hbase的基本操作

1. shell操作

常見命令：

[root@hadoop01 ~]# hbase shell      #進(jìn)入HBASE客戶端
hbase(main):001:0> help “dml”    #獲取一組命令的提示hbase(main):001:0> help "put" 獲取一個(gè)單獨(dú)命令的提示幫助
hbase(main):001:0> exit    #退出客戶端

 #查看hbase中的所有表
hbase(main):001:0>list       

創(chuàng)建表：
語法：create ‘表名’,’列簇1’,’ 列簇2’,’ 列簇3’....

# 例1:創(chuàng)建一張表，名稱為user,該表有info和data兩個(gè)列簇
hbase(main):001:0>create 'user' ,'info','data'
hbase(main):001:0>create 'user' ,{NAME=>'info'},{NAME=>'data'}

#例2:創(chuàng)建一張表叫做 user_info，包含兩個(gè)列簇 base_info 和 extra_info，并且分別指定這兩個(gè)列簇 的數(shù)據(jù)的版本數(shù)為 3 和 1
hbase(main):001:0>create 'user_info',{NAME=>'base_info',VERSIONS=>3},{NAME=>'extra_info',VERSIONS=>1}

查看表的詳細(xì)信息

#desc 或者 describe
hbase(main):001:0>desc 'user_info'
hbase(main):001:0>describe 'user_info'

插入數(shù)據(jù)
#put table ,rowkey, 列簇:列 ,value
hbase(main):001:0>put 'user','rowkey01','info:age','18'

查詢數(shù)據(jù)
#get

#獲取 user 表中 row key 為 rk0001 的所有信息
hbase(main):001:0>get 'user' ,'rk0001'

#獲取 user 表中 row key 為 rk0001，info 列簇的所有信息
hbase(main):001:0>get 'user','rk0001','info'

#獲取 user 表中 row key 為 rk0001，info 列簇的 name、age 列標(biāo)示符的信息
hbase(main):001:0>get 'user','rk0001','info:name','info:age'

#獲取 user 表中 row key 為 rk0001，列簇為 info，版本號(hào)最新 5 個(gè)的信息
hbase(main):001:0>create 'USER','rk0001',{COLUMN=>'info',VERSIONS=>5}

#獲取 user 表中 row key 為 rk0001，cell 的值為 zhangsan 的信息
hbase(main):001:0>get 'user', 'rk0001', {FILTER => "ValueFilter(=, 'binary:zhangsan')"}

#獲取 user 表中 row key 為 rk0001，列標(biāo)示符中含有 a 的信息
hbase(main):001:0>get 'user','rk0001',{FILTER=>"(QualifierFilter(=,'substring:a'))"}

查詢數(shù)據(jù)
#scan

查詢 user_info 表中的所有信息
hbase(main):001:0>scan 'user_info'

#查詢 user_info 表中的指定列簇的所有信息
hbase(main):001:0>scan 'user_info',{COLUMNS=>'base_info'}
hbase(main):001:0>scan 'user_info',{COLUMNS=>'base_info:name'}
hbase(main):001:0>scan 'user_info',{COLUMNS=>['base_info','extra_info']}

#查詢 user_info 表中的指定列簇為 base_info 的所有版本信息
hbase(main):001:0>scan 'user_info',{COLUMNS=>'base_info',VERSIONS=>5}

#查詢 user 表中列簇為 info 和 data 且列標(biāo)示符中含有 a 字符的信息
hbase(main):001:0>scan 'user', {COLUMNS => ['info', 'data'], FILTER => "(QualifierFilter(=,'substring:a'))"}

#rowkey的范圍查詢
hbase(main):001:0>scan 'user_info', {COLUMNS => 'base_info', STARTROW => 'baiyc_20150716_0003', ENDROW =>
'baiyc_20150716_0006'}

#查詢 user 表中 rowkey 以 rk 字符開頭的
hbase(main):001:0>scan 'user',{FILTER=>"PrefixFilter('rk')"}

#查詢 user_info 表中指定時(shí)間戳范圍的數(shù)據(jù)
hbase(main):001:0>scan 'user_info',{TIMERANGE=>[1540882871681,1540882888540]}

刪除數(shù)據(jù)
#delete

#刪除記錄
hbase(main):001:0>delete 'user', 'rk0001' 
#刪除字段
hbase(main):001:0>delete 'user','rk0001','info:name'
#刪除 user 表 rowkey 為 rk0001，列標(biāo)示符為 info:name，timestamp 為 1392383705316 的數(shù)據(jù)
hbase(main):001:0>delete 'user','rk0001','info:name',1392383705316

修改表結(jié)構(gòu)
#alter

#添加兩個(gè)列簇 f2 和 f3
hbase(main):001:0>alter 'user_info',NAME=>'f2'
hbase(main):001:0>alter 'user_info',NAME=>'f2'

