四、PutMerge合并文件程序

一、需求

服务器的apache日志文件可能比较小，Hadoop更适合处理大文件，效率会更高，此时就需要合并分散的文件，开发一个PutMerge程序，用于合并本地文件后存入HDFS系统中

二、java代码

package org.apache.hadoop.studyhdfs;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IOUtils;

public class PutMergeFile {
    public static void main(String[] args) throws Exception  {
        //赋初始值
        args =new String[]{"/opt/dataFile","hdfs://Hadoop-senior02.beifeng.com:8020/input/putMerge.xml"};
        //1.get conf
        Configuration conf =new Configuration();
        //2.get  filesystem
        FileSystem hdfs =FileSystem.get(conf);
        FileSystem local =FileSystem.getLocal(conf);
        //3.get path
        Path localPath=new Path(args[0]);
        Path hdfsPath=new Path(args[1]);
        //4.inputFiles
        FileStatus[] inputFiles =local.listStatus(localPath);
        //out stream
        FSDataOutputStream outputStream =hdfs.create(hdfsPath);
        FSDataInputStream inputStream;
        for(FileStatus inputFile:inputFiles){
            System.out.println(inputFile.getPath().getName());
            inputStream=local.open(inputFile.getPath());
            IOUtils.copyBytes(inputStream, outputStream, 4096,false);
            IOUtils.closeStream(inputStream);
        }
        IOUtils.closeStream(outputStream);
    }

}

 三、结果查看

代码:$bin/hdfs dfs -text /input/putMerge.xml