Hadoop学习之hdfs的操作

Hadoop学习之hdfs的操作

1.将HDFS中的文件复制到本地

package com.shujia.hdfs;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;public class Demo02Download {FileSystem fileSystem;// 创建执行对象// @Before: 前置通知, 在方法执行之前执行@Beforepublic void getFileSystem() throws IOException {Configuration entries = new Configuration();entries.set("fs.defaultFS", "hdfs://master:9000");fileSystem = FileSystem.get(entries);}// 实现文件复制到本地// @Test的作用，省略了public static void main(String[] args) {，表示测试类的方法@Testpublic void getData() throws IOException {String hdfsPath = "/NOTICE.txt";String localPath = "data/";// 将HDFS中的文件复制到本地fileSystem.copyToLocalFile(new Path(hdfsPath),new Path(localPath));}// @After: 后置通知, 在方法执行之后执行 。@Afterpublic void close() throws IOException {fileSystem.close();}}

2.上传数据到HDFS中

package com.shujia.hdfs;import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;public class Demo04PutData {public static void main(String[] args) throws IOException, URISyntaxException, InterruptedException {/*上传数据到HDFS中*/putData();putData2();}public static void putData() throws IOException {// 没有设置用户信息Configuration entries = new Configuration();entries.set("fs.defaultFS","hdfs://master:9000");FileSystem fileSystem = FileSystem.get(entries);// 从本地上传文件到HDFS上fileSystem.copyFromLocalFile(new Path("hadoop/data/students.txt"),new Path("/data/"));fileSystem.close();}public static void putData2() throws IOException, URISyntaxException, InterruptedException {// 设置了用户信息Configuration entries = new Configuration();entries.set("fs.defaultFS","hdfs://master:9000");/*FileSystem get(final URI uri, final Configuration conf,final String user)*/URI uri = new URI("hdfs://master:9000");// 获取FileSystem的实体类对象（传递uri到get函数中吗，会更改上传到HDFS中文件的用户信息）FileSystem fileSystem = FileSystem.get(uri,entries,"root");fileSystem.copyFromLocalFile(new Path("hadoop/data/students.txt"),new Path("/data/"));fileSystem.close();}
}

3.在HDFS上创建文件目录

package com.shujia.hdfs;import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;public class Demo05MakeDir {public static void main(String[] args) throws IOException, URISyntaxException, InterruptedException {/*上传数据到HDFS中*/mkdir();}public static void mkdir() throws IOException, URISyntaxException, InterruptedException {// 设置了用户信息Configuration entries = new Configuration();entries.set("fs.defaultFS","hdfs://master:9000");URI uri = new URI("hdfs://master:9000");FileSystem fileSystem = FileSystem.get(uri,entries,"root");fileSystem.mkdirs(new Path("/api"));
//        fileSystem.mkdirs(new Path("/api/1/2"));fileSystem.close();}
}

4.删除HDFS上的文件目录

package com.shujia.hdfs;import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;public class Demo06Delete {public static void main(String[] args) throws IOException, URISyntaxException, InterruptedException {delete();}public static void delete() throws IOException, URISyntaxException, InterruptedException {// 设置了用户信息Configuration entries = new Configuration();entries.set("fs.defaultFS","hdfs://master:9000");URI uri = new URI("hdfs://master:9000");FileSystem fileSystem = FileSystem.get(uri,entries,"root");//        fileSystem.delete(new Path("/api/1/2"));//TODO 参数recursive:如果path是一个目录并设置为true，则删除该目录，否则抛出异常。//               在文件的情况下，递归可以设置为true或false。fileSystem.delete(new Path("/api"),true);fileSystem.close();}
}

5.查看HDFS文件系统中文件和目录的元数据

package com.shujia.hdfs;import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;public class Demo07Liststatus {public static void main(String[] args) throws IOException, URISyntaxException, InterruptedException {getBlockLocation();}public static void getBlockLocation() throws IOException, URISyntaxException, InterruptedException {Configuration entries = new Configuration();entries.set("fs.defaultFS","hdfs://master:9000");URI uri = new URI("hdfs://master:9000");FileSystem fileSystem = FileSystem.get(uri,entries,"root");FileStatus fileStatus = fileSystem.getFileStatus(new Path("/hadoop-3.1.3.tar.gz"));System.out.println("路径:"+fileStatus.getPath());System.out.println("长度:"+fileStatus.getLen());System.out.println("副本数:"+fileStatus.getReplication());/*获取一个文件的文件指定开始和结束的部分数据所在的Block块位置BlockLocation[] getFileBlockLocations(FileStatus file,long start, long len)*/BlockLocation[] fileBlockLocations = fileSystem.getFileBlockLocations(fileStatus, 0, fileStatus.getLen());for (BlockLocation fileBlockLocation : fileBlockLocations) {System.out.println("整个长度："+fileBlockLocation.getLength());System.out.println("偏移量，从文件的什么位置开始："+fileBlockLocation.getOffset());System.out.println("整个主机："+fileBlockLocation.getHosts());System.out.println("整个名称："+fileBlockLocation.getNames());}fileSystem.close();}public static void getFileStatus() throws IOException, URISyntaxException, InterruptedException {Configuration entries = new Configuration();entries.set("fs.defaultFS","hdfs://master:9000");URI uri = new URI("hdfs://master:9000");FileSystem fileSystem = FileSystem.get(uri,entries,"root");// getFileStatus()获取FileStatus对象// FileStatus对象封装了文件系统中文件和目录的元数据，包括文件的长度、块大小、备份数、修改时间、所有者以及权限等信息。FileStatus fileStatus = fileSystem.getFileStatus(new Path("/hadoop-3.1.3.tar.gz"));System.out.println("路径:"+fileStatus.getPath());System.out.println("长度:"+fileStatus.getLen());System.out.println("副本数:"+fileStatus.getReplication());fileSystem.close();}public static void listStatus() throws IOException, URISyntaxException, InterruptedException {// 没有设置用户信息Configuration entries = new Configuration();entries.set("fs.defaultFS","hdfs://master:9000");URI uri = new URI("hdfs://master:9000");FileSystem fileSystem = FileSystem.get(uri,entries,"root");// listStatus()获取FileStatus对象数组，遍历根目录下的所有文件和目录的元数据FileStatus[] fileStatuses = fileSystem.listStatus(new Path("/"));for (FileStatus fileStatus : fileStatuses) {// 判断其是否为文件（检查这个抽象路径名表示的文件是否是普通文件），若为目录则输出其路径if (fileStatus.isFile()) {long blockSize = fileStatus.getBlockSize();System.out.println(fileStatus.getPath());System.out.println("Block块大小:"+blockSize);System.out.println("长度:"+fileStatus.getLen());}else {System.out.println(fileStatus.getPath());}}fileSystem.close();}
}