hbase开发实例
1、put/checkAndPut
package com.testdata;
import java.io.IOException;<br/>
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import org.apache.hadoop.hbase.client.Put;<br/>
import org.apache.hadoop.hbase.util.Bytes;
public class TestPut {
public static void main(String[] args) throws IOException {
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);
HTableInterface table = conn.getTable("testdata");<br/>
Put testput = new Put(Bytes.toBytes("row1"));<br/>
testput.add(Bytes.toBytes("cf"),Bytes.toBytes("col1"),Bytes.toBytes("E"));<br/>
table.put(testput);<br/>
//使用checkAndPut<br/>
table.checkAndPut(Bytes.toBytes("row1"), Bytes.toBytes("cf"),Bytes.toBytes("col5"),Bytes.toBytes("E"),testput);<br/>
table.close();<br/>
conn.close();
}
}
使用checkAndPut，需要先对数据进行验证，上面的例子中，向row1中的cf:col1写入数据”E”，而验证的是row1中的cf:col5的值是否为”E”，注意这一点，相当于加了条件。
2、使用get读取数据
package com.testdata;
import java.io.IOException;<br/>
import java.util.List;
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.Cell;<br/>
import org.apache.hadoop.hbase.CellUtil;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.Get;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import org.apache.hadoop.hbase.client.Result;
import org.apache.hadoop.hbase.util.Bytes;
public class TestGet {
public static void main(String[] args) throws IOException {<br/>
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);<br/>
HTableInterface table = conn.getTable("testdata");
Get testget = new Get(Bytes.toBytes("row1"));<br/>
Result row1 = table.get(testget);<br/>
String value = new String(row1.getValue(Bytes.toBytes("cf"), Bytes.toBytes("col1")));<br/>
System.out.println(value);<br/>
//下面限定到具体的列<br/>
testget.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("col2"));<br/>
Result result = table.get(testget);<br/>
if(result.getValue(Bytes.toBytes("cf"), Bytes.toBytes("col2")) != null){<br/>
String value2 = new String(result.getValue(Bytes.toBytes("cf"), Bytes.toBytes("col2")));<br/>
System.out.println(value2);<br/>
}<br/>
//另外一种读取方式<br/>
List<Cell> cells = row1.listCells();<br/>
for(Cell cell : cells){<br/>
String rowkey = new String(CellUtil.cloneRow(cell));<br/>
String family = new String(CellUtil.cloneFamily(cell));<br/>
String collumn = new String(CellUtil.cloneQualifier(cell));<br/>
String cvalue = new String(CellUtil.cloneValue(cell));<br/>
System.out.println("rowkey:" + rowkey + " family:" + family + " column:" + collumn +" value:" + cvalue);
}
//注意要关闭<br/>
table.close();<br/>
conn.close();
}
}
参考结果：
3、使用scan获取数据
package com.testdata;
import java.io.IOException;<br/>
import java.util.List;<br/>
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.Cell;<br/>
import org.apache.hadoop.hbase.CellUtil;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import org.apache.hadoop.hbase.client.Result;<br/>
import org.apache.hadoop.hbase.client.ResultScanner;<br/>
import org.apache.hadoop.hbase.client.Scan;
public class TestScan {
public static void main(String[] args) throws IOException {<br/>
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);<br/>
HTableInterface table = conn.getTable("testdata");
Scan testscan =new Scan();<br/>
ResultScanner rs = table.getScanner(testscan);<br/>
for(Result r : rs ){<br/>
List<Cell> cells = r.listCells();<br/>
for(Cell cell : cells){<br/>
String rowkey = new String(CellUtil.cloneRow(cell));<br/>
String family = new String(CellUtil.cloneFamily(cell));<br/>
String collumn = new String(CellUtil.cloneQualifier(cell));<br/>
String cvalue = new String(CellUtil.cloneValue(cell));<br/>
System.out.println("rowkey:" + rowkey + " family:" + family + " column:" + collumn +" value:" + cvalue);
}<br/>
}<br/>
rs.close();<br/>
table.close();<br/>
conn.close();
}
}
4、delete/checkAndDelete
package com.testdata;
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.Delete;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import java.io.IOException;
import org.apache.hadoop.hbase.util.Bytes;
public class TestDelete {
public static void main(String[] args) throws IOException {<br/>
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);<br/>
