Hadoop的MR编程实现partition、sort和自定义outputformat

Partitioner

public abstract class Partitioner<KEY, VALUE>

(key.hashCode() & Integer.MAX_VALUE) % numReduceTasks;

key.hashCode()% numReduceTasks;

按照我们的需求：3分区
13*
15*
其他*

reducer = 分区数 1
reducer > 分区数 产生很多空的没用的文件
200 200 199
reducer < 分区数

自定义分区器

package com.ccj.pxj.phone.partition;
import org.apache.hadoop.io.Writable;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
public class Access implements Writable {
    private  String phone;
    private  long up;
    private  long down;
    private  long sum;
    public String getPhone() {
        return phone;
    }
    public void setPhone(String phone) {
        this.phone = phone;
    }
    public long getUp() {
        return up;
    }
    public void setUp(long up) {
        this.up = up;
    }
    public long getDown() {
        return down;
    }
    public void setDown(long down) {
        this.down = down;
    }
    public Access(String phone, long up, long down) {
        this.phone = phone;
        this.up = up;
        this.down = down;
        this.sum=up+down;
    }
    public Access() {
    }
    @Override
    public void write(DataOutput out) throws IOException {
            out.writeUTF(phone);
            out.writeLong(up);
            out.writeLong(down);
            out.writeLong(sum);
    }
    @Override
    public void readFields(DataInput in) throws IOException {
         this.phone= in.readUTF();
         this.up= in.readLong();
         this.down=in.readLong();
         this.sum=in.readLong();
    }
    @Override
    public String toString() {
        return
                phone + '\t' +
                 up +
                "\t" + down +
                "\t" + sum ;
    }
}
package com.ccj.pxj.phone.partition;
import com.ccj.pxj.phone.utils.FileUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import java.io.IOException;
public class SerDriver {
    public static void main(String[] args) throws Exception {
        String input = "data/phone_data .txt";
        String output = "out";
        // 1）获取Job对象
        Configuration configuration = new Configuration();
        Job job = Job.getInstance(configuration);
        FileUtils.deleteOutput(configuration, output);
        // 2）本job对应要执行的主类是哪个
        job.setJarByClass(SerDriver.class);
        // 3）设置Mapper和Reducer
        job.setMapperClass(MyMaper .class);
        job.setReducerClass(MyReduce.class);
        // 4）设置Mapper阶段输出数据的类型
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(Access.class);
//        设置自定义分区器
        job.setPartitionerClass(PxjPartition.class);
//        设置reduce个数=>决定最终文件的个数
       job.setNumReduceTasks(3);
        // 5）设置Reducer阶段输出数据的类型
        job.setOutputKeyClass(NullWritable.class);
        job.setOutputValueClass(Access.class);
        // 6）设置输入和输出路径
        FileInputFormat.setInputPaths(job, new Path(input));
        FileOutputFormat.setOutputPath(job, new Path(output));
        // 7）提交作业
        boolean result = job.waitForCompletion(true);
        System.exit(result ? 0 : 1);
    }
   public static class  MyMaper extends Mapper<LongWritable, Text,Text, Access>{
       @Override
       protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
           String[] data = value.toString().split("\t");
           String phone = data[1];
           // 上行流量
           long up = Long.parseLong(data[data.length - 3]);
           // 下行流量
           long down = Long.parseLong(data[data.length - 2]);
           context.write(new Text(phone),new Access(phone,up,down));
       }
   }
   public  static  class MyReduce extends Reducer<Text, Access, NullWritable, Access>{
       @Override
       protected void reduce(Text key, Iterable<Access> values, Context context) throws IOException, InterruptedException {
            long ups=0;
            long downs=0;
           for (Access value : values) {
               ups+=value.getUp();
               downs+=value.getDown();
           }
           context.write(NullWritable.get(),new Access(key.toString(),ups,downs));
       }
   }
}
package com.ccj.pxj.phone.partition;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Partitioner;
public class PxjPartition  extends Partitioner<Text,Access> {
    @Override
    public int getPartition(Text text, Access access, int numPartition) {
        String phone = text.toString();
       /* if("13".equals(phone)){
            return  0;
        }else if("15".equals(phone)){
            return 1;
        }else {
            return 2;
        }
    }*/
       if(phone.startsWith("13")){
           return 0;
       }else if(phone.startsWith("15")){
           return 1;
       }else{
           return 2;
       }
    }
}

说明：
reduce个数=>决定最终文件的个数
reducer = 分区数 1
reducer > 分区数 产生很多空的没用的文件
200 200 199
reducer < 分区数 报错

