python hadoop streaming二次排序

data

1	123
1	12
2	234
1	24
1	9

map.py

import sys


base_count = 10000
for line in sys.stdin:
	key,value = line.strip().split("\t")
	key = base_count + int(key)
	print("%s\t%s"%(str(key),value))

reduce.py

import sys
current_key = None
result = []
base_count = 10000
for line in sys.stdin:
	key,value = line.strip().split("\t")
	key = int(key) - base_count
	if current_key == None:
		current_key = key
	if current_key != key:
		result.sort()
		result = [str(i) for i in result]
		print("%s\t%s"%(str(current_key),",".join(result)))
		current_key = key
		result = []
	result.append(int(value))
result.sort()
result = [str(i) for i in result]
print("%s\t%s"%(str(current_key),",".join(result)))

result

1	9,12,24,123
2	234