HTableInterface table = conn.getTable("testdata");
Delete testdelete = new Delete(Bytes.toBytes("row1"));<br/>
testdelete.deleteColumns(Bytes.toBytes("cf"), Bytes.toBytes("col1"));<br/>
//区别只是checkAndDelete需要进行验证，相当于加了前提条件<br/>
//table.delete(testdelete);<br/>
table.checkAndDelete(Bytes.toBytes("row1"), Bytes.toBytes("cf"), Bytes.toBytes("col2"),Bytes.toBytes("BC"), testdelete);<br/>
table.close();<br/>
conn.close();
}
}
5、append
package com.testdata;
import java.io.IOException;<br/>
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.Append;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import org.apache.hadoop.hbase.util.Bytes;
public class TestAppend {
public static void main(String[] args) throws IOException {<br/>
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);<br/>
HTableInterface table = conn.getTable("testdata");
Append testappend = new Append(Bytes.toBytes("row1"));<br/>
testappend.add(Bytes.toBytes("cf"),Bytes.toBytes("col1"),Bytes.toBytes("F"));<br/>
table.append(testappend);<br/>
table.close();<br/>
conn.close();<br/>
}
}
下面是结果，注意append是在原有的值之上附加，先前的值为”E”，现在变为”EF”
6、计数器
计数器可以用于统计用户数，点击量等信息
package com.testdata;
import java.io.IOException;<br/>
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import org.apache.hadoop.hbase.util.Bytes;
public class TestIncrement {
public static void main(String[] args) throws IOException {<br/>
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);<br/>
HTableInterface table = conn.getTable("testdata");
long result = table.incrementColumnValue(Bytes.toBytes("row1"),Bytes.toBytes("cf"),Bytes.toBytes("coli"), 10);
System.out.println(result);<br/>
table.close();<br/>
conn.close();
}
}
注意 long result = table.incrementColumnValue(Bytes.toBytes(“row1”),Bytes.toBytes(“cf”),Bytes.toBytes(“coli”), 10);
最后一个参数，可以为0，意味着读取，也可以是负数。
可以使用get_counter可以获取对应的计数器的值，也可以使用以下命令进行操作
incr '<table>', '<row>', '<column>', |<increment-value>|
7、filter
使用时注意性能
package com.testdata;
import java.io.IOException;<br/>
import java.util.List;<br/>
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.Cell;<br/>
import org.apache.hadoop.hbase.CellUtil;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import org.apache.hadoop.hbase.client.Result;<br/>
import org.apache.hadoop.hbase.client.ResultScanner;<br/>
import org.apache.hadoop.hbase.client.Scan;<br/>
import org.apache.hadoop.hbase.filter.BinaryComparator;<br/>
import org.apache.hadoop.hbase.filter.BinaryPrefixComparator;<br/>
import org.apache.hadoop.hbase.filter.CompareFilter;<br/>
import org.apache.hadoop.hbase.filter.Filter;<br/>
import org.apache.hadoop.hbase.filter.QualifierFilter;<br/>
import org.apache.hadoop.hbase.filter.RowFilter;<br/>
import org.apache.hadoop.hbase.filter.SubstringComparator;<br/>
import org.apache.hadoop.hbase.util.Bytes;
public class TestSimplefilter {
public static void main(String[] args) throws IOException {<br/>
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);<br/>
HTableInterface table = conn.getTable("testdata");
Scan sc = new Scan();<br/>
sc.setCacheBlocks(false);<br/>
//行过滤器，判断"row1"与行的key是否相等<br/>
//Filter filter = new RowFilter(CompareFilter.CompareOp.EQUAL,new BinaryComparator(Bytes.toBytes("row1")));<br/>
//是否以"row"为前缀<br/>
//Filter filter = new RowFilter(CompareFilter.CompareOp.EQUAL,new BinaryPrefixComparator(Bytes.toBytes("row")));<br/>
//是否包含"row"<br/>
//Filter filter = new RowFilter(CompareFilter.CompareOp.EQUAL,new SubstringComparator("row"));
//列过滤器，与行类似<br/>
Filter filter = new QualifierFilter(CompareFilter.CompareOp.EQUAL,new BinaryComparator(Bytes.toBytes("col1")));
sc.setFilter(filter);
ResultScanner rs = table.getScanner(sc);<br/>
for(Result r : rs ){<br/>
List<Cell> cells = r.listCells();<br/>
for(Cell cell : cells){<br/>
String rowkey = new String(CellUtil.cloneRow(cell));<br/>
String family = new String(CellUtil.cloneFamily(cell));<br/>
String collumn = new String(CellUtil.cloneQualifier(cell));<br/>
String cvalue = new String(CellUtil.cloneValue(cell));<br/>
System.out.println("rowkey:" + rowkey + " family:" + family + " column:" + collumn +" value:" + cvalue);<br/>
}<br/>
}<br/>
rs.close();<br/>
