HBase API 操作范例

时间:2021-10-05 05:22:17

package com.test.hbase.api;


import org.apache.hadoop.conf.Configuration;

import org.apache.hadoop.hbase.*;

import org.apache.hadoop.hbase.client.*;

import org.apache.hadoop.hbase.util.Bytes;


import java.io.IOException;

import java.util.ArrayList;

import java.util.List;


/**

* Created by zsq.

* Date: 2017/6/14

* desc:Hbase api操作

*/

public class HBaseAPIDemo {


public static void main(String[] args) throws IOException {

Configuration conf = HBaseConfiguration.create();

//设置zk的地址

conf.set("hbase.zookeeper.quorum", Constants.hostNames);


//获取链接hbase数据的链接对象

Connection conn = ConnectionFactory.createConnection(conf);

//过去时接口

//  HBaseAdmin admin=new HBaseAdmin(conn);


//获取操作hbase数据库的对象

HBaseAdmin ha = (HBaseAdmin) conn.getAdmin();


//        createNamespace(ha);

//        listNamespace(ha);

//        createTables(ha);

//        listNamespaceTables(ha);

//        putDataToTables(ha, conn);

//        putMultilDataToTables(ha, conn);

//        scanTablesData(ha, conn);

//        getTablesData(ha, conn);

deleteNsAndTable(ha, conn);



}



/**

* shell操作:create_namespace ‘ns2‘

* 说明:创建命名空间

*

* @param ha

* @throws IOException

*/

public static void createNamespace(HBaseAdmin ha) throws IOException {


ha.createNamespace(NamespaceDescriptor.create("ns2").build());

ha.close();

System.out.println("-----createNamespace----over");


}



/**

* shell操作:list_namespace

* 描述: 查看所有命名空间

*

* @param ha

* @throws IOException

*/

public static void listNamespace(HBaseAdmin ha) throws IOException {


NamespaceDescriptor[] listns = ha.listNamespaceDescriptors();

for (NamespaceDescriptor ns : listns) {

System.out.println(ns.getName());

}

ha.close();

System.out.println("-----listNamespace----over");

}



/**

* 创建表:help ‘create‘

* shell操作: create ‘ns2:stu‘, ‘cf_info‘, ‘cf_beizhu‘

*

* @param ha

* @throws IOException

*/

public static void createTables(HBaseAdmin ha) throws IOException {

//先判断表是否存在

if (!ha.tableExists("ns2:stu")) {

HTableDescriptor htable = new HTableDescriptor(TableName.valueOf("ns2:stu"));

//创建一个表至少需要添加一个列族

htable.addFamily(new HColumnDescriptor("cf_info"));

htable.addFamily(new HColumnDescriptor("cf_beizhu"));

ha.createTable(htable);

}

ha.close();

System.out.println("-----createTables----over");

}


/**

* shell操作:

* 查看命名空下的表:

* list_namespace_tables ‘ns2‘

*

* @param ha

* @throws IOException

*/

public static void listNamespaceTables(HBaseAdmin ha) throws IOException {


HTableDescriptor[] htables = ha.listTableDescriptorsByNamespace("ns2");

for (HTableDescriptor tb : htables) {

System.out.println(tb.getTableName());

}

ha.close();

System.out.println("-----createTables----over");

}



/**

* 添加数据到表:help ‘put‘

* shell操作:put ‘ns1:t1‘, ‘r1‘, ‘c1‘, ‘value‘

*

* @param ha

* @param conn

* @throws IOException

*/

public static void putDataToTables(HBaseAdmin ha, Connection conn) throws IOException {


//判断表是否存在

if (ha.tableExists("ns2:stu")) {

Table mTable = conn.getTable(TableName.valueOf("ns2:stu"));

//创建Put对象并且添加rowkey

Put put = new Put("soft_20170101".getBytes());

