[9] | 1 | /** |
---|
| 2 | * Program: HBaseRecord.java |
---|
| 3 | * Editor: Waue Chen |
---|
| 4 | * From : NCHC. Taiwn |
---|
| 5 | * Last Update Date: 06/01/2008 |
---|
[7] | 6 | */ |
---|
| 7 | |
---|
| 8 | /** |
---|
[9] | 9 | * Purpose : |
---|
| 10 | * Parse your record and then store in HBase. |
---|
| 11 | * |
---|
| 12 | * HowToUse : |
---|
| 13 | * Make sure Hadoop file system and Hbase are running correctly. |
---|
| 14 | * 1. put test.txt in t1 directory which content is |
---|
| 15 | --------------- |
---|
| 16 | name:locate:years |
---|
| 17 | waue:taiwan:1981 |
---|
| 18 | shellon:taiwan:1981 |
---|
| 19 | --------------- |
---|
| 20 | * 2. hadoop_root/$ bin/hadoop dfs -put t1 t1 |
---|
| 21 | * 3. hbase_root/$ bin/hbase shell |
---|
| 22 | * 4. hql > create table t1_table("person"); |
---|
| 23 | * 5. Come to Eclipse and run this code, and we will let database as that |
---|
| 24 | t1_table -> person |
---|
| 25 | ---------------- |
---|
| 26 | | name | locate | years | |
---|
| 27 | | waue | taiwan | 1981 | |
---|
| 28 | | shellon | taiwan | 1981 | |
---|
| 29 | ---------------- |
---|
| 30 | * Check Result: |
---|
| 31 | * Go to hbase console, type : |
---|
| 32 | * hql > select * from t1_table; |
---|
[7] | 33 | 08/06/06 12:20:48 INFO hbase.HTable: Creating scanner over t1_table starting at key |
---|
| 34 | +-------------------------+-------------------------+-------------------------+ |
---|
| 35 | | Row | Column | Cell | |
---|
| 36 | +-------------------------+-------------------------+-------------------------+ |
---|
| 37 | | 0 | person:locate | locate | |
---|
| 38 | +-------------------------+-------------------------+-------------------------+ |
---|
| 39 | | 0 | person:name | name | |
---|
| 40 | +-------------------------+-------------------------+-------------------------+ |
---|
| 41 | | 0 | person:years | years | |
---|
| 42 | +-------------------------+-------------------------+-------------------------+ |
---|
| 43 | | 19 | person:locate | taiwan | |
---|
| 44 | +-------------------------+-------------------------+-------------------------+ |
---|
| 45 | | 19 | person:name | waue | |
---|
| 46 | +-------------------------+-------------------------+-------------------------+ |
---|
| 47 | | 19 | person:years | 1981 | |
---|
| 48 | +-------------------------+-------------------------+-------------------------+ |
---|
| 49 | | 36 | person:locate | taiwan | |
---|
| 50 | +-------------------------+-------------------------+-------------------------+ |
---|
| 51 | | 36 | person:name | shellon | |
---|
| 52 | +-------------------------+-------------------------+-------------------------+ |
---|
| 53 | | 36 | person:years | 1981 | |
---|
| 54 | +-------------------------+-------------------------+-------------------------+ |
---|
| 55 | 3 row(s) in set. (0.04 sec) |
---|
[9] | 56 | */ |
---|
| 57 | |
---|
| 58 | |
---|
| 59 | |
---|
| 60 | |
---|
| 61 | package tw.org.nchc.code; |
---|
| 62 | |
---|
| 63 | import java.io.IOException; |
---|
| 64 | import java.util.Iterator; |
---|
| 65 | |
---|
| 66 | import org.apache.hadoop.fs.Path; |
---|
| 67 | import org.apache.hadoop.hbase.io.ImmutableBytesWritable; |
---|
| 68 | import org.apache.hadoop.hbase.mapred.TableReduce; |
---|
| 69 | import org.apache.hadoop.io.LongWritable; |
---|
| 70 | import org.apache.hadoop.io.MapWritable; |
---|
| 71 | import org.apache.hadoop.io.Text; |
---|
| 72 | import org.apache.hadoop.mapred.JobClient; |
---|
| 73 | import org.apache.hadoop.mapred.JobConf; |
---|
| 74 | import org.apache.hadoop.mapred.OutputCollector; |
---|
| 75 | import org.apache.hadoop.mapred.Reporter; |
---|
| 76 | import org.apache.hadoop.mapred.lib.IdentityMapper; |
---|
| 77 | import org.apache.hadoop.mapred.lib.IdentityReducer; |
---|
| 78 | |
---|
| 79 | |
---|
[7] | 80 | public class HBaseRecord { |
---|
| 81 | |