#刪除一個(gè)列簇 f1
hbase(main):001:0>alter 'user_info',NAME=>'f1',METHOD=>'delete'

#將 user_info 表的 base_info 列簇版本號(hào)改為 5
hbase(main):001:0>alter 'user_info',NAME=>'base_info',VERSIONS=>5

清空表
#truncate

#清空 user 表中的數(shù)據(jù)
hbase(main):001:0>truncate 'user_info'

停用表/啟用表

#disable 和 enable
hbase(main):001:0>disable 'user'
hbase(main):001:0>enable 'user'

刪除表
#drop

#刪除一張表
hbase(main):001:0>disable 'user   #需要先停用這張表，在刪除
hbase(main):001:0>drop 'user'

2. API操作（javaAPI）

這里實(shí)現(xiàn)基本的增刪改查：
pom依賴

<dependency>
            <groupId>org.apache.hbase</groupId>
            <artifactId>hbase-client</artifactId>
            <version>1.2.6</version>
        </dependency>

        <dependency>
            <groupId>org.apache.hbase</groupId>
            <artifactId>hbase-server</artifactId>
            <version>1.2.6</version>
        </dependency>

代碼實(shí)現(xiàn)：

package com.zy.hbase;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.hbase.HBaseConfiguration;
import org.apache.hadoop.hbase.HColumnDescriptor;
import org.apache.hadoop.hbase.HTableDescriptor;
import org.apache.hadoop.hbase.KeyValue;
import org.apache.hadoop.hbase.client.*;
import org.apache.hadoop.hbase.util.Bytes;
import org.junit.Test;

import java.io.IOException;

public class HbaseTest {
    // 聲明靜態(tài)配置
    static Configuration conf = null;
    private static final String ZK_CONNECT_STR = "hadoop01:2181,hadoop02:2181,hadoop03:2181";

    static {
        conf = HBaseConfiguration.create();
        conf.set("hbase.zookeeper.quorum", ZK_CONNECT_STR);
    }

    /**
     * 1.創(chuàng)建表
     */
    @Test
    public void creatTable() {
        //表名
        String tableName = "stu_info";
        //列簇名
        String[] cf = {"base_info", "extra_info"};
        try {
            //創(chuàng)建HBaseAdmin對(duì)象
            HBaseAdmin admin = new HBaseAdmin(conf);
            //創(chuàng)建表的描述信息對(duì)象
            HTableDescriptor des = new HTableDescriptor(tableName);
            for (int i = 0; i < cf.length; i++) {
                des.addFamily(new HColumnDescriptor(cf[i]));
            }
            if (admin.tableExists(tableName)) {
                System.out.println("table Exists!");
                System.exit(0);
            } else {
                admin.createTable(des);
                System.out.println("create table Success!");
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 為表添加數(shù)據(jù)
     */
    @Test
    public void addData() {
        //表名
        String tableName = "user_info";
        //行鍵
        String rowkey = "rk0011";
        //列簇
        String cf = "base_info";
        Put put = new Put(rowkey.getBytes());
        try {
            //創(chuàng)建操作hbase表的對(duì)象
            HTable table = new HTable(conf, tableName.getBytes());
            //列簇  列  值
            put.addColumn(cf.getBytes(), "name".getBytes(), "zs".getBytes());
            table.put(put);
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 根據(jù) rwokey 查詢
     */
    @Test
    public void getResult() {
        //表名
        String tableName = "user_info";
        //行鍵
        String rowKey = "rk0001";

