http://git-wip-us.apache.org/repos/asf/hbase-site/blob/aea328be/apidocs/src-html/org/apache/hadoop/hbase/mapred/GroupingTableMap.html ---------------------------------------------------------------------- diff --git a/apidocs/src-html/org/apache/hadoop/hbase/mapred/GroupingTableMap.html b/apidocs/src-html/org/apache/hadoop/hbase/mapred/GroupingTableMap.html index ca89bf0..c42d8f3 100644 --- a/apidocs/src-html/org/apache/hadoop/hbase/mapred/GroupingTableMap.html +++ b/apidocs/src-html/org/apache/hadoop/hbase/mapred/GroupingTableMap.html @@ -32,137 +32,136 @@ <span class="sourceLineNo">024</span>import org.apache.yetus.audience.InterfaceAudience;<a name="line.24"></a> <span class="sourceLineNo">025</span>import org.apache.hadoop.hbase.Cell;<a name="line.25"></a> <span class="sourceLineNo">026</span>import org.apache.hadoop.hbase.CellUtil;<a name="line.26"></a> -<span class="sourceLineNo">027</span>import org.apache.hadoop.hbase.KeyValue;<a name="line.27"></a> -<span class="sourceLineNo">028</span>import org.apache.hadoop.hbase.client.Result;<a name="line.28"></a> -<span class="sourceLineNo">029</span>import org.apache.hadoop.hbase.io.ImmutableBytesWritable;<a name="line.29"></a> -<span class="sourceLineNo">030</span>import org.apache.hadoop.hbase.util.Bytes;<a name="line.30"></a> -<span class="sourceLineNo">031</span>import org.apache.hadoop.mapred.JobConf;<a name="line.31"></a> -<span class="sourceLineNo">032</span>import org.apache.hadoop.mapred.MapReduceBase;<a name="line.32"></a> -<span class="sourceLineNo">033</span>import org.apache.hadoop.mapred.OutputCollector;<a name="line.33"></a> -<span class="sourceLineNo">034</span>import org.apache.hadoop.mapred.Reporter;<a name="line.34"></a> +<span class="sourceLineNo">027</span>import org.apache.hadoop.hbase.client.Result;<a name="line.27"></a> +<span class="sourceLineNo">028</span>import org.apache.hadoop.hbase.io.ImmutableBytesWritable;<a name="line.28"></a> +<span class="sourceLineNo">029</span>import org.apache.hadoop.hbase.util.Bytes;<a name="line.29"></a> +<span class="sourceLineNo">030</span>import org.apache.hadoop.mapred.JobConf;<a name="line.30"></a> +<span class="sourceLineNo">031</span>import org.apache.hadoop.mapred.MapReduceBase;<a name="line.31"></a> +<span class="sourceLineNo">032</span>import org.apache.hadoop.mapred.OutputCollector;<a name="line.32"></a> +<span class="sourceLineNo">033</span>import org.apache.hadoop.mapred.Reporter;<a name="line.33"></a> +<span class="sourceLineNo">034</span><a name="line.34"></a> <span class="sourceLineNo">035</span><a name="line.35"></a> -<span class="sourceLineNo">036</span><a name="line.36"></a> -<span class="sourceLineNo">037</span>/**<a name="line.37"></a> -<span class="sourceLineNo">038</span> * Extract grouping columns from input record<a name="line.38"></a> -<span class="sourceLineNo">039</span> */<a name="line.39"></a> -<span class="sourceLineNo">040</span>@InterfaceAudience.Public<a name="line.40"></a> -<span class="sourceLineNo">041</span>public class GroupingTableMap<a name="line.41"></a> -<span class="sourceLineNo">042</span>extends MapReduceBase<a name="line.42"></a> -<span class="sourceLineNo">043</span>implements TableMap<ImmutableBytesWritable,Result> {<a name="line.43"></a> -<span class="sourceLineNo">044</span><a name="line.44"></a> -<span class="sourceLineNo">045</span> /**<a name="line.45"></a> -<span class="sourceLineNo">046</span> * JobConf parameter to specify the columns used to produce the key passed to<a name="line.46"></a> -<span class="sourceLineNo">047</span> * collect from the map phase<a name="line.47"></a> -<span class="sourceLineNo">048</span> */<a name="line.48"></a> -<span class="sourceLineNo">049</span> public static final String GROUP_COLUMNS =<a name="line.49"></a> -<span class="sourceLineNo">050</span> "hbase.mapred.groupingtablemap.columns";<a name="line.50"></a> -<span class="sourceLineNo">051</span><a name="line.51"></a> -<span class="sourceLineNo">052</span> protected byte [][] columns;<a name="line.52"></a> -<span class="sourceLineNo">053</span><a name="line.53"></a> -<span class="sourceLineNo">054</span> /**<a name="line.54"></a> -<span class="sourceLineNo">055</span> * Use this before submitting a TableMap job. It will appropriately set up the<a name="line.55"></a> -<span class="sourceLineNo">056</span> * JobConf.<a name="line.56"></a> -<span class="sourceLineNo">057</span> *<a name="line.57"></a> -<span class="sourceLineNo">058</span> * @param table table to be processed<a name="line.58"></a> -<span class="sourceLineNo">059</span> * @param columns space separated list of columns to fetch<a name="line.59"></a> -<span class="sourceLineNo">060</span> * @param groupColumns space separated list of columns used to form the key<a name="line.60"></a> -<span class="sourceLineNo">061</span> * used in collect<a name="line.61"></a> -<span class="sourceLineNo">062</span> * @param mapper map class<a name="line.62"></a> -<span class="sourceLineNo">063</span> * @param job job configuration object<a name="line.63"></a> -<span class="sourceLineNo">064</span> */<a name="line.64"></a> -<span class="sourceLineNo">065</span> @SuppressWarnings("unchecked")<a name="line.65"></a> -<span class="sourceLineNo">066</span> public static void initJob(String table, String columns, String groupColumns,<a name="line.66"></a> -<span class="sourceLineNo">067</span> Class<? extends TableMap> mapper, JobConf job) {<a name="line.67"></a> -<span class="sourceLineNo">068</span><a name="line.68"></a> -<span class="sourceLineNo">069</span> TableMapReduceUtil.initTableMapJob(table, columns, mapper,<a name="line.69"></a> -<span class="sourceLineNo">070</span> ImmutableBytesWritable.class, Result.class, job);<a name="line.70"></a> -<span class="sourceLineNo">071</span> job.set(GROUP_COLUMNS, groupColumns);<a name="line.71"></a> -<span class="sourceLineNo">072</span> }<a name="line.72"></a> -<span class="sourceLineNo">073</span><a name="line.73"></a> -<span class="sourceLineNo">074</span> @Override<a name="line.74"></a> -<span class="sourceLineNo">075</span> public void configure(JobConf job) {<a name="line.75"></a> -<span class="sourceLineNo">076</span> super.configure(job);<a name="line.76"></a> -<span class="sourceLineNo">077</span> String[] cols = job.get(GROUP_COLUMNS, "").split(" ");<a name="line.77"></a> -<span class="sourceLineNo">078</span> columns = new byte[cols.length][];<a name="line.78"></a> -<span class="sourceLineNo">079</span> for(int i = 0; i < cols.length; i++) {<a name="line.79"></a> -<span class="sourceLineNo">080</span> columns[i] = Bytes.toBytes(cols[i]);<a name="line.80"></a> -<span class="sourceLineNo">081</span> }<a name="line.81"></a> -<span class="sourceLineNo">082</span> }<a name="line.82"></a> -<span class="sourceLineNo">083</span><a name="line.83"></a> -<span class="sourceLineNo">084</span> /**<a name="line.84"></a> -<span class="sourceLineNo">085</span> * Extract the grouping columns from value to construct a new key.<a name="line.85"></a> -<span class="sourceLineNo">086</span> *<a name="line.86"></a> -<span class="sourceLineNo">087</span> * Pass the new key and value to reduce.