输入数据:
1997-01-02 phone
1998-10-01 window
1997-01-02 phone
2001-11-23 xbox
2013-08-16 vr
1997-01-02 phone
2001-11-23 xbox
2013-08-16 vr
需求:去除其中的重复元素,每个日期对应的商品只保存一份
输出:
1997-01-02 phone
1998-10-01 window
2001-11-23 xbox
2013-08-16 vr
思路:
根据mapreduce特性,在reduce函数执行之前;会对相同key的数据经进行分组,将相同key的value放入一组(实际是一个集合)
分组之后每个key都是唯一的。即shuffle的过程,就可以利用key达到数据去重的效果。
这里将原数据不做任何处理,直接让其作为key原样输出,value输出类型为NullWritable;就能实现去重。
public class DistinctDataDemo {
public static class MyMapper extends Mapper<LongWritable,Text,Text,NullWritable>{
@Override
protected void map(LongWritable key, Text value,Context context)
throws IOException, InterruptedException {
//原数据作为key直接输出
context.write(value,NullWritable.get());
}
}
public static class MyReucer extends Reducer<Text,NullWritable,Text,NullWritable>{
@Override
protected void reduce(Text key, Iterable<NullWritable> value,
Context context) throws IOException, InterruptedException{
//原样输出
context.write(key,NullWritable.get());
}
}
public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException {
//获取配置对象
Configuration conf = new Configuration();
//获取FileSystem对象
FileSystem fs = FSUtil.getFS();
//创建作业对象
Job job = Job.getInstance(conf,"distinctdataDemo");
//设置运行主类
job.setJarByClass(DistinctDataDemo.class);
//设置mapper参数
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(NullWritable.class);
job.setMapperClass(MyMapper.class);
//设置输入文件路径
FileInputFormat.setInputPaths(job,new Path(args[0]));
//设置reducer参数
job.setReducerClass(MyReucer.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(NullWritable.class);
//设置输出文件路径
Path outPath = new Path(args[1]);
if(fs.exists(outPath)){
//目录已存在,则删除
fs.delete(outPath,true);
}
FileOutputFormat.setOutputPath(job,outPath);
//提交作业
boolean res = job.waitForCompletion(true);
System.exit(res ? 0 : -1);
}
}