in particular, my source files are dated CSVs. i want to have a pipeline that can essentially take a date as an argument and then do all the operations required for a given day, figuring out file names along the way. i'd like to then be able to use that to run, say, an entire year at once