        Get get = new Get(Bytes.toBytes(rowKey));
        try {
            //創(chuàng)建操作hbase表的對(duì)象
            HTable table = new HTable(conf, tableName.getBytes());
            Result result = table.get(get);
            for (KeyValue kv : result.list()) {
                System.out.println("rowkey:" + kv.getRow());  //行鍵
                System.out.println("family" + kv.getFamily()); //列簇
                System.out.println("qualifier" + kv.getQualifier());  //列
                System.out.println("value:" + new String(kv.getValue())); //值
                System.out.println("Timestamp" + kv.getTimestamp());  //時(shí)間戳
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 遍歷查詢 hbase 表
     */
    @Test
    public void getResultScann() {
        String tableName = "user_info";
        Scan scan = new Scan();
        try {
            HTable table = new HTable(conf, tableName.getBytes());
            ResultScanner scanner = table.getScanner(scan);
            for (Result rs : scanner) {
                for (KeyValue kv : rs.list()) {
                    System.out.println("rowkey:" + kv.getRow());  //行鍵
                    System.out.println("family" + kv.getFamily()); //列簇
                    System.out.println("qualifier" + kv.getQualifier());  //列
                    System.out.println("value:" + new String(kv.getValue())); //值
                    System.out.println("Timestamp" + kv.getTimestamp());  //時(shí)間戳
                }
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 遍歷查詢 hbase 表  指定rowkey的查詢范圍
     */
    @Test
    public void getResultScannRange() {
        String tableName = "user_info";
        Scan scan = new Scan();
        scan.setStartRow("rk001".getBytes());
        scan.setStopRow("rk005".getBytes());
        try {
            HTable table = new HTable(conf, tableName.getBytes());
            ResultScanner scanner = table.getScanner(scan);
            for (Result rs : scanner) {
                for (KeyValue kv : rs.list()) {
                    System.out.println("rowkey:" + kv.getRow());  //行鍵
                    System.out.println("family" + kv.getFamily()); //列簇
                    System.out.println("qualifier" + kv.getQualifier());  //列
                    System.out.println("value:" + new String(kv.getValue())); //值
                    System.out.println("Timestamp" + kv.getTimestamp());  //時(shí)間戳
                }
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 查詢表中的某一列
     */
    @Test
    public void getResultByColumn() {
        String tableName = "user_info";  //表
        String rowkey = "rk0001";   //行鍵
        String cf = "base_info";    //列簇
        String column = "name";     //列
        try {
            HTable table = new HTable(conf, tableName.getBytes());
            Get get = new Get(rowkey.getBytes());
            get.addColumn(cf.getBytes(), column.getBytes());
            Result result = table.get(get);
            for (KeyValue kv : result.list()) {
                System.out.println("rowkey:" + kv.getRow());  //行鍵
                System.out.println("family" + kv.getFamily()); //列簇
                System.out.println("qualifier" + kv.getQualifier());  //列
                System.out.println("value:" + new String(kv.getValue())); //值
                System.out.println("Timestamp" + kv.getTimestamp());  //時(shí)間戳
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 更新表中的某一列,因?yàn)樵趆base中可以以時(shí)間戳定義多個(gè)版本的值
     * 所有，更新操作就是put操作
     */
    @Test
    public void updateTable() {
        String tableName = "user_info";  //表
        String rowkey = "rk0001";   //行鍵
        String cf = "base_info";    //列簇
        String column = "name";     //列
        String value = "ll";       //值
        try {
            HTable table = new HTable(conf, Bytes.toBytes(tableName));
            Put put = new Put(rowkey.getBytes());
            put.addColumn(cf.getBytes(), column.getBytes(), value.getBytes());
            table.put(put);
        } catch (IOException e) {
            e.printStackTrace();
        }

    }

    /**
     * 查詢某列數(shù)據(jù)的多個(gè)版本
     */
    @Test
    public void getResultByVersion() {
        String tableName = "user_info";  //表
        String rowkey = "rk0001";   //行鍵
        String cf = "base_info";    //列簇
        String column = "name";     //列
        int version = 5;          //hbase的列的版本
        try {
            HTable table = new HTable(conf, Bytes.toBytes(tableName));
            Get get = new Get(rowkey.getBytes());
            get.addColumn(cf.getBytes(), column.getBytes());
            get.setMaxVersions(version);
            Result result = table.get(get);
            if (result.list() != null) {
                for (KeyValue kv : result.list()) {
                    System.out.println("rowkey:" + kv.getRow());  //行鍵
                    System.out.println("family" + kv.getFamily()); //列簇
                    System.out.println("qualifier" + kv.getQualifier());  //列
                    System.out.println("value:" + new String(kv.getValue())); //值
                    System.out.println("Timestamp" + kv.getTimestamp());  //時(shí)間戳
                }
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 刪除指定的列
     */
    @Test
    public void deleteColumn() {
        String tableName = "user_info";  //表
        String rowkey = "rk0001";   //行鍵
        String cf = "base_info";    //列簇
        String column = "name";     //列
        try {
            HTable table = new HTable(conf, Bytes.toBytes(tableName));
            Delete delete = new Delete(rowkey.getBytes());
            delete.addColumn(cf.getBytes(), column.getBytes());
            table.delete(delete);
            System.out.println(cf + ":" + column + "is deleted!");
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 刪除指定rowKey的所有列
     */
    @Test
    public void deleteAllColumn() {
        String tableName = "user_info";  //表
        String rowkey = "rk0001";   //行鍵
        try {
            HTable table = new HTable(conf, Bytes.toBytes(tableName));
            Delete deleteAll = new Delete(rowkey.getBytes());
            table.delete(deleteAll);
            System.out.println("all columns are deleted!");
        } catch (IOException e) {
            e.printStackTrace();
        }
    }
    /*
     * 刪除表
     */
    @Test
    public void deleteTable(){
        String tableName = "user_info";  //表
        try {
            HBaseAdmin admin = new HBaseAdmin(conf);
            admin.disableTable(tableName.getBytes());
            admin.deleteTable(tableName.getBytes());
            System.out.println(tableName + "is deleted!");
        } catch (IOException e) {
            e.printStackTrace();
        }

    }
}

詳細(xì)的源代碼：http://down.51cto.com/data/2457979