<a name="line.87"></a> -<span class="sourceLineNo">088</span> * If any of the grouping columns are not found in the value, the record is skipped.<a name="line.88"></a> -<span class="sourceLineNo">089</span> * @param key<a name="line.89"></a> -<span class="sourceLineNo">090</span> * @param value<a name="line.90"></a> -<span class="sourceLineNo">091</span> * @param output<a name="line.91"></a> -<span class="sourceLineNo">092</span> * @param reporter<a name="line.92"></a> -<span class="sourceLineNo">093</span> * @throws IOException<a name="line.93"></a> -<span class="sourceLineNo">094</span> */<a name="line.94"></a> -<span class="sourceLineNo">095</span> public void map(ImmutableBytesWritable key, Result value,<a name="line.95"></a> -<span class="sourceLineNo">096</span> OutputCollector<ImmutableBytesWritable,Result> output,<a name="line.96"></a> -<span class="sourceLineNo">097</span> Reporter reporter) throws IOException {<a name="line.97"></a> -<span class="sourceLineNo">098</span><a name="line.98"></a> -<span class="sourceLineNo">099</span> byte[][] keyVals = extractKeyValues(value);<a name="line.99"></a> -<span class="sourceLineNo">100</span> if(keyVals != null) {<a name="line.100"></a> -<span class="sourceLineNo">101</span> ImmutableBytesWritable tKey = createGroupKey(keyVals);<a name="line.101"></a> -<span class="sourceLineNo">102</span> output.collect(tKey, value);<a name="line.102"></a> -<span class="sourceLineNo">103</span> }<a name="line.103"></a> -<span class="sourceLineNo">104</span> }<a name="line.104"></a> -<span class="sourceLineNo">105</span><a name="line.105"></a> -<span class="sourceLineNo">106</span> /**<a name="line.106"></a> -<span class="sourceLineNo">107</span> * Extract columns values from the current record. This method returns<a name="line.107"></a> -<span class="sourceLineNo">108</span> * null if any of the columns are not found.<a name="line.108"></a> -<span class="sourceLineNo">109</span> *<a name="line.109"></a> -<span class="sourceLineNo">110</span> * Override this method if you want to deal with nulls differently.<a name="line.110"></a> -<span class="sourceLineNo">111</span> *<a name="line.111"></a> -<span class="sourceLineNo">112</span> * @param r<a name="line.112"></a> -<span class="sourceLineNo">113</span> * @return array of byte values<a name="line.113"></a> -<span class="sourceLineNo">114</span> */<a name="line.114"></a> -<span class="sourceLineNo">115</span> protected byte[][] extractKeyValues(Result r) {<a name="line.115"></a> -<span class="sourceLineNo">116</span> byte[][] keyVals = null;<a name="line.116"></a> -<span class="sourceLineNo">117</span> ArrayList<byte[]> foundList = new ArrayList<>();<a name="line.117"></a> -<span class="sourceLineNo">118</span> int numCols = columns.length;<a name="line.118"></a> -<span class="sourceLineNo">119</span> if (numCols > 0) {<a name="line.119"></a> -<span class="sourceLineNo">120</span> for (Cell value: r.listCells()) {<a name="line.120"></a> -<span class="sourceLineNo">121</span> byte [] column = KeyValue.makeColumn(CellUtil.cloneFamily(value),<a name="line.121"></a> -<span class="sourceLineNo">122</span> CellUtil.cloneQualifier(value));<a name="line.122"></a> -<span class="sourceLineNo">123</span> for (int i = 0; i < numCols; i++) {<a name="line.123"></a> -<span class="sourceLineNo">124</span> if (Bytes.equals(column, columns[i])) {<a name="line.124"></a> -<span class="sourceLineNo">125</span> foundList.add(CellUtil.cloneValue(value));<a name="line.125"></a> -<span class="sourceLineNo">126</span> break;<a name="line.126"></a> -<span class="sourceLineNo">127</span> }<a name="line.127"></a> -<span class="sourceLineNo">128</span> }<a name="line.128"></a> -<span class="sourceLineNo">129</span> }<a name="line.129"></a> -<span class="sourceLineNo">130</span> if(foundList.size() == numCols) {<a name="line.130"></a> -<span class="sourceLineNo">131</span> keyVals = foundList.toArray(new byte[numCols][]);<a name="line.131"></a> -<span class="sourceLineNo">132</span> }<a name="line.132"></a> -<span class="sourceLineNo">133</span> }<a name="line.133"></a> -<span class="sourceLineNo">134</span> return keyVals;<a name="line.134"></a> -<span class="sourceLineNo">135</span> }<a name="line.135"></a> -<span class="sourceLineNo">136</span><a name="line.136"></a> -<span class="sourceLineNo">137</span> /**<a name="line.137"></a> -<span class="sourceLineNo">138</span> * Create a key by concatenating multiple column values.<a name="line.138"></a> -<span class="sourceLineNo">139</span> * Override this function in order to produce different types of keys.<a name="line.139"></a> -<span class="sourceLineNo">140</span> *<a name="line.140"></a> -<span class="sourceLineNo">141</span> * @param vals<a name="line.141"></a> -<span class="sourceLineNo">142</span> * @return key generated by concatenating multiple column values<a name="line.142"></a> -<span class="sourceLineNo">143</span> */<a name="line.143"></a> -<span class="sourceLineNo">144</span> protected ImmutableBytesWritable createGroupKey(byte[][] vals) {<a name="line.144"></a> -<span class="sourceLineNo">145</span> if(vals == null) {<a name="line.145"></a> -<span class="sourceLineNo">146</span> return null;<a name="line.146"></a> -<span class="sourceLineNo">147</span> }<a name="line.147"></a> -<span class="sourceLineNo">148</span> StringBuilder sb = new StringBuilder();<a name="line.148"></a> -<span class="sourceLineNo">149</span> for(int i = 0; i < vals.length; i++) {<a name="line.149"></a> -<span class="sourceLineNo">150</span> if(i > 0) {<a name="line.150"></a> -<span class="sourceLineNo">151</span> sb.append(" ");<a name="line.151"></a> -<span class="sourceLineNo">152</span> }<a name="line.152"></a> -<span class="sourceLineNo">153</span> sb.append(Bytes.toString(vals[i]));<a name="line.153"></a> -<span class="sourceLineNo">154</span> }<a name="line.154"></a> -<span class="sourceLineNo">155</span> return new ImmutableBytesWritable(Bytes.toBytesBinary(sb.toString()));<a name="line.155"></a> -<span class="sourceLineNo">156</span> }<a name="line.156"></a> -<span class="sourceLineNo">157</span>}<a name="line.157"></a> +<span class="sourceLineNo">036</span>/**<a name="line.36"></a> +<span class="sourceLineNo">037</span> * Extract grouping columns from input record<a name="line.37"></a> +<span class="sourceLineNo">038</span> */<a name="line.38"></a> +<span class="sourceLineNo">039</span>@InterfaceAudience.Public<a name="line.39"></a> +<span class="sourceLineNo">040</span>public class GroupingTableMap<a name="line.40"></a> +<span class="sourceLineNo">041</span>extends MapReduceBase<a name="line.41"></a> +<span class="sourceLineNo">042</span>implements TableMap<ImmutableBytesWritable,Result> {<a name="line.42"></a> +<span class="sourceLineNo">043</span><a name="line.43"></a> +<span class="sourceLineNo">044</span> /**<a name="line.44"></a> +<span class="sourceLineNo">045</span> * JobConf parameter to specify the columns used to produce the key passed to<a name="line.45"></a> +<span class="sourceLineNo">046</span> * collect from the map phase<a name="line.46"></a> +<span class="sourceLineNo">047</span> */<a name="line.47"></a> +<span class="sourceLineNo">048</span> public static final String GROUP_COLUMNS =<a name="line.48"></a> +<span class="sourceLineNo">049</span> "hbase.mapred.groupingtablemap.columns";<a name="line.49"></a> +<span class="sourceLineNo">050</span><a name="line.50"></a> +<span class="sourceLineNo">051</span> protected byte [][] columns;<a name="line.51"></a> +<span class="sourceLineNo">052</span><a name="line.52"></a> +<span class="sourceLineNo">053</span> /**<a name="line.53"></a> +<span class="sourceLineNo">054</span> * Use this before submitting a TableMap job. It will appropriately set up the<a name="line.54"></a> +<span class="sourceLineNo">055</span> * JobConf.