Combiner

map: MapTask
reduce：ReduceTask
combiner：MapTask
父类是Reducer
对于每个MapTask的输出进行局部汇总/本地聚合
业务逻辑和Reducer是一模一样的

没使用 combiner    
Combine input records=0
        Combine output records=0
使用后
                Combine input records=12
        Combine output records=7

package com.ccj.pxj.commine.wc;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
import java.io.IOException;
public class WCMapper extends Mapper<LongWritable,Text,Text,IntWritable> {
    IntWritable ONE = new IntWritable(1);
    @Override
    protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
        String line = value.toString();
        String[] words = line.split(",");
        for (String word : words) {
            context.write(new Text(word),ONE);
        }
    }
}
package com.ccj.pxj.commine.wc;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;
import java.io.IOException;
public class WCReducer extends Reducer<Text, IntWritable, Text, IntWritable> {
    @Override
    protected void reduce(Text key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException {
        int sum=0;
        for (IntWritable value : values) {
             sum+=value.get();
        }
        context.write(key,new IntWritable(sum));
    }
}
package com.ccj.pxj.commine.wc;
import com.ccj.pxj.phone.utils.FileUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
public class WCDriver {
    public static void main(String[] args)throws  Exception {
        String input = "data/1.txt";
        String output = "out1";
        // 1）获取Job对象
        Configuration configuration = new Configuration();
        Job job = Job.getInstance(configuration);
        FileUtils.deleteOutput(configuration, output);
        // 2）本job对应要执行的主类是哪个
        job.setJarByClass(WCDriver.class);
        // 3）设置Mapper和Reducer
        job.setMapperClass(WCMapper.class);
        job.setReducerClass(WCReducer.class);
        // 4）设置Mapper阶段输出数据的类型
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(IntWritable.class);
        //设置Combiner
        job.setCombinerClass(WCReducer.class);
        // 5）设置Reducer阶段输出数据的类型
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(IntWritable.class);
        //
        // 6）设置输入和输出路径
        FileInputFormat.setInputPaths(job, new Path(input));
        FileOutputFormat.setOutputPath(job, new Path(output));
        // 7）提交作业
        boolean result = job.waitForCompletion(true);
        System.exit(result ? 0 : 1);
    }
}

排序

在MapReduce中，内置的这些数据类型不仅具有Writable的功能
还具有排序的功能
WritableComparable得实现三个方法
    write
    readFields
    compareTo

