hbase RowFilter如何根据rowkey查询以及实例实现代码

阅读更多

问题导读：
1.如何根据rowkey模糊查询？
2.如何使用Comparator过滤rowkey？
3.如何查询rowkey中包含有某字符串的数据?

RowFilter用于过滤row key

Operator	Description
LESS	小于
LESS_OR_EQUAL	小于等于
[EQUAL	等于
NOT_EQUAL	不等于
GREATER_OR_EQUAL	大于等于
GREATER	大于
NO_OP	排除所有

 

Comparator	Description
BinaryComparator	使用Bytes.compareTo()比较
BinaryPrefixComparator	和BinaryComparator差不多，从前面开始比较
NullComparator	Does not compare against an actual value but whether a given one is null, or not  null.
BitComparator	Performs a bitwise comparison, providing a BitwiseOp class with OR, and XOR operators.
RegexStringComparator	正则表达式
SubstringComparator	把数据当成字符串，用contains()来判断

 

1. import java.io.IOException;
3. import org.apache.hadoop.conf.Configuration;
4. import org.apache.hadoop.hbase.HBaseConfiguration;
5. import org.apache.hadoop.hbase.HColumnDescriptor;
6. import org.apache.hadoop.hbase.HTableDescriptor;
7. import org.apache.hadoop.hbase.client.HBaseAdmin;
8. import org.apache.hadoop.hbase.client.HTable;
9. import org.apache.hadoop.hbase.client.Put;
10. import org.apache.hadoop.hbase.client.Result;
11. import org.apache.hadoop.hbase.client.ResultScanner;
12. import org.apache.hadoop.hbase.client.Scan;
13. import org.apache.hadoop.hbase.filter.BinaryComparator;
14. import org.apache.hadoop.hbase.filter.BinaryPrefixComparator;
15. import org.apache.hadoop.hbase.filter.CompareFilter;
16. import org.apache.hadoop.hbase.filter.Filter;
17. import org.apache.hadoop.hbase.filter.RegexStringComparator;
18. import org.apache.hadoop.hbase.filter.RowFilter;
19. import org.apache.hadoop.hbase.filter.SubstringComparator;
21. public class TestHbaseRowFilter {
22.         String tableName = "test_row_filter";
23.         Configuration config = HBaseConfiguration.create();
25.         
26.         public void testRowFilter() throws IOException {
28.                 HTable table = new HTable(config, tableName);
29.                 Scan scan = new Scan();
31.                 System.out.println("小于等于row010的行");
32.                 Filter filter1 = new RowFilter(CompareFilter.CompareOp.LESS_OR_EQUAL,
33.                                 new BinaryComparator("row010".getBytes()));
34.                 scan.setFilter(filter1);
35.                 ResultScanner scanner1 = table.getScanner(scan);
36.                 for (Result res : scanner1) {
37.                         System.out.println(res);
38.                 }
39.                 scanner1.close();
41.                 System.out.println("正则获取结尾为5的行");
42.                 Filter filter2 = new RowFilter(CompareFilter.CompareOp.EQUAL,
43.                                 new RegexStringComparator(".*5[        DISCUZ_CODE_0        ]quot;));
44.                 scan.setFilter(filter2);
45.                 ResultScanner scanner2 = table.getScanner(scan);
46.                 for (Result res : scanner2) {
47.                         System.out.println(res);
48.                 }
49.                 scanner2.close();
51.                 System.out.println("包含有5的行");
52.                 Filter filter3 = new RowFilter(CompareFilter.CompareOp.EQUAL,
53.                                 new SubstringComparator("5"));
54.                 scan.setFilter(filter3);
55.                 ResultScanner scanner3 = table.getScanner(scan);
56.                 for (Result res : scanner3) {
57.                         System.out.println(res);
58.                 }
59.                 scanner3.close();
61.                 System.out.println("开头是row01的");
62.                 Filter filter4 = new RowFilter(CompareFilter.CompareOp.EQUAL,
63.                                 new BinaryPrefixComparator("row01".getBytes()));
64.                 scan.setFilter(filter4);
65.                 ResultScanner scanner4 = table.getScanner(scan);
66.                 for (Result res : scanner4) {
67.                         System.out.println(res);
68.                 }
69.                 scanner3.close();
70.         }
72.         
73.         public void init() {
74.                 // 创建表和初始化数据
75.                 try {
76.                         HBaseAdmin admin = new HBaseAdmin(config);
77.                         if (!admin.tableExists(tableName)) {
78.                                 HTableDescriptor htd = new HTableDescriptor(tableName);
79.                                 HColumnDescriptor hcd1 = new HColumnDescriptor("data");
80.                                 htd.addFamily(hcd1);
81.                                 HColumnDescriptor hcd2 = new HColumnDescriptor("url");
82.                                 htd.addFamily(hcd2);
84.                                 admin.createTable(htd);
85.                         }
87.                         HTable table = new HTable(config, tableName);
89.                         table.setAutoFlush(false);
90.                         int count = 50;
91.                         for (int i = 1; i <= count; ++i) {
92.                                 Put p = new Put(String.format("rowd", i).getBytes());
93.                                 p.add("data".getBytes(), String.format("cold", i % 10)
94.                                                 .getBytes(), String.format("datad", i).getBytes());
95.                                 p.add("url".getBytes(), String.format("cold", i % 10)
96.                                                 .getBytes(), String.format("urld", i).getBytes());
97.                                 table.put(p);
98.                         }
99.                         table.close();
101.                 } catch (IOException e) {
102.                         e.printStackTrace();
103.                 }
104.         }
106.         
107.         public static void main(String[] args) throws IOException {
108.                 TestHbaseRowFilter test = new TestHbaseRowFilter();
109.                 test.init();
110.                 test.testRowFilter();
111.         }
113. }