<a name="line.55"></a> +<span class="sourceLineNo">056</span> *<a name="line.56"></a> +<span class="sourceLineNo">057</span> * @param table table to be processed<a name="line.57"></a> +<span class="sourceLineNo">058</span> * @param columns space separated list of columns to fetch<a name="line.58"></a> +<span class="sourceLineNo">059</span> * @param groupColumns space separated list of columns used to form the key<a name="line.59"></a> +<span class="sourceLineNo">060</span> * used in collect<a name="line.60"></a> +<span class="sourceLineNo">061</span> * @param mapper map class<a name="line.61"></a> +<span class="sourceLineNo">062</span> * @param job job configuration object<a name="line.62"></a> +<span class="sourceLineNo">063</span> */<a name="line.63"></a> +<span class="sourceLineNo">064</span> @SuppressWarnings("unchecked")<a name="line.64"></a> +<span class="sourceLineNo">065</span> public static void initJob(String table, String columns, String groupColumns,<a name="line.65"></a> +<span class="sourceLineNo">066</span> Class<? extends TableMap> mapper, JobConf job) {<a name="line.66"></a> +<span class="sourceLineNo">067</span><a name="line.67"></a> +<span class="sourceLineNo">068</span> TableMapReduceUtil.initTableMapJob(table, columns, mapper,<a name="line.68"></a> +<span class="sourceLineNo">069</span> ImmutableBytesWritable.class, Result.class, job);<a name="line.69"></a> +<span class="sourceLineNo">070</span> job.set(GROUP_COLUMNS, groupColumns);<a name="line.70"></a> +<span class="sourceLineNo">071</span> }<a name="line.71"></a> +<span class="sourceLineNo">072</span><a name="line.72"></a> +<span class="sourceLineNo">073</span> @Override<a name="line.73"></a> +<span class="sourceLineNo">074</span> public void configure(JobConf job) {<a name="line.74"></a> +<span class="sourceLineNo">075</span> super.configure(job);<a name="line.75"></a> +<span class="sourceLineNo">076</span> String[] cols = job.get(GROUP_COLUMNS, "").split(" ");<a name="line.76"></a> +<span class="sourceLineNo">077</span> columns = new byte[cols.length][];<a name="line.77"></a> +<span class="sourceLineNo">078</span> for(int i = 0; i < cols.length; i++) {<a name="line.78"></a> +<span class="sourceLineNo">079</span> columns[i] = Bytes.toBytes(cols[i]);<a name="line.79"></a> +<span class="sourceLineNo">080</span> }<a name="line.80"></a> +<span class="sourceLineNo">081</span> }<a name="line.81"></a> +<span class="sourceLineNo">082</span><a name="line.82"></a> +<span class="sourceLineNo">083</span> /**<a name="line.83"></a> +<span class="sourceLineNo">084</span> * Extract the grouping columns from value to construct a new key.<a name="line.84"></a> +<span class="sourceLineNo">085</span> *<a name="line.85"></a> +<span class="sourceLineNo">086</span> * Pass the new key and value to reduce.<a name="line.86"></a> +<span class="sourceLineNo">087</span> * If any of the grouping columns are not found in the value, the record is skipped.<a name="line.87"></a> +<span class="sourceLineNo">088</span> * @param key<a name="line.88"></a> +<span class="sourceLineNo">089</span> * @param value<a name="line.89"></a> +<span class="sourceLineNo">090</span> * @param output<a name="line.90"></a> +<span class="sourceLineNo">091</span> * @param reporter<a name="line.91"></a> +<span class="sourceLineNo">092</span> * @throws IOException<a name="line.92"></a> +<span class="sourceLineNo">093</span> */<a name="line.93"></a> +<span class="sourceLineNo">094</span> public void map(ImmutableBytesWritable key, Result value,<a name="line.94"></a> +<span class="sourceLineNo">095</span> OutputCollector<ImmutableBytesWritable,Result> output,<a name="line.95"></a> +<span class="sourceLineNo">096</span> Reporter reporter) throws IOException {<a name="line.96"></a> +<span class="sourceLineNo">097</span><a name="line.97"></a> +<span class="sourceLineNo">098</span> byte[][] keyVals = extractKeyValues(value);<a name="line.98"></a> +<span class="sourceLineNo">099</span> if(keyVals != null) {<a name="line.99"></a> +<span class="sourceLineNo">100</span> ImmutableBytesWritable tKey = createGroupKey(keyVals);<a name="line.100"></a> +<span class="sourceLineNo">101</span> output.collect(tKey, value);<a name="line.101"></a> +<span class="sourceLineNo">102</span> }<a name="line.102"></a> +<span class="sourceLineNo">103</span> }<a name="line.103"></a> +<span class="sourceLineNo">104</span><a name="line.104"></a> +<span class="sourceLineNo">105</span> /**<a name="line.105"></a> +<span class="sourceLineNo">106</span> * Extract columns values from the current record. This method returns<a name="line.106"></a> +<span class="sourceLineNo">107</span> * null if any of the columns are not found.<a name="line.107"></a> +<span class="sourceLineNo">108</span> *<a name="line.108"></a> +<span class="sourceLineNo">109</span> * Override this method if you want to deal with nulls differently.<a name="line.109"></a> +<span class="sourceLineNo">110</span> *<a name="line.110"></a> +<span class="sourceLineNo">111</span> * @param r<a name="line.111"></a> +<span class="sourceLineNo">112</span> * @return array of byte values<a name="line.112"></a> +<span class="sourceLineNo">113</span> */<a name="line.113"></a> +<span class="sourceLineNo">114</span> protected byte[][] extractKeyValues(Result r) {<a name="line.114"></a> +<span class="sourceLineNo">115</span> byte[][] keyVals = null;<a name="line.115"></a> +<span class="sourceLineNo">116</span> ArrayList<byte[]> foundList = new ArrayList<>();<a name="line.116"></a> +<span class="sourceLineNo">117</span> int numCols = columns.length;<a name="line.117"></a> +<span class="sourceLineNo">118</span> if (numCols > 0) {<a name="line.118"></a> +<span class="sourceLineNo">119</span> for (Cell value: r.listCells()) {<a name="line.119"></a> +<span class="sourceLineNo">120</span> byte [] column = CellUtil.makeColumn(CellUtil.cloneFamily(value),<a name="line.120"></a> +<span class="sourceLineNo">121</span> CellUtil.cloneQualifier(value));<a name="line.121"></a> +<span class="sourceLineNo">122</span> for (int i = 0; i < numCols; i++) {<a name="line.122"></a> +<span class="sourceLineNo">123</span> if (Bytes.equals(column, columns[i])) {<a name="line.123"></a> +<span class="sourceLineNo">124</span> foundList.add(CellUtil.cloneValue(value));<a name="line.124"></a> +<span class="sourceLineNo">125</span> break;<a name="line.125"></a> +<span class="sourceLineNo">126</span> }<a name="line.126"></a> +<span class="sourceLineNo">127</span> }<a name="line.127"></a> +<span class="sourceLineNo">128</span> }<a name="line.128"></a> +<span class="sourceLineNo">129</span> if(foundList.size() == numCols) {<a name="line.129"></a> +<span class="sourceLineNo">130</span> keyVals = foundList.toArray(new byte[numCols][]);<a name="line.130"></a> +<span class="sourceLineNo">131</span> }<a name="line.131"></a> +<span class="sourceLineNo">132</span> }<a name="line.132"></a> +<span class="sourceLineNo">133</span> return keyVals;<a name="line.133"></a> +<span class="sourceLineNo">134</span> }<a name="line.134"></a> +<span class="sourceLineNo">135</span><a name="line.135"></a> +<span class="sourceLineNo">136</span> /**<a name="line.136"></a> +<span class="sourceLineNo">137</span> * Create a key by concatenating multiple column values.