//第一个参数是列族,第二个参数是列名,,第三个参数是列的值

put.addColumn("cf_info".getBytes(), "name".getBytes(), "laowang".getBytes());

put.addColumn("cf_info".getBytes(), "age".getBytes(), "20".getBytes());

put.addColumn("cf_info".getBytes(), "sex".getBytes(), "nan".getBytes());

put.addColumn("cf_beizhu".getBytes(), "address".getBytes(), "fangshan".getBytes());

mTable.put(put);

}


ha.close();

System.out.println("-----putDataToTables----over");

}


/**

* 批量添加数据到Hbase表

*

* @param ha

* @param conn

* @throws IOException

*/

public static void putMultilDataToTables(HBaseAdmin ha, Connection conn) throws IOException {


//判断表是否存在

if (ha.tableExists("ns2:stu")) {

Table mTable = conn.getTable(TableName.valueOf("ns2:stu"));


List<Put> listput = new ArrayList<Put>();

for (int i = 0; i < 100; i++) {

//创建Put对象并且添加rowkey

Put put = new Put(("soft_20170101" + i).getBytes());

//第一个参数是列族,第二个参数是列名,第三个参数是列的值

put.addColumn("cf_info".getBytes(), "name".getBytes(), ("laowang" + i).getBytes());

put.addColumn("cf_info".getBytes(), "age".getBytes(), "20".getBytes());

if (i % 2 == 1) {

put.addColumn("cf_info".getBytes(), "sex".getBytes(), "nv".getBytes());

} else {

put.addColumn("cf_info".getBytes(), "sex".getBytes(), "nan".getBytes());

}

put.addColumn("cf_beizhu".getBytes(), "address".getBytes(), ("fangshan" + i).getBytes());

listput.add(put);

}


//批量插入

mTable.put(listput);

}


ha.close();

System.out.println("-----putMultilDataToTables----over");

}



/**

* 查看表数据:

* scan ‘ns2:stu‘

* 查看一个列族

* scan ‘ns2:stu‘,{COLUMNS => ‘cf_info‘}

* 查看多个列族

* scan ‘ns2:stu‘,{COLUMNS => [‘cf_info‘,‘cf_beizhu‘]}

* 查看某个列的数据:

* scan ‘ns2:stu‘,{COLUMNS => ‘cf_info:name‘}

*

* @param ha

* @param conn

* @throws IOException

*/

public static void scanTablesData(HBaseAdmin ha, Connection conn) throws IOException {


//判断表是否存在

if (ha.tableExists("ns2:stu")) {

Table mTable = conn.getTable(TableName.valueOf("ns2:stu"));


//扫描整张表

Scan scan = new Scan();

//scan.addFamily("cf_info".getBytes());

//            scan.addColumn("cf_info".getBytes(), "name".getBytes());

ResultScanner rs = mTable.getScanner(scan);

for (Result result : rs) {

System.out.println("name:" + Bytes.toString(result.getValue("cf_info".getBytes(), "name".getBytes())));

System.out.println("age:" + Bytes.toString(result.getValue("cf_info".getBytes(), "age".getBytes())));

System.out.println("sex:" + Bytes.toString(result.getValue("cf_info".getBytes(), "sex".getBytes())));

System.out.println("address:" + Bytes.toString(result.getValue("cf_beizhu".getBytes(), "address".getBytes())));

}

}


ha.close();

System.out.println("-----scanTablesData----over");

}


/**

* 查看某一行数据:help ‘get‘

* <p>

* get ‘ns1:stu‘,‘data_20170102‘

* <p>

* 查看某行的其中的一个列族的数据:

* get ‘ns1:stu‘,‘data_20170102‘,{COLUMN => ‘cf_info‘}

* <p>

* 查看某行的其中的多个列族的数据:

* get ‘ns1:stu‘,‘data_20170102‘,{COLUMN => [‘cf_info‘,‘cf_beizhu‘]}

* <p>

* 获取某一个列的值:

* get ‘ns1:stu‘,‘data_20170102‘,‘cf_info:name‘

*

* @param ha

* @param conn

* @throws IOException

*/


public static void getTablesData(HBaseAdmin ha, Connection conn) throws IOException {