table.close();<br/>
conn.close();<br/>
}<br/>
}
使用filterlist
package com.testdata;
import java.io.IOException;<br/>
import java.util.ArrayList;<br/>
import java.util.List;<br/>
import org.apache.hadoop.conf.Configuration;<br/>
import org.apache.hadoop.hbase.Cell;<br/>
import org.apache.hadoop.hbase.CellUtil;<br/>
import org.apache.hadoop.hbase.HBaseConfiguration;<br/>
import org.apache.hadoop.hbase.client.HConnection;<br/>
import org.apache.hadoop.hbase.client.HConnectionManager;<br/>
import org.apache.hadoop.hbase.client.HTableInterface;<br/>
import org.apache.hadoop.hbase.client.Result;<br/>
import org.apache.hadoop.hbase.client.ResultScanner;<br/>
import org.apache.hadoop.hbase.client.Scan;<br/>
import org.apache.hadoop.hbase.filter.BinaryComparator;<br/>
import org.apache.hadoop.hbase.filter.BinaryPrefixComparator;<br/>
import org.apache.hadoop.hbase.filter.CompareFilter;<br/>
import org.apache.hadoop.hbase.filter.Filter;<br/>
import org.apache.hadoop.hbase.filter.FilterList;<br/>
import org.apache.hadoop.hbase.filter.QualifierFilter;<br/>
import org.apache.hadoop.hbase.filter.RegexStringComparator;<br/>
import org.apache.hadoop.hbase.filter.RowFilter;<br/>
import org.apache.hadoop.hbase.filter.SingleColumnValueFilter;<br/>
import org.apache.hadoop.hbase.filter.SubstringComparator;<br/>
import org.apache.hadoop.hbase.util.Bytes;
public class TestFilterList {
public static void main(String[] args) throws IOException {<br/>
Configuration conf = HBaseConfiguration.create();<br/>
conf.set("hbase.zookeeper.quorum","localhost");<br/>
conf.set("hbase.zookeeper.property.clientPort","2181");<br/>
HConnection conn = HConnectionManager.createConnection(conf);<br/>
HTableInterface table = conn.getTable("testdata");
Scan sc = new Scan();<br/>
Filter filter1 = new RowFilter(CompareFilter.CompareOp.EQUAL,new BinaryComparator(Bytes.toBytes("row2")));<br/>
SingleColumnValueFilter filter2 = new SingleColumnValueFilter(Bytes.toBytes("cf"),Bytes.toBytes("col1"),CompareFilter.CompareOp.EQUAL,new BinaryPrefixComparator(Bytes.toBytes("B")));
SingleColumnValueFilter filter3 = new SingleColumnValueFilter(Bytes.toBytes("cf"),Bytes.toBytes("col1"),CompareFilter.CompareOp.EQUAL,new RegexStringComparator("B|C"));<br/>
filter2.setFilterIfMissing(true);<br/>
filter3.setFilterIfMissing(true);
//List<Filter> filters = new ArrayList<Filter>();<br/>
//filters.add(filter1);<br/>
//filters.add(filter2);<br/>
//FilterList filterlist = new FilterList(filters);
//也可以这样写,MUST_PASS_ALL标识满足所有的filter,当然也可以使用MUST_PASS_ONE，标识只需要满足一个<br/>
FilterList filterlist = new FilterList(FilterList.Operator.MUST_PASS_ALL);<br/>
filterlist.addFilter(filter1);<br/>
filterlist.addFilter(filter2);<br/>
filterlist.addFilter(filter3);<br/>
sc.addColumn(Bytes.toBytes("cf"),Bytes.toBytes("col1"));<br/>
sc.setFilter(filterlist);
ResultScanner rs = table.getScanner(sc);<br/>
for(Result r : rs ){<br/>
List<Cell> cells = r.listCells();<br/>
for(Cell cell : cells){<br/>
String rowkey = new String(CellUtil.cloneRow(cell));<br/>
String family = new String(CellUtil.cloneFamily(cell));<br/>
String collumn = new String(CellUtil.cloneQualifier(cell));<br/>
String cvalue = new String(CellUtil.cloneValue(cell));<br/>
System.out.println("rowkey:" + rowkey + " family:" + family + " column:" + collumn +" value:" + cvalue);<br/>
}<br/>
}<br/>
rs.close();<br/>
table.close();<br/>
conn.close();<br/>
}<br/>
}
以上一组filter标识了这样的条件，即行的key必须为”row2″，列名必须为”col1″，值必须为”B”
结果参考：
如果没有 sc.addColumn(Bytes.toBytes(“cf”),Bytes.toBytes(“col1”));这一句，结果会是下面的样子
rowkey:row2 family:cf column:col1 value:B<br/> rowkey:row2 family:cf column:colb value:U
问题出在 SingleColumnValueFilter filter2 = new SingleColumnValueFilter(Bytes.toBytes(“cf”),Bytes.toBytes(“col1”),CompareFilter.CompareOp.EQUAL,new BinaryPrefixComparator(Bytes.toBytes(“B”)));这一句，如果打印Bytes.toBytes(“B”)与Bytes.toBytes(“U”)，会发现都是以”B”开头的。即使换成BinaryComparator，也不会解决问题。
这里是值得注意的地方，搜索网络可以发现一样的结论，使用时务必使用sc.addColumn(Bytes.toBytes(“cf”),Bytes.toBytes(“col1”))类似的语句。
rowkey:row2 family:cf column:col1 value:B
转发申明：
本文转自互联网，由小站整理并发布，在于分享相关技术和知识。版权归原作者所有，如有侵权，请联系本站 top8488@163.com，将在24小时内删除。谢谢
