【简单易懂版】使用IDEA操作Hadoop（增删改查）

---

前提：服务器中已经配置好了hadoop
本人亲测，以下代码已经跑通，基础功能都可以完成！！！希望对大家有用！！！

一、引入hdfs依赖

1. 创建一个maven项目cn.et
2. 本地maven配置阿里镜像，用于快速下载依赖（重启加载）
3. pow文件中引入hadoop依赖

		<!--	hdfs依赖	-->
		<dependency>
			<groupId>org.apache.hadoop</groupId>
			<artifactId>hadoop-client</artifactId>
			<version>3.1.3</version>
		</dependency>

二、创建hdfs工具类

创建一个HdfsApiUtils 类，用于实现hdfs的增删改查：

1. 获取hdfs的代码需要封装到静态代码块(先执行而且只执行一次)
2. 创建文件或目录（mkdirs，建议写绝对路径hdfs://地址:9000/新目录）
3. 删除文件或目录（delete）
4. 修改或移动文件或目录（rename）
5. 查询当前路径下所有的文件或目录（显示时间和MB）

package com.example.springbootonline.utils;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.*;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.context.annotation.Bean;
import org.springframework.stereotype.Component;

import java.io.IOException;
import java.net.URI;
import java.util.List;


/**
 * Created with IntelliJ IDEA.
 *
 * @Author: Print
 * @Date: 2023/07/17/10:07
 * @Description:idea集成hdfs；地址hdfs://地址:9000
 */

@Component
public class HdfsApiUtils {
    private static String hdfsUrl = "hdfs://地址:9000";


    private static String hdfsUsername = "root";

    private static FileSystem hdfs;

    static {
        Configuration conf = new Configuration();
        // 上传到云服务器需要配置下面这个句话
        conf.set("dfs.client.use.datanode.hostname","true");
        try {
            hdfs = FileSystem.get(URI.create(hdfsUrl), conf, hdfsUsername);
        } catch (Exception e) {
            throw new RuntimeException(e);
        }
    }

    /**
     * 创建文件或目录
     */
    public boolean mkdir(String path) {
        boolean res = false;
        try {
            hdfs.mkdirs(new Path(path));
            res = true;
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
        return res;
    }

    /**
     * 删除文件或目录
     */
    public boolean delete(String path) {
        boolean res = false;
        try {
            res = hdfs.delete(new Path(path), true);
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
        return res;
    }

    /**
     * 修改或移动文件或目录
     */
    public boolean rename(String oldFile, String newFlie) {
        boolean res = false;
        try {
            res = hdfs.rename(new Path(oldFile), new Path(newFlie));
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
        return res;
    }

    /**
     * 查询当前路径下所有的文件或目录（只查当前目录下）
     */
    public FileStatus[] findCurrent(String path) {
        FileStatus[] res = null;
        try {
            res = hdfs.listStatus(new Path(path));
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
        return res;
    }

    /**
     * 查询当前路径下所有的文件或目录（递归查下面所有）
     */
    public RemoteIterator<LocatedFileStatus> findAll(String path){
        RemoteIterator<LocatedFileStatus> iterator = null;
        try {
            iterator = hdfs.listFiles(new Path(path),true);
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
        return iterator;
    }

    /**
     * 上传
     */
    public boolean upload(String localPath, String path) {
        boolean res = false;
        try {
            hdfs.copyFromLocalFile(new Path(localPath), new Path(path));
            res = true;
        } catch (IOException e) {
            res = false;
            throw new RuntimeException(e);
        }
        return res;
    }

    /**
     * 下载
     */
    public boolean download(String hdfsPath, String localPath) {
        boolean res = false;
        try {
            hdfs.copyToLocalFile(new Path(hdfsPath), new Path(localPath));
            res = true;
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
        return res;
    }
}

三、测试hdfs工具类

import com.example.springbootonline.utils.HdfsApiUtils;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.RemoteIterator;
import org.junit.jupiter.api.Test;
import org.springframework.stereotype.Component;

import java.io.IOException;

/**
 * Created with IntelliJ IDEA.
 *
 * @Author: Print
 * @Date: 2023/07/17/10:59
 * @Description:
 */
@Component
public class HdfsApiUtilsTest {
    HdfsApiUtils hdfsApiUtils = new HdfsApiUtils();

    @Test
    public void mkdir(){
        String newFile = "/file";
        System.out.println(hdfsApiUtils.mkdir(newFile));
    }

    @Test
    public void delete(){
        String path = "/aaa";
        System.out.println(hdfsApiUtils.delete(path));
    }

    @Test
    public void rename(){
        String oldFile = "/aaa",newFile = "/newAAA";
        System.out.println(hdfsApiUtils.rename(oldFile,newFile));
    }

    @Test
    public void upload(){
        String localPath = "F:\\Users\\HP\\Videos\\Captures\\demo.mp4",path = "/abc/aaa";
        System.out.println(hdfsApiUtils.upload(localPath,path));
    }

    @Test
    public void findCurrent(){
        String path = "/file";
        FileStatus[] fss = hdfsApiUtils.findCurrent(path);
        for (FileStatus fs:fss) {
            System.out.println(fs.toString()+"\n");
        }
        System.out.println();
    }

    @Test
    public void findAll() throws IOException {
        String path = "/file";
        RemoteIterator<LocatedFileStatus> iterator = hdfsApiUtils.findAll(path);
        while (iterator.hasNext()){
            System.out.println(iterator.next().toString());
        }
    }
}

反思

好像应该再写一个服务器如何配置hadoop，后面再看有没有时间吧