Hadoop：使用原生python编写MapReduce

foo foo quux labs foo bar quux abc bar see you by test welcome test
abc labs foo me python hadoop ab ac bc bec python

#!/usr/bin/env python
# -*- coding:utf-8 -*-

import sys

#输入为标准输入stdin
for line in sys.stdin:
#删除开头和结尾的空行
line = line.strip()
#以默认空格分隔单词到words列表
words = line.split()
for word in words:
#输出所有单词，格式为“单词,1”以便作为Reduce的输入
print '%s\t%s' % (word,1)0

#!/usr/bin/env python
# -*- coding:utf-8 -*-

from operator import itemgetter
import sys

current_word = None
current_count = 0
word = None

#获取标准输入，即mapper.py的标准输出
for line in sys.stdin:
#删除开头和结尾的空行
line = line.strip()

#解析mapper.py输出作为程序的输入，以tab作为分隔符
word,count = line.split('\t',1)

#转换count从字符型到整型
try:
count = int(count)
except ValueError:
#count非数字时，忽略此行
continue

#要求mapper.py的输出做排序（sort）操作，以便对连续的word做判断
if current_word == word:
current_count += count
else:
if current_word:
#输出当前word统计结果到标准输出
print '%s\t%s' % (current_word,current_count)
current_count = count
current_word = word

#输出最后一个word统计
if current_word == word:
print '%s\t%s' % (current_word,current_count)

[root@wx ~]# cd /root/hadooptest/
[root@wx hadooptest]# cat input.txt | ./mapper.py 
foo 1
foo 1
quux 1
labs 1
foo 1
bar 1
quux 1
abc 1
bar 1
see 1
you 1
by 1
test 1
welcome 1
test 1
abc 1
labs 1
foo 1
me 1
python 1
hadoop 1
ab 1
ac 1
bc 1
bec 1
python 1


[root@wx hadooptest]# cat input.txt | ./mapper.py | sort -k1,1 | ./reducer.py 
ab 1
abc 2
ac 1
bar 2
bc 1
bec 1
by 1
foo 4
hadoop 1
labs 2
me 1
python 2
quux 2
see 1
test 2
welcome 1
you 1

/usr/local/hadoop-2.6.4/bin/hadoop fs -mkdir -p /user/root/word

/usr/local/hadoop-2.6.4/bin/hadoop fs -put /root/hadooptest/input.txt /user/root/word

/usr/local/hadoop-2.6.4/bin/hadoop fs -ls /user/root/word
#结果：
Found 1 items
-rw-r--r-- 2 root supergroup 118 2016-03-22 13:36 /user/root/word/input.txt

/usr/local/hadoop-2.6.4/bin/hadoop jar /usr/local/hadoop-2.6.4/share/hadoop/tools/lib/hadoop-streaming-2.6.4.jar -files 'mapper.py,reducer.py' -input /user/root/word -output /output/word -mapper ./mapper.py -reducer ./reducer.py

/usr/local/hadoop-2.6.4/bin/hadoop jar /usr/local/hadoop-2.6.4/share/hadoop/tools/lib/hadoop-streaming-2.6.4.jar \
-input <输入目录> \ # 可以指定多个输入路径，例如：-input '/user/foo/dir1' -input '/user/foo/dir2'
-inputformat <输入格式 JavaClassName> \
-output <输出目录> \
-outputformat <输出格式 JavaClassName> \
-mapper <mapper executable or JavaClassName> \
-reducer <reducer executable or JavaClassName> \
-combiner <combiner executable or JavaClassName> \
-partitioner <JavaClassName> \
-cmdenv <name=value> \ # 可以传递环境变量，可以当作参数传入到任务中，可以配置多个
-file <依赖的文件> \ # 配置文件，字典等依赖
-D <name=value> \ # 作业的属性配置

[root@wx hadooptest]# /usr/local/hadoop-2.6.4/bin/hadoop fs -ls /output/word
Found 2 items
-rw-r--r-- 2 root supergroup 0 2016-03-22 13:47 /output/word/_SUCCESS
-rw-r--r-- 2 root supergroup 110 2016-03-22 13:47 /output/word/part-00000

[root@wx hadooptest]# /usr/local/hadoop-2.6.4/bin/hadoop fs -cat /output/word/part-00000
ab 1
abc 2
ac 1
bar 2
bc 1
bec 1
by 1
foo 4
hadoop 1
labs 2
me 1
python 2
quux 2
see 1
test 2
welcome 1
you 1