There are many duplicated functions in the dataset parsers. We should pull those out into a usable library.