<a name="line.137"></a> +<span class="sourceLineNo">138</span> * Override this function in order to produce different types of keys.<a name="line.138"></a> +<span class="sourceLineNo">139</span> *<a name="line.139"></a> +<span class="sourceLineNo">140</span> * @param vals<a name="line.140"></a> +<span class="sourceLineNo">141</span> * @return key generated by concatenating multiple column values<a name="line.141"></a> +<span class="sourceLineNo">142</span> */<a name="line.142"></a> +<span class="sourceLineNo">143</span> protected ImmutableBytesWritable createGroupKey(byte[][] vals) {<a name="line.143"></a> +<span class="sourceLineNo">144</span> if(vals == null) {<a name="line.144"></a> +<span class="sourceLineNo">145</span> return null;<a name="line.145"></a> +<span class="sourceLineNo">146</span> }<a name="line.146"></a> +<span class="sourceLineNo">147</span> StringBuilder sb = new StringBuilder();<a name="line.147"></a> +<span class="sourceLineNo">148</span> for(int i = 0; i < vals.length; i++) {<a name="line.148"></a> +<span class="sourceLineNo">149</span> if(i > 0) {<a name="line.149"></a> +<span class="sourceLineNo">150</span> sb.append(" ");<a name="line.150"></a> +<span class="sourceLineNo">151</span> }<a name="line.151"></a> +<span class="sourceLineNo">152</span> sb.append(Bytes.toString(vals[i]));<a name="line.152"></a> +<span class="sourceLineNo">153</span> }<a name="line.153"></a> +<span class="sourceLineNo">154</span> return new ImmutableBytesWritable(Bytes.toBytesBinary(sb.toString()));<a name="line.154"></a> +<span class="sourceLineNo">155</span> }<a name="line.155"></a> +<span class="sourceLineNo">156</span>}<a name="line.156"></a>
http://git-wip-us.apache.org/repos/asf/hbase-site/blob/aea328be/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/GroupingTableMapper.html ---------------------------------------------------------------------- diff --git a/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/GroupingTableMapper.html b/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/GroupingTableMapper.html index 0e533ea..54bb0e5 100644 --- a/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/GroupingTableMapper.html +++ b/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/GroupingTableMapper.html @@ -34,155 +34,154 @@ <span class="sourceLineNo">026</span>import org.apache.hadoop.conf.Configuration;<a name="line.26"></a> <span class="sourceLineNo">027</span>import org.apache.hadoop.hbase.Cell;<a name="line.27"></a> <span class="sourceLineNo">028</span>import org.apache.hadoop.hbase.CellUtil;<a name="line.28"></a> -<span class="sourceLineNo">029</span>import org.apache.hadoop.hbase.KeyValue;<a name="line.29"></a> -<span class="sourceLineNo">030</span>import org.apache.hadoop.hbase.client.Result;<a name="line.30"></a> -<span class="sourceLineNo">031</span>import org.apache.hadoop.hbase.client.Scan;<a name="line.31"></a> -<span class="sourceLineNo">032</span>import org.apache.hadoop.hbase.io.ImmutableBytesWritable;<a name="line.32"></a> -<span class="sourceLineNo">033</span>import org.apache.hadoop.hbase.util.Bytes;<a name="line.33"></a> -<span class="sourceLineNo">034</span>import org.apache.hadoop.mapreduce.Job;<a name="line.34"></a> -<span class="sourceLineNo">035</span><a name="line.35"></a> -<span class="sourceLineNo">036</span>/**<a name="line.36"></a> -<span class="sourceLineNo">037</span> * Extract grouping columns from input record.<a name="line.37"></a> -<span class="sourceLineNo">038</span> */<a name="line.38"></a> -<span class="sourceLineNo">039</span>@InterfaceAudience.Public<a name="line.39"></a> -<span class="sourceLineNo">040</span>public class GroupingTableMapper<a name="line.40"></a> -<span class="sourceLineNo">041</span>extends TableMapper<ImmutableBytesWritable,Result> implements Configurable {<a name="line.41"></a> -<span class="sourceLineNo">042</span><a name="line.42"></a> -<span class="sourceLineNo">043</span> /**<a name="line.43"></a> -<span class="sourceLineNo">044</span> * JobConf parameter to specify the columns used to produce the key passed to<a name="line.44"></a> -<span class="sourceLineNo">045</span> * collect from the map phase.<a name="line.45"></a> -<span class="sourceLineNo">046</span> */<a name="line.46"></a> -<span class="sourceLineNo">047</span> public static final String GROUP_COLUMNS =<a name="line.47"></a> -<span class="sourceLineNo">048</span> "hbase.mapred.groupingtablemap.columns";<a name="line.48"></a> -<span class="sourceLineNo">049</span><a name="line.49"></a> -<span class="sourceLineNo">050</span> /** The grouping columns. */<a name="line.50"></a> -<span class="sourceLineNo">051</span> protected byte [][] columns;<a name="line.51"></a> -<span class="sourceLineNo">052</span> /** The current configuration. */<a name="line.52"></a> -<span class="sourceLineNo">053</span> private Configuration conf = null;<a name="line.53"></a> -<span class="sourceLineNo">054</span><a name="line.54"></a> -<span class="sourceLineNo">055</span> /**<a name="line.55"></a> -<span class="sourceLineNo">056</span> * Use this before submitting a TableMap job. It will appropriately set up<a name="line.56"></a> -<span class="sourceLineNo">057</span> * the job.<a name="line.57"></a> -<span class="sourceLineNo">058</span> *<a name="line.58"></a> -<span class="sourceLineNo">059</span> * @param table The table to be processed.<a name="line.59"></a> -<span class="sourceLineNo">060</span> * @param scan The scan with the columns etc.<a name="line.60"></a> -<span class="sourceLineNo">061</span> * @param groupColumns A space separated list of columns used to form the<a name="line.61"></a> -<span class="sourceLineNo">062</span> * key used in collect.<a name="line.62"></a> -<span class="sourceLineNo">063</span> * @param mapper The mapper class.<a name="line.63"></a> -<span class="sourceLineNo">064</span> * @param job The current job.<a name="line.64"></a> -<span class="sourceLineNo">065</span> * @throws IOException When setting up the job fails.