org.sf.xrime.algorithms.partitions.connected.bi.EdgeSetJoin.java Source code

Introduction

Here is the source code for org.sf.xrime.algorithms.partitions.connected.bi.EdgeSetJoin.java

Source

/*
 * Copyright (C) IBM Corp. 2009.
 * 
 * Licensed under the Apache License, Version 2.0 (the "License"); you may not
 * use this file except in compliance with the License. You may obtain a copy of
 * the License at
 * 
 * http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
 * License for the specific language governing permissions and limitations under
 * the License.
 */
package org.sf.xrime.algorithms.partitions.connected.bi;

import java.io.IOException;
import java.util.Iterator;
import java.util.TreeSet;

import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.compress.GzipCodec;
import org.apache.hadoop.mapred.FileInputFormat;
import org.apache.hadoop.mapred.FileOutputFormat;
import org.apache.hadoop.mapred.JobClient;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.Mapper;
import org.apache.hadoop.mapred.OutputCollector;
import org.apache.hadoop.mapred.Reducer;
import org.apache.hadoop.mapred.Reporter;
import org.apache.hadoop.mapred.SequenceFileInputFormat;
import org.apache.hadoop.mapred.SequenceFileOutputFormat;
import org.sf.xrime.ProcessorExecutionException;
import org.sf.xrime.algorithms.GraphAlgorithm;
import org.sf.xrime.algorithms.utils.GraphAlgorithmMapReduceBase;
import org.sf.xrime.model.edge.Edge;
import org.sf.xrime.model.edge.EdgeComparator;
import org.sf.xrime.model.edge.EdgeSet;

/**
 * Just like weakly.alg_1.VertexSetJoin, used to calculate the transitive closure of the
 * share edge relationship between circles (edge sets).
 * @author xue
 */
public class EdgeSetJoin extends GraphAlgorithm {
    /**
     * Default constructor.
     */
    public EdgeSetJoin() {
        super();
    }

    /**
     * Emit the edge at lower layer as k2, the edge at higher layer as v2. 
     * @author xue
     */
    public static class MapClass extends GraphAlgorithmMapReduceBase
            implements Mapper<Text, EdgeSet, Text, EdgeSet> {

        @Override
        public void map(Text key, EdgeSet value, OutputCollector<Text, EdgeSet> output, Reporter reporter)
                throws IOException {
            // Generate the higher layer edge first.
            int index_of_sharp = key.toString().indexOf(ConstantLabels.NON_ID_CHAR);
            String from = key.toString().substring(0, index_of_sharp);
            String to = key.toString().substring(index_of_sharp + 1, key.toString().length());
            Edge higher_edge = new Edge(from, to);

            // Generate the higher layer edge set.
            EdgeSet result_set = new EdgeSet();
            result_set.addEdge(higher_edge);
            // Emit.
            for (Edge edge : value.getEdges()) {
                String k2 = edge.getFrom() + ConstantLabels.NON_ID_CHAR + edge.getTo();
                output.collect(new Text(k2), result_set);
            }
        }
    }

    /**
     * Merge higher layer edge sets with the same lower edge.
     * @author xue
     */
    public static class ReduceClass extends GraphAlgorithmMapReduceBase
            implements Reducer<Text, EdgeSet, Text, EdgeSet> {

        @Override
        public void reduce(Text key, Iterator<EdgeSet> values, OutputCollector<Text, EdgeSet> output,
                Reporter reporter) throws IOException {
            // Sort the edges in the higher layer edge sets.
            TreeSet<Edge> edge_set = new TreeSet<Edge>(new EdgeComparator());
            while (values.hasNext()) {
                EdgeSet curr_set = values.next();
                // No need to do deep clone, since the edges are generated by readFields.
                edge_set.addAll(curr_set.getEdges());
            }
            // k3 is the lexically smallest edge in the set.
            String k3 = edge_set.first().getFrom() + ConstantLabels.NON_ID_CHAR + edge_set.first().getTo();
            EdgeSet v3 = new EdgeSet();
            v3.setEdges(edge_set);
            // Emit.
            output.collect(new Text(k3), v3);
        }
    }

    @Override
    public void execute() throws ProcessorExecutionException {
        JobConf conf = new JobConf(context, EdgeSetJoin.class);
        conf.setJobName("EdgeSetJoin");

        conf.setOutputKeyClass(Text.class);
        conf.setOutputValueClass(EdgeSet.class);
        conf.setMapperClass(MapClass.class);
        // Since this is a join operation, combiner is not permitted here.
        conf.setReducerClass(ReduceClass.class);
        // makes the file format suitable for machine processing.
        conf.setInputFormat(SequenceFileInputFormat.class);
        conf.setOutputFormat(SequenceFileOutputFormat.class);
        try {
            FileInputFormat.setInputPaths(conf, getSource().getPath());
            FileOutputFormat.setOutputPath(conf, getDestination().getPath());
        } catch (IllegalAccessException e1) {
            throw new ProcessorExecutionException(e1);
        }
        conf.setNumMapTasks(getMapperNum());
        conf.setNumReduceTasks(getReducerNum());
        conf.setMapOutputCompressorClass(GzipCodec.class);
        conf.setCompressMapOutput(true);

        try {
            this.runningJob = JobClient.runJob(conf);
        } catch (IOException e) {
            throw new ProcessorExecutionException(e);
        }
    }
}