利用Java連接Hadoop進(jìn)行編程

更新時間：2022年06月28日 15:40:16 作者：wr456wr

這篇文章主要介紹了利用Java連接Hadoop進(jìn)行編程，文章圍繞主題展開詳細(xì)的內(nèi)容介紹，具有一定的參考價值，感興趣的小伙伴可以參考一下

實驗環(huán)境

hadoop版本：3.3.2
jdk版本：1.8
hadoop安裝系統(tǒng)：ubuntu18.04
編程環(huán)境：IDEA
編程主機(jī)：windows

實驗內(nèi)容

測試Java遠(yuǎn)程連接hadoop

創(chuàng)建maven工程，引入以下依賴：

<dependency>
                <groupId>org.testng</groupId>
                <artifactId>testng</artifactId>
                <version>RELEASE</version>
                <scope>compile</scope>
            </dependency>
            <dependency>
                <groupId>org.apache.hadoop</groupId>
                <artifactId>hadoop-common</artifactId>
            </dependency>
            <dependency>
                <groupId>org.apache.hadoop</groupId>
                <artifactId>hadoop-hdfs</artifactId>
                <version>3.3.2</version>
            </dependency>
            <dependency>
                <groupId>org.apache.hadoop</groupId>
                <artifactId>hadoop-common</artifactId>
                <version>3.3.2</version>
            </dependency>
            <dependency>
                <groupId>org.apache.hadoop</groupId>
                <artifactId>hadoop-core</artifactId>
                <version>1.2.1</version>
            </dependency>

虛擬機(jī)的/etc/hosts配置

hdfs-site.xml配置

<configuration>
        <property>
                <name>dfs.replication</name>
                <value>1</value>
        </property>
        <property>
                <name>dfs.namenode.name.dir</name>
                <value>file:/root/rDesk/hadoop-3.3.2/tmp/dfs/name</value>
        </property>
        <property>
                <name>dfs.datanode.http.address</name>
                <value>VM-12-11-ubuntu:50010</value>
        </property>
        <property>
                <name>dfs.client.use.datanode.hostname</name>
                <value>true</value>
        </property>
        <property>
                <name>dfs.datanode.data.dir</name>
                <value>file:/root/rDesk/hadoop-3.3.2/tmp/dfs/data</value>
        </property>
</configuration>

core-site.xml配置

<configuration>
  <property>
          <name>hadoop.tmp.dir</name>
          <value>file:/root/rDesk/hadoop-3.3.2/tmp</value>
          <description>Abase for other temporary directories.</description>
  </property>
  <property>
          <name>fs.defaultFS</name>
          <value>hdfs://VM-12-11-ubuntu:9000</value>
  </property>
</configuration>

啟動hadoop

sbin/start-dfs.sh

主機(jī)的hosts（C:\Windows\System32\drivers\etc）文件配置

嘗試連接到虛擬機(jī)的hadoop并讀取文件內(nèi)容，這里我讀取hdfs下的/root/iinput文件內(nèi)容

Java代碼：

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.hdfs.DistributedFileSystem;
public class TestConnectHadoop {
    public static void main(String[] args) throws Exception {

        String hostname = "VM-12-11-ubuntu";
        String HDFS_PATH = "hdfs://" + hostname + ":9000";
        Configuration conf = new Configuration();
        conf.set("fs.defaultFS", HDFS_PATH);
        conf.set("fs.hdfs.impl", DistributedFileSystem.class.getName());
        conf.set("dfs.client.use.datanode.hostname", "true");

        FileSystem fs = FileSystem.get(conf);
        FileStatus[] fileStatuses = fs.listStatus(new Path("/"));
        for (FileStatus fileStatus : fileStatuses) {
            System.out.println(fileStatus.toString());
        }
        FileStatus fileStatus = fs.getFileStatus(new Path("/root/iinput"));
        System.out.println(fileStatus.getOwner());
        System.out.println(fileStatus.getGroup());

        System.out.println(fileStatus.getPath());
        FSDataInputStream open = fs.open(fileStatus.getPath());
        byte[] buf = new byte[1024];
        int n = -1;
        StringBuilder sb = new StringBuilder();
        while ((n = open.read(buf)) > 0) {
            sb.append(new String(buf, 0, n));
        }
        System.out.println(sb);
    }
}

運行結(jié)果：

編程實現(xiàn)一個類“MyFSDataInputStream”，該類繼承“org.apache.hadoop.fs.FSDataInputStream",要求如下： ①實現(xiàn)按行讀取HDFS中指定文件的方法”readLine（）“，如果讀到文件末尾，則返回為空，否則返回文件一行的文本