复制代码

输出结果

1. 小于等于row010的行
2. keyvalues={row001/data:col1/1364133382268/Put/vlen=7, row001/url:col1/1364133382268/Put/vlen=6}
3. keyvalues={row002/data:col2/1364133382268/Put/vlen=7, row002/url:col2/1364133382268/Put/vlen=6}
4. keyvalues={row003/data:col3/1364133382268/Put/vlen=7, row003/url:col3/1364133382268/Put/vlen=6}
5. keyvalues={row004/data:col4/1364133382268/Put/vlen=7, row004/url:col4/1364133382268/Put/vlen=6}
6. keyvalues={row005/data:col5/1364133382268/Put/vlen=7, row005/url:col5/1364133382268/Put/vlen=6}
7. keyvalues={row006/data:col6/1364133382268/Put/vlen=7, row006/url:col6/1364133382268/Put/vlen=6}
8. keyvalues={row007/data:col7/1364133382268/Put/vlen=7, row007/url:col7/1364133382268/Put/vlen=6}
9. keyvalues={row008/data:col8/1364133382268/Put/vlen=7, row008/url:col8/1364133382268/Put/vlen=6}
10. keyvalues={row009/data:col9/1364133382268/Put/vlen=7, row009/url:col9/1364133382268/Put/vlen=6}
11. keyvalues={row010/data:col0/1364133382268/Put/vlen=7, row010/url:col0/1364133382268/Put/vlen=6}
12. 正则获取结尾为5的行
13. keyvalues={row005/data:col5/1364133382268/Put/vlen=7, row005/url:col5/1364133382268/Put/vlen=6}
14. keyvalues={row015/data:col5/1364133382268/Put/vlen=7, row015/url:col5/1364133382268/Put/vlen=6}
15. keyvalues={row025/data:col5/1364133382268/Put/vlen=7, row025/url:col5/1364133382268/Put/vlen=6}
16. keyvalues={row035/data:col5/1364133382268/Put/vlen=7, row035/url:col5/1364133382268/Put/vlen=6}
17. keyvalues={row045/data:col5/1364133382268/Put/vlen=7, row045/url:col5/1364133382268/Put/vlen=6}
18. 包行有5的行
19. keyvalues={row005/data:col5/1364133382268/Put/vlen=7, row005/url:col5/1364133382268/Put/vlen=6}
20. keyvalues={row015/data:col5/1364133382268/Put/vlen=7, row015/url:col5/1364133382268/Put/vlen=6}
21. keyvalues={row025/data:col5/1364133382268/Put/vlen=7, row025/url:col5/1364133382268/Put/vlen=6}
22. keyvalues={row035/data:col5/1364133382268/Put/vlen=7, row035/url:col5/1364133382268/Put/vlen=6}
23. keyvalues={row045/data:col5/1364133382268/Put/vlen=7, row045/url:col5/1364133382268/Put/vlen=6}
24. keyvalues={row050/data:col0/1364133382268/Put/vlen=7, row050/url:col0/1364133382268/Put/vlen=6}
25. 开头是row01的
26. keyvalues={row010/data:col0/1364133382268/Put/vlen=7, row010/url:col0/1364133382268/Put/vlen=6}
27. keyvalues={row011/data:col1/1364133382268/Put/vlen=7, row011/url:col1/1364133382268/Put/vlen=6}
28. keyvalues={row012/data:col2/1364133382268/Put/vlen=7, row012/url:col2/1364133382268/Put/vlen=6}
29. keyvalues={row013/data:col3/1364133382268/Put/vlen=7, row013/url:col3/1364133382268/Put/vlen=6}
30. keyvalues={row014/data:col4/1364133382268/Put/vlen=7, row014/url:col4/1364133382268/Put/vlen=6}
31. keyvalues={row015/data:col5/1364133382268/Put/vlen=7, row015/url:col5/1364133382268/Put/vlen=6}
32. keyvalues={row016/data:col6/1364133382268/Put/vlen=7, row016/url:col6/1364133382268/Put/vlen=6}
33. keyvalues={row017/data:col7/1364133382268/Put/vlen=7, row017/url:col7/1364133382268/Put/vlen=6}
34. keyvalues={row018/data:col8/1364133382268/Put/vlen=7, row018/url:col8/1364133382268/Put/vlen=6}
35. keyvalues={row019/data:col9/1364133382268/Put/vlen=7, row019/url:col9/1364133382268/Put/vlen=6}

 

 

http://blog.sina.com.cn/s/blog_68674da70102v6q4.html