//判断表是否存在

if (ha.tableExists("ns2:stu")) {

Table mTable = conn.getTable(TableName.valueOf("ns2:stu"));


//获取整行数据

Get get = new Get("soft_2017010196".getBytes());

//            get.addFamily("cf_info".getBytes());

get.addColumn("cf_info".getBytes(), "name".getBytes());

Result result = mTable.get(get);


System.out.println("name:" + Bytes.toString(result.getValue("cf_info".getBytes(), "name".getBytes())));

System.out.println("age:" + Bytes.toString(result.getValue("cf_info".getBytes(), "age".getBytes())));

System.out.println("sex:" + Bytes.toString(result.getValue("cf_info".getBytes(), "sex".getBytes())));

System.out.println("address:" + Bytes.toString(result.getValue("cf_beizhu".getBytes(), "address".getBytes())));

}


ha.close();

System.out.println("-----getTablesData----over");

}


/**

* 操作:删除命名空间和命名空间所有表

*

* @param ha

* @param conn

* @throws IOException

*/

public static void deleteNsAndTable(HBaseAdmin ha, Connection conn) throws IOException {


//遍历命名空间下所有表

HTableDescriptor[] htables = ha.listTableDescriptorsByNamespace("ns1");

for (HTableDescriptor tb : htables) {

System.out.println(tb.getTableName());

if (ha.isTableEnabled(tb.getTableName())) {

ha.disableTable(tb.getTableName());

}

ha.deleteTable(tb.getTableName());

}

ha.deleteNamespace("ns1");

ha.close();

System.out.println("-----deleteNsAndTable----over");

}



}


// HBase 的过滤器操作


public class HbaseAPIFilter {


public static void main(String[] args) throws Exception {

//        singleColumnValueFilter();


//        qualifierFilter();

familyFilter();

}



/**

* 1、单个列值过滤   SingleColumnValueFilter

*

* @throws IOException

*/

public static void singleColumnValueFilter() throws IOException {

Table mTable = TableUtils.getTable(Constants.tableName);



//如果匹配,那么可用获取行数据

//单个列值的匹配:比较基类为BinaryComparator   BinaryComparator 匹配完整字节数组

// SingleColumnValueFilter scvf = new SingleColumnValueFilter("cf_info".getBytes(), "name".getBytes(), CompareFilter.CompareOp.EQUAL, "laowang88".getBytes());

SingleColumnValueFilter scvf = new SingleColumnValueFilter("cf_info".getBytes(), "name".getBytes(), CompareFilter.CompareOp.NOT_EQUAL, new BinaryComparator("laowang88".getBytes()));


//BinaryPrefixComparator 匹配开始的部分字节数组 name的值开始部分为"xiao"的行都过滤出来

SingleColumnValueFilter scvf2 = new SingleColumnValueFilter("cf_info".getBytes(), "name".getBytes(), CompareFilter.CompareOp.GREATER_OR_EQUAL, new BinaryPrefixComparator("xiao".getBytes()));


//        Only EQUAL or NOT_EQUAL comparisons are valid with this comparator.

//        所以对于RegexStringComparator基类只能用EQUAL或者NOT_EQUAL

SingleColumnValueFilter scvf3 = new SingleColumnValueFilter("cf_info".getBytes(), "name".getBytes(), CompareFilter.CompareOp.EQUAL, new RegexStringComparator("^[x].*$"));



//        Only EQUAL or NOT_EQUAL tests are valid with this comparator.

