学步园

Creating secondary indexes in HBase-0.19.3:

You
need to enable indexing in HBase before you can create a secondary
index on columns. Edit the file $HBASE_INSTALL_DIR/conf/hbase-site.xml
and add the following property to it.

<property><br /> <name>hbase.regionserver.class</name><br /> <value>org.apache.hadoop.hbase.ipc.IndexedRegionInterface</value><br /> </property><br /> <property><br /> <name>hbase.regionserver.impl</name><br /> <value><br /> org.apache.hadoop.hbase.regionserver.tableindexed.IndexedRegionServer<br /> </value><br /> </property><br />
Adding secondary index while creating table:

HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> HTableDescriptor desc = new HTableDescriptor("test_table");<br /> desc.addFamily(new HColumnDescriptor("columnfamily1:"));<br /> desc.addFamily(new HColumnDescriptor("columnfamily2:"));<br /> desc.addIndex(new IndexSpecification("column1",<br /> Bytes.toBytes("columnfamily1:column1")));<br /> desc.addIndex(new IndexSpecification("column2",<br /> Bytes.toBytes("columnfamily1:column2")));<br /> IndexedTableAdmin admin = null;<br /> admin = new IndexedTableAdmin(conf);<br /> admin.createTable(desc);

Adding index in an existing table:

HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTableAdmin admin = null;<br /> admin = new IndexedTableAdmin(conf);<br /> admin.addIndex(Bytes.toBytes("test_table"), new IndexSpecification("column2",<br /> Bytes.toBytes("columnfamily1:column2")));

Deleting existing index from a table.

HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTableAdmin admin = null;<br /> admin = new IndexedTableAdmin(conf);<br /> admin.removeIndex(Bytes.toBytes("test_table"), "column2");

Reading from secondary indexed columns:

To read from a secondary index, get a scanner for the index and scan through the data.

HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTable table = new IndexedTable(conf, Bytes.toBytes("test_table"));<br /> // You need to specify which columns to get<br /> Scanner scanner = table.getIndexedScanner("column1",<br /> HConstants.EMPTY_START_ROW, null, null, new byte[][] {<br /> Bytes.toBytes("columnfamily1:column1"),<br /> Bytes.toBytes("columnfamily1:column2") });<br /> for (RowResult rowResult : scanner) {<br /> String value1 = new String(<br /> rowResult.get(Bytes.toBytes("columnfamily1:column1")).getValue());<br /> String value2 = new String(<br /> rowResult.get(Bytes.toBytes("columnfamily1:column2")).getValue());<br /> System.out.println(value1 + ", " + value2);<br /> }<br /> table.close();

To get a scanner to a subset of the rows specify a column filter.

ColumnValueFilter filter =<br /> new ColumnValueFilter(Bytes.toBytes("columnfamily1:column1"),<br /> CompareOp.LESS, Bytes.toBytes("value1-10"));<br /> scanner = table.getIndexedScanner("column1", HConstants.EMPTY_START_ROW,<br /> null, filter, new byte[][] { Bytes.toBytes("columnfamily1:column1"),<br /> Bytes.toBytes("columnfamily1:column2"));<br /> for (RowResult rowResult : scanner) {<br /> String value1 = new String(<br /> rowResult.get(Bytes.toBytes("columnfamily1:column1")).getValue());<br /> String value2 = new String(<br /> rowResult.get(Bytes.toBytes("columnfamily1:column2")).getValue());<br /> System.out.println(value1 + ", " + value2);<br /> }

Example Code:

