Hadoop Streaming： c++编写uniq程序

1. 描述
使用Hadoop streaming对输入数据的第一列进行uniq去重计算，采用c++编写map-reduce程序。

2. mapper程序
mapper.cpp

#include <iostream> #include <string> using namespace std; #define BUFFER_SIZE 102400 #define DELIM "\t" int main(int argc, char** argv) { char buf[BUFFER_SIZE]; char *strCur = NULL; //读取第一个字段并输出 while(fgets(buf, BUFFER_SIZE-1, stdin)) { strCur = strtok(buf, DELIM); cout << buf << endl; } return 0; }
3. reducer
reducer.cpp
#include <string> #include <set> #include <iostream> using namespace std; int main(int argc, char** argv) { string key; set<string> myset; while(cin >> key) { myset.insert(key); } cout << myset.size() << endl; return 0; }

4. 测试数据
test.txt

a 15 1 a 15 1 a 15 1 a 15 1 b 20 1 c 15 1 c 15 1 d 16 1 a 16 1
5. 本地测试
$ g++ -o mapper mapper.cpp $ g++ -o reducer reducer.cpp $ cat test.txt | ./mapper | ./reducer 4

6. Hadoop streaming 测试

$ hadoop fs -put test.txt /user/test.txt $ hadoop streaming -input /user/test.txt / -output /user/tmp_1324 / -mapper ./mapper -reducer ./reducer / -file mapper -file reducer / -jobconf mapred.reduce.tasks=1 / -jobconf mapre.job.name="uniq_test"

7. 查看结果

$ hadoop fs -cat /user/tmp_1324/part-00000 4