大数据案例（九）——自定义Outputformat

代码下载地址：https://github.com/tazhigang/big-data-github.git

一、概述

要在一个mapreduce程序中根据数据的不同输出两类结果到不同目录，这类灵活的输出需求可以通过自定义outputformat来实现。

• 自定义outputformat，
• 改写recordwriter，具体改写输出数据的方法write()

二、案例需求

1. 需求:过滤输入的log日志中是否包含baidu
   • （1）包含atguigu的网站输出到j:/url/baidu_url.txt
   • （2）不包含atguigu的网站输出到j:/url/other_url.txt
2. 输入数据

=================log.txt====================
https://www.baidu.com
http://news.baidu.com
https://map.baidu.com
http://www.google.com
http://cn.bing.com
http://www.sohu.com
http://www.sina.com
https://github.com
https://my.oschina.net

3. 输出结果

=================baidu_url.txt====================
http://news.baidu.com	
https://map.baidu.com	
https://www.baidu.com	
=================other_url.txt====================
http://cn.bing.com	
http://www.google.com	
http://www.sina.com	
http://www.sohu.com	
https://github.com	
https://my.oschina.net	

三、创建maven项目

1. 项目结构
2. 代码实现

• HDFSUtil.java

  	package com.ittzg.hadoop.outputformat;
  
  	import org.apache.hadoop.conf.Configuration;
  	import org.apache.hadoop.fs.FileSystem;
  	import org.apache.hadoop.fs.Path;
  	import org.junit.After;
  	import org.junit.Before;
  	import org.junit.Test;
  
  	import java.io.IOException;
  	import java.net.URI;
  
  	/**
  	 * @email: tazhigang095@163.com
  	 * @author: ittzg
  	 * @date: 2019/7/7 22:54
  	 */
  	public class HDFSUtil {
  		Configuration configuration = new Configuration();
  		FileSystem fileSystem = null;
  
  		/**
  		 * 每次执行添加有@Test注解的方法之前调用
  		 */
  		@Before
  		public void init(){
  			configuration.set("fs.defaultFs","hadoop-ip-101:9000");
  			try {
  				fileSystem = FileSystem.get(new URI("hdfs://hadoop-ip-101:9000"),configuration,"hadoop");
  			} catch (Exception e) {
  				throw new RuntimeException("获取hdfs客户端连接异常");
  			}
  		}
  		/**
  		 * 每次执行添加有@Test注解的方法之后调用
  		 */
  		@After
  		public void closeRes(){
  			if(fileSystem!=null){
  				try {
  					fileSystem.close();
  				} catch (IOException e) {
  					throw new RuntimeException("关闭hdfs客户端连接异常");
  				}
  			}
  		}
  		/**
  		 * 上传文件
  		 */
  		@Test
  		public void putFileToHDFS(){
  			try {
  				fileSystem.copyFromLocalFile(new Path("F:\\big-data-github\\hadoop-parent\\hadoop-outputformat\\src\\main\\resources\\file\\log.txt"),new Path("/user/hadoop/outputformat/input/log.txt"));
  			} catch (IOException e) {
  				e.printStackTrace();
  				System.out.println(e.getMessage());
  			}
  		}
  		/**
  		 * 创建hdfs的目录
  		 * 支持多级目录
  		 */
  		@Test
  		public void mkdirAtHDFS(){
  			try {
  				boolean mkdirs = fileSystem.mkdirs(new Path("/user/hadoop/outputformat/input"));
  				System.out.println(mkdirs);
  			} catch (IOException e) {
  				e.printStackTrace();
  			}
  		}
  	}
  

• MyRecordWriter.java

  	package com.ittzg.hadoop.outputformat;
  
  	import org.apache.hadoop.conf.Configuration;
  	import org.apache.hadoop.fs.FSDataOutputStream;
  	import org.apache.hadoop.fs.FileSystem;
  	import org.apache.hadoop.fs.Path;
  	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;
  