import java.io.IOException;<br /> import java.util.Date;<br /> import org.apache.hadoop.fs.Path;<br /> import org.apache.hadoop.hbase.HBaseConfiguration;<br /> import org.apache.hadoop.hbase.HColumnDescriptor;<br /> import org.apache.hadoop.hbase.HConstants;<br /> import org.apache.hadoop.hbase.HTableDescriptor;<br /> import org.apache.hadoop.hbase.client.Scanner;<br /> import org.apache.hadoop.hbase.client.tableindexed.IndexSpecification;<br /> import org.apache.hadoop.hbase.client.tableindexed.IndexedTable;<br /> import org.apache.hadoop.hbase.client.tableindexed.IndexedTableAdmin;<br /> import org.apache.hadoop.hbase.filter.ColumnValueFilter;<br /> import org.apache.hadoop.hbase.filter.ColumnValueFilter.CompareOp;<br /> import org.apache.hadoop.hbase.io.BatchUpdate;<br /> import org.apache.hadoop.hbase.io.RowResult;<br /> import org.apache.hadoop.hbase.util.Bytes;<br /> public class SecondaryIndexTest {<br /> public void writeToTable() throws IOException {<br /> HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTable table = new IndexedTable(conf, Bytes.toBytes("test_table"));<br /> String row = "test_row";<br /> BatchUpdate update = null;<br /> for (int i = 0; i < 100; i++) {<br /> update = new BatchUpdate(row + i);<br /> update.put("columnfamily1:column1", Bytes.toBytes("value1-" + i));<br /> update.put("columnfamily1:column2", Bytes.toBytes("value2-" + i));<br /> table.commit(update);<br /> }<br /> table.close();<br /> }<br /> public void readAllRowsFromSecondaryIndex() throws IOException {<br /> HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTable table = new IndexedTable(conf, Bytes.toBytes("test_table"));<br /> Scanner scanner = table.getIndexedScanner("column1",<br /> HConstants.EMPTY_START_ROW, null, null, new byte[][] {<br /> Bytes.toBytes("columnfamily1:column1"),<br /> Bytes.toBytes("columnfamily1:column2") });<br /> for (RowResult rowResult : scanner) {<br /> System.out.println(Bytes.toString(<br /> rowResult.get(Bytes.toBytes("columnfamily1:column1")).getValue())<br /> + ", " + Bytes.toString(rowResult.get(<br /> Bytes.toBytes("columnfamily1:column2")).getValue()<br /> ));<br /> }<br /> table.close();<br /> }<br /> public void readFilteredRowsFromSecondaryIndex() throws IOException {<br /> HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTable table = new IndexedTable(conf, Bytes.toBytes("test_table"));<br /> ColumnValueFilter filter =<br /> new ColumnValueFilter(Bytes.toBytes("columnfamily1:column1"),<br /> CompareOp.LESS, Bytes.toBytes("value1-40"));<br /> Scanner scanner = table.getIndexedScanner("column1",<br /> HConstants.EMPTY_START_ROW, null, filter,<br /> new byte[][] { Bytes.toBytes("columnfamily1:column1"),<br /> Bytes.toBytes("columnfamily1:column2")<br /> });<br /> for (RowResult rowResult : scanner) {<br /> System.out.println(Bytes.toString(<br /> rowResult.get(Bytes.toBytes("columnfamily1:column1")).getValue())<br /> + ", " + Bytes.toString(rowResult.get(<br /> Bytes.toBytes("columnfamily1:column2")).getValue()<br /> ));<br /> }<br /> table.close();<br /> }<br /> public void createTableWithSecondaryIndexes() throws IOException {<br /> HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> HTableDescriptor desc = new HTableDescriptor("test_table");<br /> desc.addFamily(new HColumnDescriptor("columnfamily1:column1"));<br /> desc.addFamily(new HColumnDescriptor("columnfamily1:column2"));<br /> desc.addIndex(new IndexSpecification("column1",<br /> Bytes.toBytes("columnfamily1:column1")));<br /> desc.addIndex(new IndexSpecification("column2",<br /> Bytes.toBytes("columnfamily1:column2")));<br /> IndexedTableAdmin admin = null;<br /> admin = new IndexedTableAdmin(conf);<br /> if (admin.tableExists(Bytes.toBytes("test_table"))) {<br /> if (admin.isTableEnabled("test_table")) {<br /> admin.disableTable(Bytes.toBytes("test_table"));<br /> }<br /> admin.deleteTable(Bytes.toBytes("test_table"));<br /> }<br /> if (admin.tableExists(Bytes.toBytes("test_table-column1"))) {<br /> if (admin.isTableEnabled("test_table-column1")) {<br /> admin.disableTable(Bytes.toBytes("test_table-column1"));<br /> }<br /> admin.deleteTable(Bytes.toBytes("test_table-column1"));<br /> }<br /> admin.createTable(desc);<br /> }<br /> public void addSecondaryIndexToExistingTable() throws IOException {<br /> HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTableAdmin admin = null;<br /> admin = new IndexedTableAdmin(conf);<br /> admin.addIndex(Bytes.toBytes("test_table"),<br /> new IndexSpecification("column2",<br /> Bytes.toBytes("columnfamily1:column2")));<br /> }<br /> public void removeSecondaryIndexToExistingTable() throws IOException {<br /> HBaseConfiguration conf = new HBaseConfiguration();<br /> conf.addResource(new Path("/opt/hbase-0.19.3/conf/hbase-site.xml"));<br /> IndexedTableAdmin admin = null;<br /> admin = new IndexedTableAdmin(conf);<br /> admin.removeIndex(Bytes.toBytes("test_table"), "column2");<br /> }<br /> public static void main(String[] args) throws IOException {<br /> SecondaryIndexTest test = new SecondaryIndexTest();<br /> test.createTableWithSecondaryIndexes();<br /> test.writeToTable();<br /> test.addSecondaryIndexToExistingTable();<br /> test.removeSecondaryIndexToExistingTable();<br /> test.readAllRowsFromSecondaryIndex();<br /> test.readFilteredRowsFromSecondaryIndex();<br /> System.out.println("Done!");<br /> }<br /> }