<a name="line.65"></a> -<span class="sourceLineNo">066</span> */<a name="line.66"></a> -<span class="sourceLineNo">067</span> @SuppressWarnings("unchecked")<a name="line.67"></a> -<span class="sourceLineNo">068</span> public static void initJob(String table, Scan scan, String groupColumns,<a name="line.68"></a> -<span class="sourceLineNo">069</span> Class<? extends TableMapper> mapper, Job job) throws IOException {<a name="line.69"></a> -<span class="sourceLineNo">070</span> TableMapReduceUtil.initTableMapperJob(table, scan, mapper,<a name="line.70"></a> -<span class="sourceLineNo">071</span> ImmutableBytesWritable.class, Result.class, job);<a name="line.71"></a> -<span class="sourceLineNo">072</span> job.getConfiguration().set(GROUP_COLUMNS, groupColumns);<a name="line.72"></a> -<span class="sourceLineNo">073</span> }<a name="line.73"></a> -<span class="sourceLineNo">074</span><a name="line.74"></a> -<span class="sourceLineNo">075</span> /**<a name="line.75"></a> -<span class="sourceLineNo">076</span> * Extract the grouping columns from value to construct a new key. Pass the<a name="line.76"></a> -<span class="sourceLineNo">077</span> * new key and value to reduce. If any of the grouping columns are not found<a name="line.77"></a> -<span class="sourceLineNo">078</span> * in the value, the record is skipped.<a name="line.78"></a> -<span class="sourceLineNo">079</span> *<a name="line.79"></a> -<span class="sourceLineNo">080</span> * @param key The current key.<a name="line.80"></a> -<span class="sourceLineNo">081</span> * @param value The current value.<a name="line.81"></a> -<span class="sourceLineNo">082</span> * @param context The current context.<a name="line.82"></a> -<span class="sourceLineNo">083</span> * @throws IOException When writing the record fails.<a name="line.83"></a> -<span class="sourceLineNo">084</span> * @throws InterruptedException When the job is aborted.<a name="line.84"></a> -<span class="sourceLineNo">085</span> */<a name="line.85"></a> -<span class="sourceLineNo">086</span> @Override<a name="line.86"></a> -<span class="sourceLineNo">087</span> public void map(ImmutableBytesWritable key, Result value, Context context)<a name="line.87"></a> -<span class="sourceLineNo">088</span> throws IOException, InterruptedException {<a name="line.88"></a> -<span class="sourceLineNo">089</span> byte[][] keyVals = extractKeyValues(value);<a name="line.89"></a> -<span class="sourceLineNo">090</span> if(keyVals != null) {<a name="line.90"></a> -<span class="sourceLineNo">091</span> ImmutableBytesWritable tKey = createGroupKey(keyVals);<a name="line.91"></a> -<span class="sourceLineNo">092</span> context.write(tKey, value);<a name="line.92"></a> -<span class="sourceLineNo">093</span> }<a name="line.93"></a> -<span class="sourceLineNo">094</span> }<a name="line.94"></a> -<span class="sourceLineNo">095</span><a name="line.95"></a> -<span class="sourceLineNo">096</span> /**<a name="line.96"></a> -<span class="sourceLineNo">097</span> * Extract columns values from the current record. This method returns<a name="line.97"></a> -<span class="sourceLineNo">098</span> * null if any of the columns are not found.<a name="line.98"></a> -<span class="sourceLineNo">099</span> * <p><a name="line.99"></a> -<span class="sourceLineNo">100</span> * Override this method if you want to deal with nulls differently.<a name="line.100"></a> -<span class="sourceLineNo">101</span> *<a name="line.101"></a> -<span class="sourceLineNo">102</span> * @param r The current values.<a name="line.102"></a> -<span class="sourceLineNo">103</span> * @return Array of byte values.<a name="line.103"></a> -<span class="sourceLineNo">104</span> */<a name="line.104"></a> -<span class="sourceLineNo">105</span> protected byte[][] extractKeyValues(Result r) {<a name="line.105"></a> -<span class="sourceLineNo">106</span> byte[][] keyVals = null;<a name="line.106"></a> -<span class="sourceLineNo">107</span> ArrayList<byte[]> foundList = new ArrayList<>();<a name="line.107"></a> -<span class="sourceLineNo">108</span> int numCols = columns.length;<a name="line.108"></a> -<span class="sourceLineNo">109</span> if (numCols > 0) {<a name="line.109"></a> -<span class="sourceLineNo">110</span> for (Cell value: r.listCells()) {<a name="line.110"></a> -<span class="sourceLineNo">111</span> byte [] column = KeyValue.makeColumn(CellUtil.cloneFamily(value),<a name="line.111"></a> -<span class="sourceLineNo">112</span> CellUtil.cloneQualifier(value));<a name="line.112"></a> -<span class="sourceLineNo">113</span> for (int i = 0; i < numCols; i++) {<a name="line.113"></a> -<span class="sourceLineNo">114</span> if (Bytes.equals(column, columns[i])) {<a name="line.114"></a> -<span class="sourceLineNo">115</span> foundList.add(CellUtil.cloneValue(value));<a name="line.115"></a> -<span class="sourceLineNo">116</span> break;<a name="line.116"></a> -<span class="sourceLineNo">117</span> }<a name="line.117"></a> -<span class="sourceLineNo">118</span> }<a name="line.118"></a> -<span class="sourceLineNo">119</span> }<a name="line.119"></a> -<span class="sourceLineNo">120</span> if(foundList.size() == numCols) {<a name="line.120"></a> -<span class="sourceLineNo">121</span> keyVals = foundList.toArray(new byte[numCols][]);<a name="line.121"></a> -<span class="sourceLineNo">122</span> }<a name="line.122"></a> -<span class="sourceLineNo">123</span> }<a name="line.123"></a> -<span class="sourceLineNo">124</span> return keyVals;<a name="line.124"></a> -<span class="sourceLineNo">125</span> }<a name="line.125"></a> -<span class="sourceLineNo">126</span><a name="line.126"></a> -<span class="sourceLineNo">127</span> /**<a name="line.127"></a> -<span class="sourceLineNo">128</span> * Create a key by concatenating multiple column values.<a name="line.128"></a> -<span class="sourceLineNo">129</span> * <p><a name="line.129"></a> -<span class="sourceLineNo">130</span> * Override this function in order to produce different types of keys.<a name="line.130"></a> -<span class="sourceLineNo">131</span> *<a name="line.131"></a> -<span class="sourceLineNo">132</span> * @param vals The current key/values.<a name="line.132"></a> -<span class="sourceLineNo">133</span> * @return A key generated by concatenating multiple column values.<a name="line.133"></a> -<span class="sourceLineNo">134</span> */<a name="line.134"></a> -<span class="sourceLineNo">135</span> protected ImmutableBytesWritable createGroupKey(byte[][] vals) {<a name="line.135"></a> -<span class="sourceLineNo">136</span> if(vals == null) {<a name="line.136"></a> -<span class="sourceLineNo">137</span> return null;<a name="line.137"></a> -<span class="sourceLineNo">138</span> }<a name="line.138"></a> -<span class="sourceLineNo">139</span> StringBuilder sb = new StringBuilder();<a name="line.139"></a> -<span class="sourceLineNo">140</span> for(int i = 0; i < vals.length; i++) {<a name="line.140"></a> -<span class="sourceLineNo">141</span> if(i > 0) {<a name="line.141"></a> -<span class="sourceLineNo">142</span> sb.append(" ");<a name="line.142"></a> -<span class="sourceLineNo">143</span> }<a name="line.143"></a> -<span class="sourceLineNo">144</span> sb.append(Bytes.toString(vals[i]));<a name="line.144"></a> -<span class="sourceLineNo">145</span> }<a name="line.145"></a> -<span class="sourceLineNo">146</span> return new ImmutableBytesWritable(Bytes.toBytesBinary(sb.toString()));<a name="line.146"></a> -<span class="sourceLineNo">147</span> }<a name="line.147"></a> -<span class="sourceLineNo">148</span><a name="line.148"></a> -<span class="sourceLineNo">149</span> /**<a name="line.149"></a> -<span class="sourceLineNo">150</span> * Returns the current configuration.