全排： order by
分区排序：sort by

package com.ccj.pxj.sort;
import org.apache.hadoop.io.WritableComparable;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
/**
 * 自定义序列化类且排序
 */
public class Traffic implements WritableComparable<Traffic> {
    private  String phone;
    private  long up;
    private  long down;
    private  long sum;
    public String getPhone() {
        return phone;
    }
    public void setPhone(String phone) {
        this.phone = phone;
    }
    public long getUp() {
        return up;
    }
    public void setUp(long up) {
        this.up = up;
    }
    public long getDown() {
        return down;
    }
    public void setDown(long down) {
        this.down = down;
    }
    public Traffic(String phone, long up, long down) {
        this.phone = phone;
        this.up = up;
        this.down = down;
        this.sum=up+down;
    }
    public Traffic() {
    }
    @Override
    public void write(DataOutput out) throws IOException {
            out.writeUTF(phone);
            out.writeLong(up);
            out.writeLong(down);
            out.writeLong(sum);
    }
    @Override
    public void readFields(DataInput in) throws IOException {
         this.phone= in.readUTF();
         this.up= in.readLong();
         this.down=in.readLong();
         this.sum=in.readLong();
    }
    @Override
    public String toString() {
        return
                phone + '\t' +
                 up +
                "\t" + down +
                "\t" + sum ;
    }
    @Override
    public int compareTo(Traffic o) {
        return this.sum>o.sum?-1:1;
    }
}
package com.ccj.pxj.sort;
import com.ccj.pxj.phone.utils.FileUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import java.io.IOException;
public class AllSortDriver {
    public static void main(String[] args) throws Exception {
        String input = "data/phone_data .txt";
        String output = "out";
        // 1）获取Job对象
        Configuration configuration = new Configuration();
        Job job = Job.getInstance(configuration);
        FileUtils.deleteOutput(configuration, output);
        // 2）本job对应要执行的主类是哪个
        job.setJarByClass(AllSortDriver.class);
        // 3）设置Mapper和Reducer
        job.setMapperClass(MyMaper.class);
        job.setReducerClass(MyReduce.class);
        // 4）设置Mapper阶段输出数据的类型
        job.setMapOutputKeyClass(Traffic.class);
        job.setMapOutputValueClass(Text.class);
        // 5）设置Reducer阶段输出数据的类型
        job.setOutputKeyClass(Traffic.class);
        job.setOutputValueClass(Text.class);
        // 6）设置输入和输出路径
        FileInputFormat.setInputPaths(job, new Path(input));
        FileOutputFormat.setOutputPath(job, new Path(output));
        // 7）提交作业
        boolean result = job.waitForCompletion(true);
        System.exit(result ? 0 : 1);
    }
   public static class  MyMaper extends Mapper<LongWritable, Text,Traffic, Text>{
       @Override
       protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
           String[] data = value.toString().split("\t");
           String phone = data[1];
           // 上行流量
           long up = Long.parseLong(data[data.length - 3]);
           // 下行流量
           long down = Long.parseLong(data[data.length - 2]);
           context.write(new Traffic(phone,up,down),new Text(phone));
       }
   }
   public  static  class MyReduce extends Reducer< Traffic,Text, Text, Traffic>{
      /* @Override
       protected void reduce(Text key, Iterable<Traffic> values, Context context) throws IOException, InterruptedException {
            long ups=0;
            long downs=0;
           for (Traffic value : values) {
               ups+=value.getUp();
               downs+=value.getDown();
           }
           context.write(NullWritable.get(),new Traffic(key.toString(),ups,downs));
       }*/
       @Override
       protected void reduce(Traffic key, Iterable<Text> values, Context context) throws IOException, InterruptedException {
           for (Text value : values) {
               context.write(new Text(value),key);
           }
       }
   }
}
分区内排序
package com.ccj.pxj.sort;
import org.apache.hadoop.io.WritableComparable;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
/**
 * 自定义序列化类且排序
 */
public class Traffic implements WritableComparable<Traffic> {
    private  String phone;
    private  long up;
    private  long down;
    private  long sum;
    public String getPhone() {
        return phone;
    }
    public void setPhone(String phone) {
        this.phone = phone;
    }
    public long getUp() {
        return up;
    }
    public void setUp(long up) {
        this.up = up;
    }
    public long getDown() {
        return down;
    }
    public void setDown(long down) {
        this.down = down;
    }
    public Traffic(String phone, long up, long down) {
        this.phone = phone;
        this.up = up;
        this.down = down;
        this.sum=up+down;
    }
    public Traffic() {
    }
    @Override
    public void write(DataOutput out) throws IOException {
            out.writeUTF(phone);
            out.writeLong(up);
            out.writeLong(down);
            out.writeLong(sum);
    }
    @Override
    public void readFields(DataInput in) throws IOException {
         this.phone= in.readUTF();
         this.up= in.readLong();
         this.down=in.readLong();
         this.sum=in.readLong();
    }
    @Override
    public String toString() {
        return
                phone + '\t' +
                 up +
                "\t" + down +
                "\t" + sum ;
    }
    @Override
    public int compareTo(Traffic o) {
        return this.sum>o.sum?-1:1;
    }
}
package com.ccj.pxj.sort;
import com.ccj.pxj.phone.utils.FileUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import java.io.IOException;
public class PartitionSortDriver {
    public static void main(String[] args) throws Exception {
        String input = "data/phone_data .txt";
        String output = "out";
        // 1）获取Job对象
        Configuration configuration = new Configuration();
        Job job = Job.getInstance(configuration);
        FileUtils.deleteOutput(configuration, output);
        // 2）本job对应要执行的主类是哪个
        job.setJarByClass(PartitionSortDriver.class);
        // 3）设置Mapper和Reducer
        job.setMapperClass(MyMaper.class);
        job.setReducerClass(MyReduce.class);
        job.setPartitionerClass(PhonePartitioner.class);
        job.setNumReduceTasks(3);
        // 4）设置Mapper阶段输出数据的类型
        job.setMapOutputKeyClass(Traffic.class);
        job.setMapOutputValueClass(Text.class);
        // 5）设置Reducer阶段输出数据的类型
        job.setOutputKeyClass(Traffic.class);
        job.setOutputValueClass(Text.class);
        // 6）设置输入和输出路径
        FileInputFormat.setInputPaths(job, new Path(input));
        FileOutputFormat.setOutputPath(job, new Path(output));
        // 7）提交作业
        boolean result = job.waitForCompletion(true);
        System.exit(result ? 0 : 1);
    }
   public static class  MyMaper extends Mapper<LongWritable, Text,Traffic, Text>{
       @Override
       protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
           String[] data = value.toString().split("\t");
           String phone = data[1];
           // 上行流量
           long up = Long.parseLong(data[data.length - 3]);
           // 下行流量
           long down = Long.parseLong(data[data.length - 2]);
           context.write(new Traffic(phone,up,down),new Text(phone));
       }
   }
   public  static  class MyReduce extends Reducer< Traffic,Text, Text, Traffic>{
      /* @Override
       protected void reduce(Text key, Iterable<Traffic> values, Context context) throws IOException, InterruptedException {
            long ups=0;
            long downs=0;
           for (Traffic value : values) {
               ups+=value.getUp();
               downs+=value.getDown();
           }
           context.write(NullWritable.get(),new Traffic(key.toString(),ups,downs));
       }*/
       @Override
       protected void reduce(Traffic key, Iterable<Text> values, Context context) throws IOException, InterruptedException {
           for (Text value : values) {
               context.write(new Text(value),key);
           }
       }
   }
}