  	/**
  	 * @email: tazhigang095@163.com
  	 * @author: ittzg
  	 * @date: 2019/7/7 22:57
  	 */
  	public class MyRecordWriter extends RecordWriter<Text, NullWritable> {
  
  		private FSDataOutputStream baiduOut = null;
  		private FSDataOutputStream otherOut = null;
  
  
  		public MyRecordWriter(TaskAttemptContext job) {
  			Configuration configuration = job.getConfiguration();
  			try {
  				FileSystem fileSystem = FileSystem.get(configuration);
  				//创建两个输入流
  				Path baiduPath = new Path("j:/url/baidu_url.txt");
  				Path otherPath = new Path("j:/url/other_url.txt");
  				baiduOut = fileSystem.create(baiduPath);
  				otherOut = fileSystem.create(otherPath);
  			} catch (IOException e) {
  				e.printStackTrace();
  			}
  		}
  
  		public void write(Text key, NullWritable value) throws IOException, InterruptedException {
  			if(key.toString().contains("baidu")){
  				baiduOut.write(key.toString().getBytes());
  			}else{
  				otherOut.write(key.toString().getBytes());
  			}
  		}
  
  		public void close(TaskAttemptContext context) throws IOException, InterruptedException {
  			if(baiduOut != null){
  				baiduOut.close();
  			}
  			if(otherOut != null){
  				otherOut.close();
  			}
  		}
  	}
  

• MyFileOutputFormat.java

  	package com.ittzg.hadoop.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;
  
  	/**
  	 * @email: tazhigang095@163.com
  	 * @author: ittzg
  	 * @date: 2019/7/7 22:55
  	 */
  	public class MyFileOutputFormat extends FileOutputFormat<Text,NullWritable> {
  		public RecordWriter<Text, NullWritable> getRecordWriter(TaskAttemptContext job) throws IOException, InterruptedException {
  			return new MyRecordWriter(job);
  		}
  	}
  

• MyFileOutputFormatDriver.java

  	package com.ittzg.hadoop.outputformat;
  
  	import org.apache.hadoop.conf.Configuration;
  	import org.apache.hadoop.fs.FileSystem;
  	import org.apache.hadoop.fs.Path;
  	import org.apache.hadoop.io.BytesWritable;
  	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 org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat;
  
  	import java.io.IOException;
  	import java.net.URI;
  
  	/**
  	 * @email: tazhigang095@163.com
  	 * @author: ittzg
  	 * @date: 2019/7/7 23:08
  	 */
  	public class MyFileOutputFormatDriver {
  		static class MyMapper extends Mapper<LongWritable,Text,Text,NullWritable>{
  			@Override
  			protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
  				context.write(value,NullWritable.get());
  			}
  		}
  
  		static class MyReduce extends Reducer<Text,NullWritable,Text,NullWritable> {
  			Text urlFormat = new Text();
  			@Override
  			protected void reduce(Text key, Iterable<NullWritable> values, Context context) throws IOException, InterruptedException {
  				urlFormat.set(key.toString()+"\t\n");
  				context.write(urlFormat,NullWritable.get());
  			}
  		}
  
  		public static void main(String[] args) throws Exception {
  			String input = "hdfs://hadoop-ip-101:9000/user/hadoop/outputformat/input";
  			String output = "hdfs://hadoop-ip-101:9000/user/hadoop/outputformat/output";
  
  			Configuration conf = new Configuration();
  			conf.set("mapreduce.app-submission.cross-platform","true");
  			Job job = Job.getInstance(conf);
  			job.setJarByClass(MyFileOutputFormatDriver.class);
  
  			job.setMapperClass(MyMapper.class);
  			job.setReducerClass(MyReduce.class);
  
  			job.setMapOutputKeyClass(Text.class);
  			job.setMapOutputValueClass(NullWritable.class);
  
  			job.setOutputKeyClass(Text.class);
  			job.setOutputValueClass(NullWritable.class);
  
  
  			FileSystem fs = FileSystem.get(new URI("hdfs://hadoop-ip-101:9000"),conf,"hadoop");
  			Path outPath = new Path(output);
  			if(fs.exists(outPath)){
  				fs.delete(outPath,true);
  			}
  			// 将自定义的输出格式组件设置到job中
  			job.setOutputFormatClass(MyFileOutputFormat.class);
  			FileInputFormat.setInputPaths(job, new Path(input));
  
  			// 虽然我们自定义了outputformat，但是因为我们的outputformat继承自fileoutputformat
  			// 而fileoutputformat要输出一个_SUCCESS文件，所以，在这还得指定一个输出目录
  			FileOutputFormat.setOutputPath(job, outPath);
  
  			boolean result = job.waitForCompletion(true);
  
  			System.exit(result ? 0 : 1);
  
  		}
  	}
  

四、运行结果

1. 网页浏览及本地文件浏览
   
   
2. 文件内容下载浏览
   
   

转载于:https://my.oschina.net/ittzg/blog/3070814