使用MapReduce求出各年销售笔数、各年销售总额

1）将上面的数据文件上传到hdfs

hdfs dfs -put sales.csv /input/

2）采用Eclipse/IDEA创建一个Maven工程，同时修改pom.xml文件，增加dependencies，/dependencies、build，/build节点，内容如下：

    <dependency>
      <groupId>org.apache.hadoop</groupId>
      <artifactId>hadoop-common</artifactId>
      <version>2.7.7</version>
    </dependency>
    <dependency>
      <groupId>org.apache.hadoop</groupId>
      <artifactId>hadoop-hdfs</artifactId>
      <version>2.7.7</version>
    </dependency>
    <dependency>
      <groupId>org.apache.hadoop</groupId>
      <artifactId>hadoop-mapreduce-client-core</artifactId>
      <version>2.7.7</version>
    </dependency>
    <dependency>
      <groupId>org.apache.hadoop</groupId>
      <artifactId>hadoop-client</artifactId>
      <version>2.7.7</version>
    </dependency>

    <plugins>
      <plugin>
        <groupId>org.apache.maven.plugins</groupId>
        <artifactId>maven-jar-plugin</artifactId>
        <version>2.6</version>
        <configuration>
          <archive>
            <manifest>
              <!-- main()所在的类，注意修改 -->
              <mainClass>org.example.SoldMain</mainClass>
            </manifest>
          </archive>
        </configuration>
      </plugin>
    </plugins>

3）开始开发java代码，需要4个类：

首先是主输出类SoldMain(代码如下)：

package org.example;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
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 SoldMain {
 public static void main(String[] args) throws Exception {
//1. 创建一个job和任务入口(指定主类)
Job job = Job.getInstance(new Configuration());
 job.setJarByClass(SoldMain.class);
 //2. 指定job的mapper和输出的类型<k2 v2>
job.setMapperClass(SoldMapper.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(Sold.class);
//3. 指定job的reducer和输出的类型<k4  v4>
 job.setReducerClass(SoldReduce.class);
 job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
//4.指定job的输入和输出路径
FileInputFormat.setInputPaths(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
//5. 执行job
 job.waitForCompletion(true);
 }
}

然后是SoldMapper类(代码如下)：

package org.example;
import org.apache.commons.lang.StringUtils;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
import java.io.IOException;
public class SoldMapper extends Mapper<LongWritable, Text, Text, Sold> {
 @Override
protected void map(LongWritable k1, Text v1,
Context context)
throws IOException, InterruptedException {
 //字段名 prod_id,cust_id,time,channel_id,promo_id,quantity_sold,amount_sold
//数据类型：Int，Int，Date， Int，Int ，Int ,float(10,2),
//数据: 13,987,1998-01-10,3,999,1,1232.16
String data = v1.toString();
String[] words = data.split(",");
//数据: t1=987,1998-01-10,3,999,1,1232.16
 String t1 = StringUtils.substringAfter(data, ",");
//数据: t2=1998-01-10,3,999,1,1232.16 
String t2 = StringUtils.substringAfter(t1, ",");
//取年份为偏移量，数据: words2[0]=1998，words2[1]=01，words2[2]=10,3,999,1,1232.16
String[] words2 = t2.split("-");
//        StringUtils.substringAfter("dskeabcedeh", "e");
//        /*结果是：abcedeh*/
 Sold sold = new Sold();
 sold.setTime(words[2]);//数组word[]
sold.setQuantity_sold(Integer.parseInt(words[5]));
sold.setAmount_sold(Float.valueOf(words[6]));
context.write(new Text(words2[0]), sold);//数组word2[],word2[0]代表年份作为k2
 }
}

接着是SoldReduce类(代码如下)：

package org.example;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;
import java.io.IOException;
public class SoldReduce extends Reducer<Text, Sold, Text, Text> {
 protected void reduce(Text k3, Iterable<Sold> v3, Context context) throws IOException, InterruptedException {
int total1 = 0;
float total2 = 0;
for (Sold sold : v3) {
total1 = total1 + sold.getQuantity_sold();
total2 = total2 + sold.getAmount_sold();
}
String total = "销售笔数:" + Integer.toString(total1) + "," + "销售总额:" + Float.toString(total2);
context.write(k3, new Text(total));
}
}

最后是Sold类(代码如下)：

package org.example;
import org.apache.hadoop.io.Writable;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
public class Sold implements Writable {
//字段名 prod_id,cust_id,time,channel_id,promo_id,quantity_sold,amount_sold
 //数据类型：Int，Int，Date， Int，Int ，Int ,float(10,2),
        //数据: 13, 987, 1998/1/10, 3, 999,1, 1232.16
//由以上定义变量
private int prod_id;
private int cust_id;
private String time;
private int channel_id;
private int promo_id;
private int quantity_sold;
private float amount_sold;//奖金
//序列化方法：将java对象转化为可跨机器传输数据流（二进制串/字节）的一种技术
public void write(DataOutput out) throws IOException {
    out.writeInt(this.prod_id);
    out.writeInt(this.cust_id);
    out.writeUTF(this.time);
    out.writeInt(this.channel_id);
    out.writeInt(this.promo_id);
    out.writeInt(this.quantity_sold);
    out.writeFloat(this.amount_sold);
}
//反序列化方法：将可跨机器传输数据流（二进制串）转化为java对象的一种技术
 public void readFields(DataInput in) throws IOException {
    this.prod_id = in.readInt();
    this.cust_id = in.readInt();
    this.time = in.readUTF();
    this.channel_id = in.readInt();
    this.promo_id = in.readInt();
    this.quantity_sold = in.readInt();
    this.amount_sold = in.readFloat();
 }
public int getProd_id() {
    return prod_id;
 }
 public void setProd_id(int prod_id) {
    this.prod_id = prod_id;
 }
 public int getCust_id() {
 return cust_id;
 }
 public void setCust_id(int cust_id) {
    this.cust_id = cust_id;
}
 public String getTime() {
 return time;
}
public void setTime(String time) {
this.time = time;
}
public int getChannel_id() {
 return channel_id;
}
 public void setChannel_id(int channel_id) {
 this.channel_id = channel_id;
 }
public int getPromo_id() {
 return promo_id;
}
 public void setPromo_id(int promo_id) {
 this.promo_id = promo_id;
}
 public int getQuantity_sold() {
return quantity_sold;
 }
public void setQuantity_sold(int quantity_sold) {
this.quantity_sold = quantity_sold;
     }
public float getAmount_sold() {
return amount_sold;
}
 public void setAmount_sold(float amount_sold) {
 this.amount_sold = amount_sold;
 }
}

4）使用命令(如下)打包:

mvn clean package

5）将jar包通过xftp传输到linux下，在hadoop环境运行jar包，命令如下：

hadoop jar annualTotal-0.0.1-SNAPSHOT.jar  /input/sales.csv  /output/sales

jar包名和输入输出名请自行修改
6）查看执行结果(命令如下)：

hdfs dfs -cat /output/sales/part-r-00000

输出路径请自行查看