OutputFormat

有一类很常见的需求：按照一定的规则把数据给我写到某个文件中去
package com.ccj.outputformat;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IOUtils;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.RecordWriter;
import org.apache.hadoop.mapreduce.TaskAttemptContext;
import java.io.IOException;
public class PxjRecordWriter extends RecordWriter<Text,NullWritable> {
   FSDataOutputStream pxjout;
   FSDataOutputStream wfyout;
   FileSystem fileSystem;
   public PxjRecordWriter(TaskAttemptContext context){
       try{
           fileSystem=FileSystem.get(context.getConfiguration());
           pxjout=fileSystem.create(new Path("out/pxj"));
           wfyout=fileSystem.create(new Path("out/wfy"));
       }catch (Exception e){
           e.printStackTrace();
       }
   }
    @Override
    public void write(Text key, NullWritable value) throws IOException, InterruptedException {
        if(key.toString().contains("pxj")){
            pxjout.write(key.toString().getBytes());
        }else{
            wfyout.write(key.toString().getBytes());
        }
    }
    @Override
    public void close(TaskAttemptContext context) throws IOException, InterruptedException {
        IOUtils.closeStream(pxjout);
        IOUtils.closeStream(wfyout);
    }
}
package com.ccj.outputformat;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.RecordWriter;
import org.apache.hadoop.mapreduce.TaskAttemptContext;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import java.io.IOException;
public class PxjDataOutputFormat extends FileOutputFormat<Text,NullWritable> {
    @Override
    public RecordWriter<Text, NullWritable> getRecordWriter(TaskAttemptContext job) throws IOException, InterruptedException {
        return new PxjRecordWriter(job);
    }
}
package com.ccj.outputformat;
import com.ccj.pxj.phone.utils.FileUtils;
import com.ccj.pxj.sort.PhonePartitioner;
import com.ccj.pxj.sort.Traffic;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import java.io.IOException;
public class PxjDataDriver {
    public static void main(String[] args) throws Exception {
        String input = "data/log.txt";
        String output = "out";
        // 1）获取Job对象
        Configuration configuration = new Configuration();
        Job job = Job.getInstance(configuration);
        FileUtils.deleteOutput(configuration, output);
        // 2）本job对应要执行的主类是哪个
        job.setJarByClass(PxjDataDriver.class);
        // 3）设置Mapper和Reducer
        job.setMapperClass(MyMaper.class);
        job.setReducerClass(MyReduce.class);
        //job.setPartitionerClass(PhonePartitioner.class);
      //  job.setNumReduceTasks(3);
        // 4）设置Mapper阶段输出数据的类型
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(NullWritable.class);
        // 5）设置Reducer阶段输出数据的类型
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(NullWritable.class);
        job.setOutputFormatClass(PxjDataOutputFormat.class);
        // 6）设置输入和输出路径
        FileInputFormat.setInputPaths(job, new Path(input));
        FileOutputFormat.setOutputPath(job, new Path(output));
        // 7）提交作业
        boolean result = job.waitForCompletion(true);
        System.exit(result ? 0 : 1);
    }
   public static class  MyMaper extends Mapper<LongWritable, Text,Text, NullWritable>{
       @Override
       protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
           context.write(value,NullWritable.get());
       }
   }
   public  static  class MyReduce extends Reducer< Text,NullWritable, Text,NullWritable>{
       @Override
       protected void reduce(Text key, Iterable<NullWritable> values, Context context) throws IOException, InterruptedException {
           for (NullWritable value : values) {
               context.write(new Text(key+"\r\n"),value);
           }
       }
   }
}

RecordWriter<Text, NullWritable>怎么拿到呢？
==> 抄源码
return new LineRecordWriter<K, V>(fileOut, keyValueSeparator);
上下文是一个框，什么东西都往里面装，什么东西都从里面取
源码、源码里面的单元测试 这是我们学习最好的资源
代码具有工业性的参考价值

作者：pxj(潘陈)
日期：2020-01-18