<a name="line.150"></a> -<span class="sourceLineNo">151</span> *<a name="line.151"></a> -<span class="sourceLineNo">152</span> * @return The current configuration.<a name="line.152"></a> -<span class="sourceLineNo">153</span> * @see org.apache.hadoop.conf.Configurable#getConf()<a name="line.153"></a> -<span class="sourceLineNo">154</span> */<a name="line.154"></a> -<span class="sourceLineNo">155</span> @Override<a name="line.155"></a> -<span class="sourceLineNo">156</span> public Configuration getConf() {<a name="line.156"></a> -<span class="sourceLineNo">157</span> return conf;<a name="line.157"></a> -<span class="sourceLineNo">158</span> }<a name="line.158"></a> -<span class="sourceLineNo">159</span><a name="line.159"></a> -<span class="sourceLineNo">160</span> /**<a name="line.160"></a> -<span class="sourceLineNo">161</span> * Sets the configuration. This is used to set up the grouping details.<a name="line.161"></a> -<span class="sourceLineNo">162</span> *<a name="line.162"></a> -<span class="sourceLineNo">163</span> * @param configuration The configuration to set.<a name="line.163"></a> -<span class="sourceLineNo">164</span> * @see org.apache.hadoop.conf.Configurable#setConf(<a name="line.164"></a> -<span class="sourceLineNo">165</span> * org.apache.hadoop.conf.Configuration)<a name="line.165"></a> -<span class="sourceLineNo">166</span> */<a name="line.166"></a> -<span class="sourceLineNo">167</span> @Override<a name="line.167"></a> -<span class="sourceLineNo">168</span> public void setConf(Configuration configuration) {<a name="line.168"></a> -<span class="sourceLineNo">169</span> this.conf = configuration;<a name="line.169"></a> -<span class="sourceLineNo">170</span> String[] cols = conf.get(GROUP_COLUMNS, "").split(" ");<a name="line.170"></a> -<span class="sourceLineNo">171</span> columns = new byte[cols.length][];<a name="line.171"></a> -<span class="sourceLineNo">172</span> for(int i = 0; i < cols.length; i++) {<a name="line.172"></a> -<span class="sourceLineNo">173</span> columns[i] = Bytes.toBytes(cols[i]);<a name="line.173"></a> -<span class="sourceLineNo">174</span> }<a name="line.174"></a> -<span class="sourceLineNo">175</span> }<a name="line.175"></a> -<span class="sourceLineNo">176</span><a name="line.176"></a> -<span class="sourceLineNo">177</span>}<a name="line.177"></a> +<span class="sourceLineNo">029</span>import org.apache.hadoop.hbase.client.Result;<a name="line.29"></a> +<span class="sourceLineNo">030</span>import org.apache.hadoop.hbase.client.Scan;<a name="line.30"></a> +<span class="sourceLineNo">031</span>import org.apache.hadoop.hbase.io.ImmutableBytesWritable;<a name="line.31"></a> +<span class="sourceLineNo">032</span>import org.apache.hadoop.hbase.util.Bytes;<a name="line.32"></a> +<span class="sourceLineNo">033</span>import org.apache.hadoop.mapreduce.Job;<a name="line.33"></a> +<span class="sourceLineNo">034</span><a name="line.34"></a> +<span class="sourceLineNo">035</span>/**<a name="line.35"></a> +<span class="sourceLineNo">036</span> * Extract grouping columns from input record.<a name="line.36"></a> +<span class="sourceLineNo">037</span> */<a name="line.37"></a> +<span class="sourceLineNo">038</span>@InterfaceAudience.Public<a name="line.38"></a> +<span class="sourceLineNo">039</span>public class GroupingTableMapper<a name="line.39"></a> +<span class="sourceLineNo">040</span>extends TableMapper<ImmutableBytesWritable,Result> implements Configurable {<a name="line.40"></a> +<span class="sourceLineNo">041</span><a name="line.41"></a> +<span class="sourceLineNo">042</span> /**<a name="line.42"></a> +<span class="sourceLineNo">043</span> * JobConf parameter to specify the columns used to produce the key passed to<a name="line.43"></a> +<span class="sourceLineNo">044</span> * collect from the map phase.<a name="line.44"></a> +<span class="sourceLineNo">045</span> */<a name="line.45"></a> +<span class="sourceLineNo">046</span> public static final String GROUP_COLUMNS =<a name="line.46"></a> +<span class="sourceLineNo">047</span> "hbase.mapred.groupingtablemap.columns";<a name="line.47"></a> +<span class="sourceLineNo">048</span><a name="line.48"></a> +<span class="sourceLineNo">049</span> /** The grouping columns. */<a name="line.49"></a> +<span class="sourceLineNo">050</span> protected byte [][] columns;<a name="line.50"></a> +<span class="sourceLineNo">051</span> /** The current configuration. */<a name="line.51"></a> +<span class="sourceLineNo">052</span> private Configuration conf = null;<a name="line.52"></a> +<span class="sourceLineNo">053</span><a name="line.53"></a> +<span class="sourceLineNo">054</span> /**<a name="line.54"></a> +<span class="sourceLineNo">055</span> * Use this before submitting a TableMap job. It will appropriately set up<a name="line.55"></a> +<span class="sourceLineNo">056</span> * the job.<a name="line.56"></a> +<span class="sourceLineNo">057</span> *<a name="line.57"></a> +<span class="sourceLineNo">058</span> * @param table The table to be processed.<a name="line.58"></a> +<span class="sourceLineNo">059</span> * @param scan The scan with the columns etc.<a name="line.59"></a> +<span class="sourceLineNo">060</span> * @param groupColumns A space separated list of columns used to form the<a name="line.60"></a> +<span class="sourceLineNo">061</span> * key used in collect.<a name="line.61"></a> +<span class="sourceLineNo">062</span> * @param mapper The mapper class.<a name="line.62"></a> +<span class="sourceLineNo">063</span> * @param job The current job.<a name="line.63"></a> +<span class="sourceLineNo">064</span> * @throws IOException When setting up the job fails.<a name="line.64"></a> +<span class="sourceLineNo">065</span> */<a name="line.65"></a> +<span class="sourceLineNo">066</span> @SuppressWarnings("unchecked")<a name="line.66"></a> +<span class="sourceLineNo">067</span> public static void initJob(String table, Scan scan, String groupColumns,<a name="line.67"></a> +<span class="sourceLineNo">068</span> Class<? extends TableMapper> mapper, Job job) throws IOException {<a name="line.68"></a> +<span class="sourceLineNo">069</span> TableMapReduceUtil.initTableMapperJob(table, scan, mapper,<a name="line.69"></a> +<span class="sourceLineNo">070</span> ImmutableBytesWritable.class, Result.class, job);<a name="line.70"></a> +<span class="sourceLineNo">071</span> job.getConfiguration().set(GROUP_COLUMNS, groupColumns);<a name="line.71"></a> +<span class="sourceLineNo">072</span> }<a name="line.72"></a> +<span class="sourceLineNo">073</span><a name="line.73"></a> +<span class="sourceLineNo">074</span> /**<a name="line.74"></a> +<span class="sourceLineNo">075</span> * Extract the grouping columns from value to construct a new key. Pass the<a name="line.75"></a> +<span class="sourceLineNo">076</span> * new key and value to reduce. If any of the grouping columns are not found<a name="line.76"></a> +<span class="sourceLineNo">077</span> * in the value, the record is skipped.