-named-inputformat A simple hadoop inputformat that provides file information for each record. returns an rdd of tuples of type (file_name, file_line)