---|
| 82 | /* Denify parameter */ |
---|
| 83 | // one column family: person; three column qualifier: name,locate,years |
---|
[9] | 84 | static private String baseId1 ="person:name"; |
---|
[7] | 85 | static private String baseId2 ="person:locate"; |
---|
| 86 | static private String baseId3 ="person:years"; |
---|
| 87 | //split character |
---|
| 88 | static private String sp = ":"; |
---|
| 89 | // file path in hadoop file system (not phisical file system) |
---|
| 90 | String file_path = "/user/waue/t1"; |
---|
| 91 | // Hbase table name |
---|
| 92 | String table_name = "t1_table"; |
---|
| 93 | // setup MapTask and Reduce Task |
---|
| 94 | int mapTasks = 1; |
---|
| 95 | int reduceTasks = 1; |
---|
| 96 | |
---|
| 97 | private static class ReduceClass extends TableReduce<LongWritable, Text> { |
---|
| 98 | |
---|
| 99 | // Column id is created dymanically, |
---|
| 100 | private static final Text col_name = new Text(baseId1); |
---|
| 101 | private static final Text col_local = new Text(baseId2); |
---|
| 102 | private static final Text col_year = new Text(baseId3); |
---|
| 103 | |
---|
| 104 | // this map holds the columns per row |
---|
| 105 | private MapWritable map = new MapWritable(); |
---|
| 106 | |
---|
| 107 | // on this sample, map is nonuse, we use reduce to handle |
---|
| 108 | public void reduce(LongWritable key, Iterator<Text> values, |
---|
| 109 | OutputCollector<Text, MapWritable> output, Reporter reporter) |
---|
| 110 | throws IOException { |
---|
| 111 | |
---|
| 112 | // values.next().getByte() can get value and transfer to byte form, there is an other way that let decode() |
---|
| 113 | // to substitude getByte() |
---|
| 114 | String stro = new String(values.next().getBytes()); |
---|
| 115 | String str[] = stro.split(sp); |
---|
| 116 | byte b_local[] = str[0].getBytes(); |
---|
| 117 | byte b_name[] = str[1].getBytes(); |
---|
| 118 | byte b_year[] = str[2].getBytes(); |
---|
| 119 | |
---|
| 120 | // contents must be ImmutableBytesWritable |
---|
| 121 | ImmutableBytesWritable w_local = new ImmutableBytesWritable( b_local); |
---|
| 122 | ImmutableBytesWritable w_name = new ImmutableBytesWritable( b_name ); |
---|
| 123 | ImmutableBytesWritable w_year = new ImmutableBytesWritable( b_year ); |
---|
| 124 | |
---|
| 125 | // populate the current row |
---|
| 126 | map.clear(); |
---|
| 127 | map.put(col_name, w_local); |
---|
| 128 | map.put(col_local, w_name); |
---|
| 129 | map.put(col_year, w_year); |
---|
| 130 | |
---|
| 131 | // add the row with the key as the row id |
---|
| 132 | output.collect(new Text(key.toString()), map); |
---|
| 133 | } |
---|
| 134 | } |
---|
| 135 | |
---|
| 136 | private HBaseRecord() { |
---|
| 137 | } |
---|
| 138 | |
---|
| 139 | /** |
---|
| 140 | * Runs the demo. |
---|
| 141 | */ |
---|
| 142 | public static void main(String[] args) throws IOException { |
---|
| 143 | // which path of input files in Hadoop file system |
---|
| 144 | |
---|
| 145 | HBaseRecord setup = new HBaseRecord(); |
---|
| 146 | JobConf conf = new JobConf(HBaseRecord.class); |
---|
| 147 | |
---|
| 148 | //Job name; you can modify to any you like |
---|
| 149 | conf.setJobName("NCHC_PersonDataBase"); |
---|
| 150 | |
---|
| 151 | // Hbase table name must be correct , in our profile is t1_table |
---|
| 152 | TableReduce.initJob(setup.table_name, ReduceClass.class, conf); |
---|
| 153 | |
---|
| 154 | // below are map-reduce profile |
---|
| 155 | conf.setNumMapTasks(setup.mapTasks); |
---|
| 156 | conf.setNumReduceTasks(setup.reduceTasks); |
---|
| 157 | conf.setInputPath(new Path(setup.file_path)); |
---|
| 158 | conf.setMapperClass(IdentityMapper.class); |
---|
| 159 | conf.setCombinerClass(IdentityReducer.class); |
---|
| 160 | conf.setReducerClass(ReduceClass.class); |
---|
| 161 | JobClient.runJob(conf); |
---|
| 162 | } |
---|
| 163 | } |
---|