<a name="line.77"></a> +<span class="sourceLineNo">078</span> *<a name="line.78"></a> +<span class="sourceLineNo">079</span> * @param key The current key.<a name="line.79"></a> +<span class="sourceLineNo">080</span> * @param value The current value.<a name="line.80"></a> +<span class="sourceLineNo">081</span> * @param context The current context.<a name="line.81"></a> +<span class="sourceLineNo">082</span> * @throws IOException When writing the record fails.<a name="line.82"></a> +<span class="sourceLineNo">083</span> * @throws InterruptedException When the job is aborted.<a name="line.83"></a> +<span class="sourceLineNo">084</span> */<a name="line.84"></a> +<span class="sourceLineNo">085</span> @Override<a name="line.85"></a> +<span class="sourceLineNo">086</span> public void map(ImmutableBytesWritable key, Result value, Context context)<a name="line.86"></a> +<span class="sourceLineNo">087</span> throws IOException, InterruptedException {<a name="line.87"></a> +<span class="sourceLineNo">088</span> byte[][] keyVals = extractKeyValues(value);<a name="line.88"></a> +<span class="sourceLineNo">089</span> if(keyVals != null) {<a name="line.89"></a> +<span class="sourceLineNo">090</span> ImmutableBytesWritable tKey = createGroupKey(keyVals);<a name="line.90"></a> +<span class="sourceLineNo">091</span> context.write(tKey, value);<a name="line.91"></a> +<span class="sourceLineNo">092</span> }<a name="line.92"></a> +<span class="sourceLineNo">093</span> }<a name="line.93"></a> +<span class="sourceLineNo">094</span><a name="line.94"></a> +<span class="sourceLineNo">095</span> /**<a name="line.95"></a> +<span class="sourceLineNo">096</span> * Extract columns values from the current record. This method returns<a name="line.96"></a> +<span class="sourceLineNo">097</span> * null if any of the columns are not found.<a name="line.97"></a> +<span class="sourceLineNo">098</span> * <p><a name="line.98"></a> +<span class="sourceLineNo">099</span> * Override this method if you want to deal with nulls differently.<a name="line.99"></a> +<span class="sourceLineNo">100</span> *<a name="line.100"></a> +<span class="sourceLineNo">101</span> * @param r The current values.<a name="line.101"></a> +<span class="sourceLineNo">102</span> * @return Array of byte values.<a name="line.102"></a> +<span class="sourceLineNo">103</span> */<a name="line.103"></a> +<span class="sourceLineNo">104</span> protected byte[][] extractKeyValues(Result r) {<a name="line.104"></a> +<span class="sourceLineNo">105</span> byte[][] keyVals = null;<a name="line.105"></a> +<span class="sourceLineNo">106</span> ArrayList<byte[]> foundList = new ArrayList<>();<a name="line.106"></a> +<span class="sourceLineNo">107</span> int numCols = columns.length;<a name="line.107"></a> +<span class="sourceLineNo">108</span> if (numCols > 0) {<a name="line.108"></a> +<span class="sourceLineNo">109</span> for (Cell value: r.listCells()) {<a name="line.109"></a> +<span class="sourceLineNo">110</span> byte [] column = CellUtil.makeColumn(CellUtil.cloneFamily(value),<a name="line.110"></a> +<span class="sourceLineNo">111</span> CellUtil.cloneQualifier(value));<a name="line.111"></a> +<span class="sourceLineNo">112</span> for (int i = 0; i < numCols; i++) {<a name="line.112"></a> +<span class="sourceLineNo">113</span> if (Bytes.equals(column, columns[i])) {<a name="line.113"></a> +<span class="sourceLineNo">114</span> foundList.add(CellUtil.cloneValue(value));<a name="line.114"></a> +<span class="sourceLineNo">115</span> break;<a name="line.115"></a> +<span class="sourceLineNo">116</span> }<a name="line.116"></a> +<span class="sourceLineNo">117</span> }<a name="line.117"></a> +<span class="sourceLineNo">118</span> }<a name="line.118"></a> +<span class="sourceLineNo">119</span> if(foundList.size() == numCols) {<a name="line.119"></a> +<span class="sourceLineNo">120</span> keyVals = foundList.toArray(new byte[numCols][]);<a name="line.120"></a> +<span class="sourceLineNo">121</span> }<a name="line.121"></a> +<span class="sourceLineNo">122</span> }<a name="line.122"></a> +<span class="sourceLineNo">123</span> return keyVals;<a name="line.123"></a> +<span class="sourceLineNo">124</span> }<a name="line.124"></a> +<span class="sourceLineNo">125</span><a name="line.125"></a> +<span class="sourceLineNo">126</span> /**<a name="line.126"></a> +<span class="sourceLineNo">127</span> * Create a key by concatenating multiple column values.<a name="line.127"></a> +<span class="sourceLineNo">128</span> * <p><a name="line.128"></a> +<span class="sourceLineNo">129</span> * Override this function in order to produce different types of keys.<a name="line.129"></a> +<span class="sourceLineNo">130</span> *<a name="line.130"></a> +<span class="sourceLineNo">131</span> * @param vals The current key/values.<a name="line.131"></a> +<span class="sourceLineNo">132</span> * @return A key generated by concatenating multiple column values.<a name="line.132"></a> +<span class="sourceLineNo">133</span> */<a name="line.133"></a> +<span class="sourceLineNo">134</span> protected ImmutableBytesWritable createGroupKey(byte[][] vals) {<a name="line.134"></a> +<span class="sourceLineNo">135</span> if(vals == null) {<a name="line.135"></a> +<span class="sourceLineNo">136</span> return null;<a name="line.136"></a> +<span class="sourceLineNo">137</span> }<a name="line.137"></a> +<span class="sourceLineNo">138</span> StringBuilder sb = new StringBuilder();<a name="line.138"></a> +<span class="sourceLineNo">139</span> for(int i = 0; i < vals.length; i++) {<a name="line.139"></a> +<span class="sourceLineNo">140</span> if(i > 0) {<a name="line.140"></a> +<span class="sourceLineNo">141</span> sb.append(" ");<a name="line.141"></a> +<span class="sourceLineNo">142</span> }<a name="line.142"></a> +<span class="sourceLineNo">143</span> sb.append(Bytes.toString(vals[i]));<a name="line.143"></a> +<span class="sourceLineNo">144</span> }<a name="line.144"></a> +<span class="sourceLineNo">145</span> return new ImmutableBytesWritable(Bytes.toBytesBinary(sb.toString()));<a name="line.145"></a> +<span class="sourceLineNo">146</span> }<a name="line.146"></a> +<span class="sourceLineNo">147</span><a name="line.147"></a> +<span class="sourceLineNo">148</span> /**<a name="line.148"></a> +<span class="sourceLineNo">149</span> * Returns the current configuration.<a name="line.149"></a> +<span class="sourceLineNo">150</span> *<a name="line.150"></a> +<span class="sourceLineNo">151</span> * @return The current configuration.