View Javadoc

1   /**
2    *
3    * Licensed to the Apache Software Foundation (ASF) under one
4    * or more contributor license agreements.  See the NOTICE file
5    * distributed with this work for additional information
6    * regarding copyright ownership.  The ASF licenses this file
7    * to you under the Apache License, Version 2.0 (the
8    * "License"); you may not use this file except in compliance
9    * with the License.  You may obtain a copy of the License at
10   *
11   *     http://www.apache.org/licenses/LICENSE-2.0
12   *
13   * Unless required by applicable law or agreed to in writing, software
14   * distributed under the License is distributed on an "AS IS" BASIS,
15   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16   * See the License for the specific language governing permissions and
17   * limitations under the License.
18   */
19  package org.apache.hadoop.hbase.mapred;
20  
21  import java.io.IOException;
22  import java.util.ArrayList;
23  
24  import org.apache.hadoop.classification.InterfaceAudience;
25  import org.apache.hadoop.classification.InterfaceStability;
26  import org.apache.hadoop.hbase.Cell;
27  import org.apache.hadoop.hbase.CellUtil;
28  import org.apache.hadoop.hbase.KeyValue;
29  import org.apache.hadoop.hbase.client.Result;
30  import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
31  import org.apache.hadoop.hbase.util.Bytes;
32  import org.apache.hadoop.mapred.JobConf;
33  import org.apache.hadoop.mapred.MapReduceBase;
34  import org.apache.hadoop.mapred.OutputCollector;
35  import org.apache.hadoop.mapred.Reporter;
36  
37  
38  /**
39   * Extract grouping columns from input record
40   */
41  @Deprecated
42  @InterfaceAudience.Public
43  @InterfaceStability.Stable
44  public class GroupingTableMap
45  extends MapReduceBase
46  implements TableMap<ImmutableBytesWritable,Result> {
47  
48    /**
49     * JobConf parameter to specify the columns used to produce the key passed to
50     * collect from the map phase
51     */
52    public static final String GROUP_COLUMNS =
53      "hbase.mapred.groupingtablemap.columns";
54  
55    protected byte [][] columns;
56  
57    /**
58     * Use this before submitting a TableMap job. It will appropriately set up the
59     * JobConf.
60     *
61     * @param table table to be processed
62     * @param columns space separated list of columns to fetch
63     * @param groupColumns space separated list of columns used to form the key
64     * used in collect
65     * @param mapper map class
66     * @param job job configuration object
67     */
68    @SuppressWarnings("unchecked")
69    public static void initJob(String table, String columns, String groupColumns,
70      Class<? extends TableMap> mapper, JobConf job) {
71  
72      TableMapReduceUtil.initTableMapJob(table, columns, mapper,
73          ImmutableBytesWritable.class, Result.class, job);
74      job.set(GROUP_COLUMNS, groupColumns);
75    }
76  
77    @Override
78    public void configure(JobConf job) {
79      super.configure(job);
80      String[] cols = job.get(GROUP_COLUMNS, "").split(" ");
81      columns = new byte[cols.length][];
82      for(int i = 0; i < cols.length; i++) {
83        columns[i] = Bytes.toBytes(cols[i]);
84      }
85    }
86  
87    /**
88     * Extract the grouping columns from value to construct a new key.
89     *
90     * Pass the new key and value to reduce.
91     * If any of the grouping columns are not found in the value, the record is skipped.
92     * @param key
93     * @param value
94     * @param output
95     * @param reporter
96     * @throws IOException
97     */
98    public void map(ImmutableBytesWritable key, Result value,
99        OutputCollector<ImmutableBytesWritable,Result> output,
100       Reporter reporter) throws IOException {
101 
102     byte[][] keyVals = extractKeyValues(value);
103     if(keyVals != null) {
104       ImmutableBytesWritable tKey = createGroupKey(keyVals);
105       output.collect(tKey, value);
106     }
107   }
108 
109   /**
110    * Extract columns values from the current record. This method returns
111    * null if any of the columns are not found.
112    *
113    * Override this method if you want to deal with nulls differently.
114    *
115    * @param r
116    * @return array of byte values
117    */
118   protected byte[][] extractKeyValues(Result r) {
119     byte[][] keyVals = null;
120     ArrayList<byte[]> foundList = new ArrayList<byte[]>();
121     int numCols = columns.length;
122     if (numCols > 0) {
123       for (Cell value: r.listCells()) {
124         byte [] column = KeyValue.makeColumn(CellUtil.cloneFamily(value),
125             CellUtil.cloneQualifier(value));
126         for (int i = 0; i < numCols; i++) {
127           if (Bytes.equals(column, columns[i])) {
128             foundList.add(CellUtil.cloneValue(value));
129             break;
130           }
131         }
132       }
133       if(foundList.size() == numCols) {
134         keyVals = foundList.toArray(new byte[numCols][]);
135       }
136     }
137     return keyVals;
138   }
139 
140   /**
141    * Create a key by concatenating multiple column values.
142    * Override this function in order to produce different types of keys.
143    *
144    * @param vals
145    * @return key generated by concatenating multiple column values
146    */
147   protected ImmutableBytesWritable createGroupKey(byte[][] vals) {
148     if(vals == null) {
149       return null;
150     }
151     StringBuilder sb =  new StringBuilder();
152     for(int i = 0; i < vals.length; i++) {
153       if(i > 0) {
154         sb.append(" ");
155       }
156       sb.append(Bytes.toString(vals[i]));
157     }
158     return new ImmutableBytesWritable(Bytes.toBytes(sb.toString()));
159   }
160 }