Every time you add a new dataset, do you have to run the parser again all the way to the genesis block?