blob: 38ea9abb2c89d3fba8f498862a2c87baeaf86ca1 [file] [log] [blame]
<!DOCTYPE HTML>
<html lang="en">
<head>
<!-- Generated by javadoc (17) -->
<title>Source code</title>
<meta name="viewport" content="width=device-width, initial-scale=1">
<meta name="description" content="source: package: org.apache.hadoop.hbase.mapreduce, class: GroupingTableMapper">
<meta name="generator" content="javadoc/SourceToHTMLConverter">
<link rel="stylesheet" type="text/css" href="../../../../../../stylesheet.css" title="Style">
</head>
<body class="source-page">
<main role="main">
<div class="source-container">
<pre><span class="source-line-no">001</span><span id="line-1">/*</span>
<span class="source-line-no">002</span><span id="line-2"> * Licensed to the Apache Software Foundation (ASF) under one</span>
<span class="source-line-no">003</span><span id="line-3"> * or more contributor license agreements. See the NOTICE file</span>
<span class="source-line-no">004</span><span id="line-4"> * distributed with this work for additional information</span>
<span class="source-line-no">005</span><span id="line-5"> * regarding copyright ownership. The ASF licenses this file</span>
<span class="source-line-no">006</span><span id="line-6"> * to you under the Apache License, Version 2.0 (the</span>
<span class="source-line-no">007</span><span id="line-7"> * "License"); you may not use this file except in compliance</span>
<span class="source-line-no">008</span><span id="line-8"> * with the License. You may obtain a copy of the License at</span>
<span class="source-line-no">009</span><span id="line-9"> *</span>
<span class="source-line-no">010</span><span id="line-10"> * http://www.apache.org/licenses/LICENSE-2.0</span>
<span class="source-line-no">011</span><span id="line-11"> *</span>
<span class="source-line-no">012</span><span id="line-12"> * Unless required by applicable law or agreed to in writing, software</span>
<span class="source-line-no">013</span><span id="line-13"> * distributed under the License is distributed on an "AS IS" BASIS,</span>
<span class="source-line-no">014</span><span id="line-14"> * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
<span class="source-line-no">015</span><span id="line-15"> * See the License for the specific language governing permissions and</span>
<span class="source-line-no">016</span><span id="line-16"> * limitations under the License.</span>
<span class="source-line-no">017</span><span id="line-17"> */</span>
<span class="source-line-no">018</span><span id="line-18">package org.apache.hadoop.hbase.mapreduce;</span>
<span class="source-line-no">019</span><span id="line-19"></span>
<span class="source-line-no">020</span><span id="line-20">import java.io.IOException;</span>
<span class="source-line-no">021</span><span id="line-21">import java.util.ArrayList;</span>
<span class="source-line-no">022</span><span id="line-22">import org.apache.hadoop.conf.Configurable;</span>
<span class="source-line-no">023</span><span id="line-23">import org.apache.hadoop.conf.Configuration;</span>
<span class="source-line-no">024</span><span id="line-24">import org.apache.hadoop.hbase.Cell;</span>
<span class="source-line-no">025</span><span id="line-25">import org.apache.hadoop.hbase.CellUtil;</span>
<span class="source-line-no">026</span><span id="line-26">import org.apache.hadoop.hbase.client.Result;</span>
<span class="source-line-no">027</span><span id="line-27">import org.apache.hadoop.hbase.client.Scan;</span>
<span class="source-line-no">028</span><span id="line-28">import org.apache.hadoop.hbase.io.ImmutableBytesWritable;</span>
<span class="source-line-no">029</span><span id="line-29">import org.apache.hadoop.hbase.util.Bytes;</span>
<span class="source-line-no">030</span><span id="line-30">import org.apache.hadoop.mapreduce.Job;</span>
<span class="source-line-no">031</span><span id="line-31">import org.apache.yetus.audience.InterfaceAudience;</span>
<span class="source-line-no">032</span><span id="line-32"></span>
<span class="source-line-no">033</span><span id="line-33">/**</span>
<span class="source-line-no">034</span><span id="line-34"> * Extract grouping columns from input record.</span>
