mapperMaxCost.py

#! /usr/bin/env python

import sys

for line in sys.stdin:
  data = line.strip().split("\t")
  if len(data) != 6:
    continue
  date, time, store, item, cost, payment = data
  print "{0}\t{1}".format(store, cost)

mapperSales.py

#! /usr/bin/env python

import sys

for line in sys.stdin:
  data = line.strip().split("\t")
  if len(data) != 6:
    continue
  date, time, store, item, cost, payment = data
  print "{0}\t{1}".format(store, cost)

reducerCU.py

#! /usr/bin/env python

import sys

sales_total = 0
old_key = None

for line in sys.stdin:
  data = line.strip().split("\t")
  if len(data) != 2:
    continue
  key, sales = data
  if (old_key and key != old_key):
    print "{0}\t{1}".format(old_key, sales_total)
    sales_total = 0
  old_key = key
  sales_total += float(sales)
print "{0}\t{1}".format(old_key, sales_total)

reducerMaxCost.py

#! /usr/bin/env python

import sys

max_cost = 0
old_key = None

for line in sys.stdin:
  data = line.strip().split("\t")
  if len(data) != 2:
    continue
  store, cost = data
  if (old_key and key != old_key):
    print "{0}\t{1}".format(old_key, sales_total)
    max_cost = 0
  old_key = key
  max_cost = max(cost, max_cost)
print "{0}\t{1}".format(old_key, max_cost)

reducerSales.py

#! /usr/bin/env python

import sys

sales_total = 0
sales_num = 0
old_key = None

for line in sys.stdin:
  data = line.strip().split("\t")
  if len(data) != 2:
    continue
  key, sales = data
  if (old_key and key != old_key):
    print "{0}\t{1}".format(old_key, sales_total)
    sales_total = 0
  old_key = key
  sales_total += float(sales)
  sales_num += 1
print "{0}\t{1}\t{1}".format(old_key, sales_total, sales_num)