MapReduce之MapJoin案例

/缓存普通文件到Task运行节点。
job.addCacheFile(new URI("file://e:/cache/pd.txt");

public class JoinBean {
	
	private String orderId;
	private String pid;
	private String pname;
	private String amount;
	
	@Override
	public String toString() {
		return  orderId + "\t" +  pname + "\t" + amount ;
	}

	public String getOrderId() {
		return orderId;
	}

	public void setOrderId(String orderId) {
		this.orderId = orderId;
	}

	public String getPid() {
		return pid;
	}

	public void setPid(String pid) {
		this.pid = pid;
	}

	public String getPname() {
		return pname;
	}

	public void setPname(String pname) {
		this.pname = pname;
	}

	public String getAmount() {
		return amount;
	}

	public void setAmount(String amount) {
		this.amount = amount;
	}


}

/*
 * 1. 在Hadoop中，hadoop为MR提供了分布式缓存
 * 			①用来缓存一些Job运行期间的需要的文件(普通文件，jar，归档文件(har))
 * 			②通过在Job的Configuration中，使用uri代替要缓存的文件
 * 			③分布式缓存会假设当前的文件已经上传到了HDFS，并且在集群的任意一台机器都可以访问到这个URI所代表的文件
 * 			④分布式缓存会在每个节点的task运行之前，提前将文件发送到节点
 * 			⑤分布式缓存的高效是由于每个Job只会复制一次文件，且可以自动在从节点对归档文件解归档
 * 
 * 		
 * 
 * 
 */
public class MapJoinMapper extends Mapper<LongWritable, Text, JoinBean, NullWritable>{

	private JoinBean out_key=new JoinBean();
	private Map<String, String> pdDatas=new HashMap<String, String>();
	//在map之前手动读取pd.txt中的内容
	
	@Override
	protected void setup(Mapper<LongWritable, Text, JoinBean, NullWritable>.Context context)
			throws IOException, InterruptedException {
		
		//从分布式缓存中读取数据
		URI[] files = context.getCacheFiles();
		
		for (URI uri : files) {
			
			BufferedReader reader = new BufferedReader(new FileReader(new File(uri)));
			
			String line="";
			
			//循环读取pd.txt中的每一行
			while(StringUtils.isNotBlank(line=reader.readLine())) {
				
				String[] words = line.split("\t");
				
				pdDatas.put(words[0], words[1]);

			}
			
			reader.close();
			
		}
		
	}
	
	//对切片中order.txt的数据进行join，输出
	@Override
	protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, JoinBean, NullWritable>.Context context)
			throws IOException, InterruptedException {
		
		String[] words = value.toString().split("\t");
		
		out_key.setOrderId(words[0]);
		out_key.setPname(pdDatas.get(words[1]));
		out_key.setAmount(words[2]);
		
		context.write(out_key, NullWritable.get());
			
	}
	
}

public class MapJoinDriver {
	
	public static void main(String[] args) throws Exception {
		
		Path inputPath=new Path("e:/mrinput/mapjoin");
		Path outputPath=new Path("e:/mroutput/mapjoin");
		

		//作为整个Job的配置
		Configuration conf = new Configuration();
		//保证输出目录不存在
		FileSystem fs=FileSystem.get(conf);
		
		if (fs.exists(outputPath)) {
			
			fs.delete(outputPath, true);
			
		}
		
		// ①创建Job
		Job job = Job.getInstance(conf);
		
		job.setJarByClass(MapJoinDriver.class);
		
		
		// 为Job创建一个名字
		job.setJobName("wordcount");
		
		// ②设置Job
		// 设置Job运行的Mapper，Reducer类型，Mapper,Reducer输出的key-value类型
		job.setMapperClass(MapJoinMapper.class);
		
		// 设置输入目录和输出目录
		FileInputFormat.setInputPaths(job, inputPath);
		FileOutputFormat.setOutputPath(job, outputPath);
		
		// 设置分布式缓存
		job.addCacheFile(new URI("file:///e:/pd.txt"));
		
		//取消reduce阶段
		job.setNumReduceTasks(0);

		// ③运行Job
		job.waitForCompletion(true);
		
	}

}