MappReduce例子和总结02

1.计算每个人相应科目的平均分。

将同一个科目的放在一个文件中,并按照平均分从大到小排序。

部分数据文件:

MappReduce例子和总结02

1.创建对象并实现接口

public class NameScort implements WritableComparable {
    private String name;
    private String kemu;
    private Double score;
    @Override
    public int compareTo(Object o) {//按照成绩排序
        NameScort scort=(NameScort) o;
        if (scort.getScore()<this.score){
            return -1;
        }else {
            return 1;
        }
    }

    @Override
    public void write(DataOutput dataOutput) throws IOException {//读取的数据
        dataOutput.writeDouble(this.score);
        dataOutput.writeUTF(this.name);
        dataOutput.writeUTF(this.kemu);
    }

    @Override
    public void readFields(DataInput dataInput) throws IOException {//写出的数据
        this.score=dataInput.readDouble();
        this.name = dataInput.readUTF();
        this.kemu=dataInput.readUTF();
    }

2.自定义分区

public class ScoreParta extends Partitioner<NameScort, NullWritable> {//和Mapp的k,v类型一致
    @Override
    public int getPartition(NameScort text, NullWritable doubleWritable,int i) {
        if ("english".equals(text.getKemu())) {
            return 0;
        } else if ("math".equals(text.getKemu())) {
            return 1;
        } else if ("computer".equals(text.getKemu())) {
            return 2;
        } else if ("algorithm".equals(text.getKemu())) {
            return 3;
        }
        return 4;
    }
}

3.编写MappReduce方法

public class WorkTest02 {
    public static class w2Mapp extends Mapper<LongWritable, Text,NameScort, NullWritable>{
        @Override
        protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
            String[] split = value.toString().split("\n");
            String[] split1 = split[0].split(",");
//            String keys=split1[0]+" "+split1[1];
            int j=0;
            double sum=0;
            for (int i=2;i<split1.length;i++){
               sum+=Integer.parseInt(split1[i]);
               j++;
            }
            double avg=sum/j;
            NameScort scort=new NameScort();
            scort.setKemu(split1[0]);
            scort.setName(split1[1]);
            scort.setScore(avg);
            context.write(scort,NullWritable.get());
        }
    }
    public static class w2reduce extends Reducer<NameScort,NullWritable,NameScort,NullWritable>{
        @Override
        protected void reduce(NameScort key, Iterable<NullWritable> values, Context context) throws IOException, InterruptedException {
            context.write(key,NullWritable.get());
        }
    }
    public static void main(String[] args) throws Exception {
        Configuration conf=new Configuration();
        Job job = Job.getInstance(conf);
        job.setJarByClass(WorkTest02.class);
        job.setMapperClass(w2Mapp.class);
        job.setReducerClass(w2reduce.class);
        //设置map的k,v类型
        job.setMapOutputKeyClass(NameScort.class);
        job.setMapOutputValueClass(NullWritable.class);
        //设置reduce的k,v类型
        job.setOutputKeyClass(NameScort.class);
        job.setOutputValueClass(NullWritable.class);
        //设置分区数量
        job.setNumReduceTasks(4);
        //根据返回值去往指定的分区
        job.setPartitionerClass(ScoreParta.class);
        FileInputFormat.setInputPaths(job,"F:\\input\\score.txt");
        Path path=new Path("F:\\out6");
        FileSystem fs = FileSystem.get(conf);
        if (fs.exists(path)){
            fs.delete(path,true);
        }
        FileOutputFormat.setOutputPath(job,path);
        job.submit();
    }
}

4.部分结果文件

MappReduce例子和总结02

 

 

2.MapJoin:当有多个文件要一起处理时,可以先将小文件读取到内存中,在和大文件一块处理

数据文件格式:两个文件

MappReduce例子和总结02         MappReduce例子和总结02          

 

1.编写MapReducer方法

public class MappJoin {
    public static class mapp extends Mapper<LongWritable,Text,Text,NullWritable>{
        HashMap<String,String> usermap=new HashMap();
        @Override
        //先将小文件读取到内存中
        protected void setup(Context context) throws IOException, InterruptedException {
//            String line=new String(value.getBytes(),0,value.getLength(),"GBK");
            BufferedReader reader = new BufferedReader(new FileReader("F:\\input\\goods2.txt"));
            String line="";
            while ((line=reader.readLine())!=null){
                String[] split = line.split(",");
                usermap.put(split[0],split[1]);
            }
        }

        @Override
        protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
            String line=new String(value.getBytes(),0,value.getLength(),"GBK");
            String[] split = line.split(",");
            String s = usermap.get(split[0]);
//            String gbk = new String(s.getBytes(), "GBK");
            context.write(new Text(split[0]+","+split[1]+","+s), NullWritable.get());
        }

        public static void main(String[] args) throws Exception {
            Configuration conf=new Configuration();
            Job job = Job.getInstance(conf);
            job.setJarByClass(MappJoin.class);
            job.setMapperClass(mapp.class);
            job.setOutputKeyClass(Text.class);
            job.setOutputValueClass(NullWritable.class);
            job.setNumReduceTasks(0);
            FileInputFormat.setInputPaths(job,"F:\\input\\goods1.txt");
            Path path=new Path("F:\\out7");
            FileSystem fs = FileSystem.get(conf);
            if (fs.exists(path)){
                fs.delete(path,true);
            }
            FileOutputFormat.setOutputPath(job,path);
            job.submit();
        }
    }
}

 

2.处理结果文件

MappReduce例子和总结02

 

MappReduce例子和总结02

上一篇:iOS设计模式之工厂模式


下一篇:Spring中BeanFactory和ApplicationContext的区别