<a name="line.151"></a> +<span class="sourceLineNo">152</span> * @see org.apache.hadoop.conf.Configurable#getConf()<a name="line.152"></a> +<span class="sourceLineNo">153</span> */<a name="line.153"></a> +<span class="sourceLineNo">154</span> @Override<a name="line.154"></a> +<span class="sourceLineNo">155</span> public Configuration getConf() {<a name="line.155"></a> +<span class="sourceLineNo">156</span> return conf;<a name="line.156"></a> +<span class="sourceLineNo">157</span> }<a name="line.157"></a> +<span class="sourceLineNo">158</span><a name="line.158"></a> +<span class="sourceLineNo">159</span> /**<a name="line.159"></a> +<span class="sourceLineNo">160</span> * Sets the configuration. This is used to set up the grouping details.<a name="line.160"></a> +<span class="sourceLineNo">161</span> *<a name="line.161"></a> +<span class="sourceLineNo">162</span> * @param configuration The configuration to set.<a name="line.162"></a> +<span class="sourceLineNo">163</span> * @see org.apache.hadoop.conf.Configurable#setConf(<a name="line.163"></a> +<span class="sourceLineNo">164</span> * org.apache.hadoop.conf.Configuration)<a name="line.164"></a> +<span class="sourceLineNo">165</span> */<a name="line.165"></a> +<span class="sourceLineNo">166</span> @Override<a name="line.166"></a> +<span class="sourceLineNo">167</span> public void setConf(Configuration configuration) {<a name="line.167"></a> +<span class="sourceLineNo">168</span> this.conf = configuration;<a name="line.168"></a> +<span class="sourceLineNo">169</span> String[] cols = conf.get(GROUP_COLUMNS, "").split(" ");<a name="line.169"></a> +<span class="sourceLineNo">170</span> columns = new byte[cols.length][];<a name="line.170"></a> +<span class="sourceLineNo">171</span> for(int i = 0; i < cols.length; i++) {<a name="line.171"></a> +<span class="sourceLineNo">172</span> columns[i] = Bytes.toBytes(cols[i]);<a name="line.172"></a> +<span class="sourceLineNo">173</span> }<a name="line.173"></a> +<span class="sourceLineNo">174</span> }<a name="line.174"></a> +<span class="sourceLineNo">175</span><a name="line.175"></a> +<span class="sourceLineNo">176</span>}<a name="line.176"></a> http://git-wip-us.apache.org/repos/asf/hbase-site/blob/aea328be/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/TableInputFormat.html ---------------------------------------------------------------------- diff --git a/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/TableInputFormat.html b/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/TableInputFormat.html index 0c8664d..27a7727 100644 --- a/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/TableInputFormat.html +++ b/apidocs/src-html/org/apache/hadoop/hbase/mapreduce/TableInputFormat.html @@ -35,7 +35,7 @@ <span class="sourceLineNo">027</span>import org.apache.commons.logging.LogFactory;<a name="line.27"></a> <span class="sourceLineNo">028</span>import org.apache.hadoop.conf.Configurable;<a name="line.28"></a> <span class="sourceLineNo">029</span>import org.apache.hadoop.conf.Configuration;<a name="line.29"></a> -<span class="sourceLineNo">030</span>import org.apache.hadoop.hbase.KeyValue;<a name="line.30"></a> +<span class="sourceLineNo">030</span>import org.apache.hadoop.hbase.CellUtil;<a name="line.30"></a> <span class="sourceLineNo">031</span>import org.apache.hadoop.hbase.TableName;<a name="line.31"></a> <span class="sourceLineNo">032</span>import org.apache.yetus.audience.InterfaceAudience;<a name="line.32"></a> <span class="sourceLineNo">033</span>import org.apache.hadoop.hbase.client.Connection;<a name="line.33"></a> @@ -221,7 +221,7 @@ <span class="sourceLineNo">213</span> * @throws IllegalArgumentException When familyAndQualifier is invalid.<a name="line.213"></a> <span class="sourceLineNo">214</span> */<a name="line.214"></a> <span class="sourceLineNo">215</span> private static void addColumn(Scan scan, byte[] familyAndQualifier) {<a name="line.215"></a> -<span class="sourceLineNo">216</span> byte [][] fq = KeyValue.parseColumn(familyAndQualifier);<a name="line.216"></a> +<span class="sourceLineNo">216</span> byte [][] fq = CellUtil.parseColumn(familyAndQualifier);<a name="line.216"></a> <span class="sourceLineNo">217</span> if (fq.length == 1) {<a name="line.217"></a> <span class="sourceLineNo">218</span> scan.addFamily(fq[0]);<a name="line.218"></a> <span class="sourceLineNo">219</span> } else if (fq.length == 2) {<a name="line.219"></a> http://git-wip-us.apache.org/repos/asf/hbase-site/blob/aea328be/apidocs/src-html/org/apache/hadoop/hbase/rest/client/RemoteHTable.html ---------------------------------------------------------------------- diff --git a/apidocs/src-html/org/apache/hadoop/hbase/rest/client/RemoteHTable.html b/apidocs/src-html/org/apache/hadoop/hbase/rest/client/RemoteHTable.html index 0519a56..8b4a6be 100644 --- a/apidocs/src-html/org/apache/hadoop/hbase/rest/client/RemoteHTable.html +++ b/apidocs/src-html/org/apache/hadoop/hbase/rest/client/RemoteHTable.html @@ -185,7 +185,7 @@ <span class="sourceLineNo">177</span> for (RowModel row: model.getRows()) {<a name="line.177"></a> <span class="sourceLineNo">178</span> List<Cell> kvs = new ArrayList<>(row.getCells().size());<a name="line.178"></a> <span class="sourceLineNo">179</span> for (CellModel cell: row.getCells()) {<a name="line.179"></a> -<span class="sourceLineNo">180</span> byte[][] split = KeyValue.parseColumn(cell.getColumn());<a name="line.180"></a> +<span class="sourceLineNo">180</span> byte[][] split = CellUtil.parseColumn(cell.getColumn());<a name="line.180"></a> <span class="sourceLineNo">181</span> byte[] column = split[0];<a name="line.181"></a> <span class="sourceLineNo">182</span> byte[] qualifier = null;<a name="line.182"></a> <span class="sourceLineNo">183</span> if (split.length == 1) {<a name="line.183"></a> http://git-wip-us.apache.org/repos/asf/hbase-site/blob/aea328be/book.html ---------------------------------------------------------------------- diff --git a/book.html b/book.html index c746626..9ad6214 100644 --- a/book.html +++ b/book.html @@ -35210,7 +35210,7 @@ The server will return cellblocks compressed using this same compressor as long <div id="footer"> <div id="footer-text"> Version 3.0.0-SNAPSHOT<br> -Last updated 2017-09-28 14:29:37 UTC +Last updated 2017-09-29 14:29:46 UTC </div> </div> </body> http://git-wip-us.apache.org/repos/asf/hbase-site/blob/aea328be/bulk-loads.html ---------------------------------------------------------------------- diff --git a/bulk-loads.html b/bulk-loads.html index 4964ee4..02ea3bc 100644 --- a/bulk-loads.html +++ b/bulk-loads.html @@ -7,7 +7,7 @@ <head> <meta charset="UTF-8" /> <meta name="viewport" content="width=device-width, initial-scale=1.0" /> - <meta name="Date-Revision-yyyymmdd" content="20170928" /> + <meta name="Date-Revision-yyyymmdd" content="20170929" /> <meta http-equiv="Content-Language" content="en" /> <title>Apache HBase – Bulk Loads in Apache HBase (TM) @@ -311,7 +311,7 @@ under the License. --> <a href="https://www.apache.org/";>The Apache Software Foundation</a>. All rights reserved. - <li id="publishDate" class="pull-right">Last Published: 2017-09-28</li> + <li id="publishDate" class="pull-right">Last Published: 2017-09-29</li> </p> </div>
