The aggregator will then only read data1 and data2 and do the matching on those (taking mime type into account, of course).
842a1f4b0b
c3b7e4e7