<span class="source-line-no">035</span><span id="line-35"> */</span>
<span class="source-line-no">036</span><span id="line-36">@InterfaceAudience.Public</span>
<span class="source-line-no">037</span><span id="line-37">public class GroupingTableMapper extends TableMapper&lt;ImmutableBytesWritable, Result&gt;</span>
<span class="source-line-no">038</span><span id="line-38"> implements Configurable {</span>
<span class="source-line-no">039</span><span id="line-39"></span>
<span class="source-line-no">040</span><span id="line-40"> /**</span>
<span class="source-line-no">041</span><span id="line-41"> * JobConf parameter to specify the columns used to produce the key passed to collect from the map</span>
<span class="source-line-no">042</span><span id="line-42"> * phase.</span>
<span class="source-line-no">043</span><span id="line-43"> */</span>
<span class="source-line-no">044</span><span id="line-44"> public static final String GROUP_COLUMNS = "hbase.mapred.groupingtablemap.columns";</span>
<span class="source-line-no">045</span><span id="line-45"></span>
<span class="source-line-no">046</span><span id="line-46"> /** The grouping columns. */</span>
<span class="source-line-no">047</span><span id="line-47"> protected byte[][] columns;</span>
<span class="source-line-no">048</span><span id="line-48"> /** The current configuration. */</span>
<span class="source-line-no">049</span><span id="line-49"> private Configuration conf = null;</span>
<span class="source-line-no">050</span><span id="line-50"></span>
<span class="source-line-no">051</span><span id="line-51"> /**</span>
<span class="source-line-no">052</span><span id="line-52"> * Use this before submitting a TableMap job. It will appropriately set up the job.</span>
<span class="source-line-no">053</span><span id="line-53"> * @param table The table to be processed.</span>
<span class="source-line-no">054</span><span id="line-54"> * @param scan The scan with the columns etc.</span>
<span class="source-line-no">055</span><span id="line-55"> * @param groupColumns A space separated list of columns used to form the key used in collect.</span>
<span class="source-line-no">056</span><span id="line-56"> * @param mapper The mapper class.</span>
<span class="source-line-no">057</span><span id="line-57"> * @param job The current job.</span>
<span class="source-line-no">058</span><span id="line-58"> * @throws IOException When setting up the job fails.</span>
<span class="source-line-no">059</span><span id="line-59"> */</span>
<span class="source-line-no">060</span><span id="line-60"> @SuppressWarnings("unchecked")</span>
<span class="source-line-no">061</span><span id="line-61"> public static void initJob(String table, Scan scan, String groupColumns,</span>
<span class="source-line-no">062</span><span id="line-62"> Class&lt;? extends TableMapper&gt; mapper, Job job) throws IOException {</span>
<span class="source-line-no">063</span><span id="line-63"> TableMapReduceUtil.initTableMapperJob(table, scan, mapper, ImmutableBytesWritable.class,</span>
<span class="source-line-no">064</span><span id="line-64"> Result.class, job);</span>
<span class="source-line-no">065</span><span id="line-65"> job.getConfiguration().set(GROUP_COLUMNS, groupColumns);</span>
<span class="source-line-no">066</span><span id="line-66"> }</span>
<span class="source-line-no">067</span><span id="line-67"></span>
<span class="source-line-no">068</span><span id="line-68"> /**</span>
<span class="source-line-no">069</span><span id="line-69"> * Extract the grouping columns from value to construct a new key. Pass the new key and value to</span>
<span class="source-line-no">070</span><span id="line-70"> * reduce. If any of the grouping columns are not found in the value, the record is skipped.</span>
<span class="source-line-no">071</span><span id="line-71"> * @param key The current key.</span>
<span class="source-line-no">072</span><span id="line-72"> * @param value The current value.</span>
<span class="source-line-no">073</span><span id="line-73"> * @param context The current context.</span>
<span class="source-line-no">074</span><span id="line-74"> * @throws IOException When writing the record fails.</span>