//        所以对于SubstringComparator基类只能用EQUAL或者NOT_EQUAL

SingleColumnValueFilter scvf4 = new SingleColumnValueFilter("cf_info".getBytes(), "name".getBytes(), CompareFilter.CompareOp.EQUAL, new SubstringComparator("bai"));



Scan scan = new Scan();

scan.setFilter(scvf4);

ResultScanner rs = mTable.getScanner(scan);

for (Result result : rs) {

System.out.println("name:" + Bytes.toString(result.getValue("cf_info".getBytes(), "name".getBytes())));

System.out.println("age:" + Bytes.toString(result.getValue("cf_info".getBytes(), "age".getBytes())));

System.out.println("sex:" + Bytes.toString(result.getValue("cf_info".getBytes(), "sex".getBytes())));

System.out.println("address:" + Bytes.toString(result.getValue("cf_beizhu".getBytes(), "address".getBytes())));

}

}



/**

* 2. 列过滤器(列过滤器表中存在该列的数据都出来)

* QualifierFilter

*

* @throws IOException

*/

public static void qualifierFilter() throws IOException {

Table mTable = TableUtils.getTable(Constants.tableName);


//列名过滤:

//比较基类为BinaryComparator   BinaryComparator 匹配完整字节数组

QualifierFilter columnsNameFilter = new QualifierFilter(CompareFilter.CompareOp.EQUAL, new BinaryComparator("name".getBytes()));

QualifierFilter columnsNameFilter2 = new QualifierFilter(CompareFilter.CompareOp.EQUAL, new BinaryComparator("age".getBytes()));


//        Only EQUAL or NOT_EQUAL tests are valid with this comparator.

//        所以对于SubstringComparator基类只能用EQUAL或者NOT_EQUAL

QualifierFilter columnsNameFilter3 = new QualifierFilter(CompareFilter.CompareOp.EQUAL, new SubstringComparator("a"));



//BinaryPrefixComparator 匹配开始的部分字节数组, (可能用)

// RegexStringComparator, 正则表达式匹配(可能用)


Scan scan = new Scan();

scan.setFilter(columnsNameFilter3);

ResultScanner rs = mTable.getScanner(scan);

for (Result result : rs) {

System.out.println("name:" + Bytes.toString(result.getValue("cf_info".getBytes(), "name".getBytes())));

System.out.println("age:" + Bytes.toString(result.getValue("cf_info".getBytes(), "age".getBytes())));

System.out.println("sex:" + Bytes.toString(result.getValue("cf_info".getBytes(), "sex".getBytes())));

System.out.println("address:" + Bytes.toString(result.getValue("cf_beizhu".getBytes(), "address".getBytes())));

}


}



/**

* 定义列族过滤器(表中存在该列族的数据都出来)

* FamilyFilter

*

* @throws IOException

*/

public static void familyFilter() throws IOException {

Table mTable = TableUtils.getTable(Constants.tableName);


//因为是根据列族的名称过滤,所以传入列族名称

FamilyFilter familyFilter = new FamilyFilter(CompareFilter.CompareOp.GREATER_OR_EQUAL, new BinaryPrefixComparator("cf_i".getBytes()));


//        Only EQUAL or NOT_EQUAL tests are valid with this comparator.

//        所以对于RegexStringComparator基类只能用EQUAL或者NOT_EQUAL

FamilyFilter familyFilter2 = new FamilyFilter(CompareFilter.CompareOp.EQUAL, new RegexStringComparator("^[a-z_]{3}[^b].+$"));



//        BinaryComparator 匹配完整字节数组, (可能用)

//        SubstringComparator 比配子串、大小写不敏感(可能用)


Scan scan = new Scan();

scan.setFilter(familyFilter2);

ResultScanner rs = mTable.getScanner(scan);

for (Result result : rs) {

System.out.println("name:" + Bytes.toString(result.getValue("cf_info".getBytes(), "name".getBytes())));

System.out.println("age:" + Bytes.toString(result.getValue("cf_info".getBytes(), "age".getBytes())));

System.out.println("sex:" + Bytes.toString(result.getValue("cf_info".getBytes(), "sex".getBytes())));

System.out.println("address:" + Bytes.toString(result.getValue("cf_beizhu".getBytes(), "address".getBytes())));

}


}


}


注意: 前提是保证集群正常运行和HBase集群正常工作。