-- records_orders.pig -- by Joe Hahn, jhahn@spacescience.org, 7 February 2014. -- -- This Pig script reads the records and orders csv files, joins them on record_id, -- groups by city,state and then sums the $ amounts of all orders at each locale, -- and writes the results to file. -- -- To execute in local mode > pig -x local records_orders.pig records = load 'records.csv' using PigStorage(',') as (index, record_id, city, state); orders = load 'orders.csv' using PigStorage(',') as (index, order_id, record_id, amount); rec_ord = join records by record_id, orders by record_id; rec_ord_grouped = group rec_ord by (city, state); amount_s = foreach rec_ord_grouped generate group, SUM(rec_ord.amount) as amount; amount_so = order amount_s by amount desc; store amount_so into 'amount';