<span class="source-line-no">075</span><span id="line-75"> * @throws InterruptedException When the job is aborted.</span>
<span class="source-line-no">076</span><span id="line-76"> */</span>
<span class="source-line-no">077</span><span id="line-77"> @Override</span>
<span class="source-line-no">078</span><span id="line-78"> public void map(ImmutableBytesWritable key, Result value, Context context)</span>
<span class="source-line-no">079</span><span id="line-79"> throws IOException, InterruptedException {</span>
<span class="source-line-no">080</span><span id="line-80"> byte[][] keyVals = extractKeyValues(value);</span>
<span class="source-line-no">081</span><span id="line-81"> if (keyVals != null) {</span>
<span class="source-line-no">082</span><span id="line-82"> ImmutableBytesWritable tKey = createGroupKey(keyVals);</span>
<span class="source-line-no">083</span><span id="line-83"> context.write(tKey, value);</span>
<span class="source-line-no">084</span><span id="line-84"> }</span>
<span class="source-line-no">085</span><span id="line-85"> }</span>
<span class="source-line-no">086</span><span id="line-86"></span>
<span class="source-line-no">087</span><span id="line-87"> /**</span>
<span class="source-line-no">088</span><span id="line-88"> * Extract columns values from the current record. This method returns null if any of the columns</span>
<span class="source-line-no">089</span><span id="line-89"> * are not found.</span>
<span class="source-line-no">090</span><span id="line-90"> * &lt;p&gt;</span>
<span class="source-line-no">091</span><span id="line-91"> * Override this method if you want to deal with nulls differently.</span>
<span class="source-line-no">092</span><span id="line-92"> * @param r The current values.</span>
<span class="source-line-no">093</span><span id="line-93"> * @return Array of byte values.</span>
<span class="source-line-no">094</span><span id="line-94"> */</span>
<span class="source-line-no">095</span><span id="line-95"> protected byte[][] extractKeyValues(Result r) {</span>
<span class="source-line-no">096</span><span id="line-96"> byte[][] keyVals = null;</span>
<span class="source-line-no">097</span><span id="line-97"> ArrayList&lt;byte[]&gt; foundList = new ArrayList&lt;&gt;();</span>
<span class="source-line-no">098</span><span id="line-98"> int numCols = columns.length;</span>
<span class="source-line-no">099</span><span id="line-99"> if (numCols &gt; 0) {</span>
<span class="source-line-no">100</span><span id="line-100"> for (Cell value : r.listCells()) {</span>
<span class="source-line-no">101</span><span id="line-101"> byte[] column =</span>
<span class="source-line-no">102</span><span id="line-102"> CellUtil.makeColumn(CellUtil.cloneFamily(value), CellUtil.cloneQualifier(value));</span>
<span class="source-line-no">103</span><span id="line-103"> for (int i = 0; i &lt; numCols; i++) {</span>
<span class="source-line-no">104</span><span id="line-104"> if (Bytes.equals(column, columns[i])) {</span>
<span class="source-line-no">105</span><span id="line-105"> foundList.add(CellUtil.cloneValue(value));</span>
<span class="source-line-no">106</span><span id="line-106"> break;</span>
<span class="source-line-no">107</span><span id="line-107"> }</span>
<span class="source-line-no">108</span><span id="line-108"> }</span>
<span class="source-line-no">109</span><span id="line-109"> }</span>
<span class="source-line-no">110</span><span id="line-110"> if (foundList.size() == numCols) {</span>
<span class="source-line-no">111</span><span id="line-111"> keyVals = foundList.toArray(new byte[numCols][]);</span>
<span class="source-line-no">112</span><span id="line-112"> }</span>
<span class="source-line-no">113</span><span id="line-113"> }</span>
<span class="source-line-no">114</span><span id="line-114"> return keyVals;</span>
<span class="source-line-no">115</span><span id="line-115"> }</span>
<span class="source-line-no">116</span><span id="line-116"></span>
<span class="source-line-no">117</span><span id="line-117"> /**</span>
<span class="source-line-no">118</span><span id="line-118"> * Create a key by concatenating multiple column values.</span>
<span class="source-line-no">119</span><span id="line-119"> * &lt;p&gt;</span>
<span class="source-line-no">120</span><span id="line-120"> * Override this function in order to produce different types of keys.</span>
<span class="source-line-no">121</span><span id="line-121"> * @param vals The current key/values.</span>
<span class="source-line-no">122</span><span id="line-122"> * @return A key generated by concatenating multiple column values.</span>
<span class="source-line-no">123</span><span id="line-123"> */</span>
<span class="source-line-no">124</span><span id="line-124"> protected ImmutableBytesWritable createGroupKey(byte[][] vals) {</span>
<span class="source-line-no">125</span><span id="line-125"> if (vals == null) {</span>
<span class="source-line-no">126</span><span id="line-126"> return null;</span>
<span class="source-line-no">127</span><span id="line-127"> }</span>
<span class="source-line-no">128</span><span id="line-128"> StringBuilder sb = new StringBuilder();</span>
<span class="source-line-no">129</span><span id="line-129"> for (int i = 0; i &lt; vals.length; i++) {</span>
<span class="source-line-no">130</span><span id="line-130"> if (i &gt; 0) {</span>
<span class="source-line-no">131</span><span id="line-131"> sb.append(" ");</span>
<span class="source-line-no">132</span><span id="line-132"> }</span>
<span class="source-line-no">133</span><span id="line-133"> sb.append(Bytes.toString(vals[i]));</span>
<span class="source-line-no">134</span><span id="line-134"> }</span>
<span class="source-line-no">135</span><span id="line-135"> return new ImmutableBytesWritable(Bytes.toBytesBinary(sb.toString()));</span>
<span class="source-line-no">136</span><span id="line-136"> }</span>
<span class="source-line-no">137</span><span id="line-137"></span>
<span class="source-line-no">138</span><span id="line-138"> /**</span>
<span class="source-line-no">139</span><span id="line-139"> * Returns the current configuration.</span>
<span class="source-line-no">140</span><span id="line-140"> * @return The current configuration.</span>
<span class="source-line-no">141</span><span id="line-141"> * @see org.apache.hadoop.conf.Configurable#getConf()</span>
<span class="source-line-no">142</span><span id="line-142"> */</span>
<span class="source-line-no">143</span><span id="line-143"> @Override</span>
<span class="source-line-no">144</span><span id="line-144"> public Configuration getConf() {</span>
<span class="source-line-no">145</span><span id="line-145"> return conf;</span>
<span class="source-line-no">146</span><span id="line-146"> }</span>
<span class="source-line-no">147</span><span id="line-147"></span>
<span class="source-line-no">148</span><span id="line-148"> /**</span>
<span class="source-line-no">149</span><span id="line-149"> * Sets the configuration. This is used to set up the grouping details.</span>
<span class="source-line-no">150</span><span id="line-150"> * @param configuration The configuration to set.</span>
<span class="source-line-no">151</span><span id="line-151"> * @see org.apache.hadoop.conf.Configurable#setConf( org.apache.hadoop.conf.Configuration)</span>
<span class="source-line-no">152</span><span id="line-152"> */</span>
<span class="source-line-no">153</span><span id="line-153"> @Override</span>
<span class="source-line-no">154</span><span id="line-154"> public void setConf(Configuration configuration) {</span>
<span class="source-line-no">155</span><span id="line-155"> this.conf = configuration;</span>
<span class="source-line-no">156</span><span id="line-156"> String[] cols = conf.get(GROUP_COLUMNS, "").split(" ");</span>
<span class="source-line-no">157</span><span id="line-157"> columns = new byte[cols.length][];</span>
<span class="source-line-no">158</span><span id="line-158"> for (int i = 0; i &lt; cols.length; i++) {</span>
<span class="source-line-no">159</span><span id="line-159"> columns[i] = Bytes.toBytes(cols[i]);</span>
<span class="source-line-no">160</span><span id="line-160"> }</span>
<span class="source-line-no">161</span><span id="line-161"> }</span>
<span class="source-line-no">162</span><span id="line-162"></span>
<span class="source-line-no">163</span><span id="line-163">}</span>
</pre>
</div>
</main>
</body>
</html>