思路：emmm我的思路比較簡單，只適用于該要求，僅作參考。
將所有的數(shù)據(jù)讀取出來存儲起來，然后根據(jù)換行符進(jìn)行拆分，將拆分的字符串?dāng)?shù)組存儲起來，用于readline返回

Java代碼

import org.apache.hadoop.fs.FSDataInputStream;
import java.io.IOException;
import java.io.InputStream;
public class MyFSDataInputStream extends FSDataInputStream {
    private String data = null;
    private String[] lines = null;
    private int count = 0;
    private FSDataInputStream in;
    public MyFSDataInputStream(InputStream in) throws IOException {
        super(in);
        this.in = (FSDataInputStream) in;
        init();
    }
    private void init() throws IOException {
        byte[] buf = new byte[1024];
        int n = -1;
        StringBuilder sb = new StringBuilder();
        while ((n = this.in.read(buf)) > 0) {
            sb.append(new String(buf, 0, n));
        }
        data = sb.toString();
        lines = data.split("\n");
    }
    /**
     * 實現(xiàn)按行讀取HDFS中指定文件的方法”readLine（）“，如果讀到文件末尾，則返回為空，否則返回文件一行的文本
     */
    public String read_line() {
        return count < lines.length ? lines[count++] : null;
    }

}

測試類：

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.hdfs.DistributedFileSystem;
public class TestConnectHadoop {
    public static void main(String[] args) throws Exception {

        String hostname = "VM-12-11-ubuntu";
        String HDFS_PATH = "hdfs://" + hostname + ":9000";
        Configuration conf = new Configuration();
        conf.set("fs.defaultFS", HDFS_PATH);
        conf.set("fs.hdfs.impl", DistributedFileSystem.class.getName());
        conf.set("dfs.client.use.datanode.hostname", "true");
        FileSystem fs = FileSystem.get(conf);
        FileStatus fileStatus = fs.getFileStatus(new Path("/root/iinput"));
        System.out.println(fileStatus.getOwner());
        System.out.println(fileStatus.getGroup());
        System.out.println(fileStatus.getPath());
        FSDataInputStream open = fs.open(fileStatus.getPath());
        MyFSDataInputStream myFSDataInputStream = new MyFSDataInputStream(open);
        String line = null;
        int count = 0;
        while ((line = myFSDataInputStream.read_line()) != null ) {
            System.out.printf("line %d is: %s\n", count++, line);
        }
        System.out.println("end");

    }
}

運行結(jié)果：

②實現(xiàn)緩存功能，即利用”MyFSDataInputStream“讀取若干字節(jié)數(shù)據(jù)時，首先查找緩存，如果緩存中有所需要數(shù)據(jù)，則直接由緩存提供，否則從HDFS中讀取數(shù)據(jù)

import org.apache.hadoop.fs.FSDataInputStream;
import java.io.BufferedInputStream;
import java.io.IOException;
import java.io.InputStream;
public class MyFSDataInputStream extends FSDataInputStream {
    private BufferedInputStream buffer;
    private String[] lines = null;
    private int count = 0;
    private FSDataInputStream in;
    public MyFSDataInputStream(InputStream in) throws IOException {
        super(in);
        this.in = (FSDataInputStream) in;
        init();
    }
    private void init() throws IOException {
        byte[] buf = new byte[1024];
        int n = -1;
        StringBuilder sb = new StringBuilder();
        while ((n = this.in.read(buf)) > 0) {
            sb.append(new String(buf, 0, n));
        }
        //緩存數(shù)據(jù)讀取
        buffer = new BufferedInputStream(this.in);
        lines = sb.toString().split("\n");
    }
    /**
     * 實現(xiàn)按行讀取HDFS中指定文件的方法”readLine（）“，如果讀到文件末尾，則返回為空，否則返回文件一行的文本
     */
    public String read_line() {
        return count < lines.length ? lines[count++] : null;
    }
    @Override
    public int read() throws IOException {
        return this.buffer.read();
    }
    public int readWithBuf(byte[] buf, int offset, int len) throws IOException {
        return this.buffer.read(buf, offset, len);
    }
    public int readWithBuf(byte[] buf) throws IOException {
        return this.buffer.read(buf);
    }
}

到此這篇關(guān)于利用Java連接Hadoop進(jìn)行編程的文章就介紹到這了,更多相關(guān)Java連接Hadoop內(nèi)容請搜索腳本之家以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持腳本之家！

您可能感興趣的文章: