001/**
002 * Copyright (c) 2011, The University of Southampton and the individual contributors.
003 * All rights reserved.
004 *
005 * Redistribution and use in source and binary forms, with or without modification,
006 * are permitted provided that the following conditions are met:
007 *
008 *   *  Redistributions of source code must retain the above copyright notice,
009 *      this list of conditions and the following disclaimer.
010 *
011 *   *  Redistributions in binary form must reproduce the above copyright notice,
012 *      this list of conditions and the following disclaimer in the documentation
013 *      and/or other materials provided with the distribution.
014 *
015 *   *  Neither the name of the University of Southampton nor the names of its
016 *      contributors may be used to endorse or promote products derived from this
017 *      software without specific prior written permission.
018 *
019 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
020 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
021 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
022 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
023 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
024 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
025 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
026 * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
027 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
028 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
029 */
030package org.openimaj.hadoop.tools.downloader;
031
032import org.apache.hadoop.conf.Configuration;
033import org.apache.hadoop.conf.Configured;
034import org.apache.hadoop.io.BytesWritable;
035import org.apache.hadoop.io.SequenceFile.CompressionType;
036import org.apache.hadoop.io.Text;
037import org.apache.hadoop.io.compress.DefaultCodec;
038import org.apache.hadoop.mapreduce.Job;
039import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
040import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
041import org.apache.hadoop.mapreduce.lib.map.MultithreadedMapper;
042import org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat;
043import org.apache.hadoop.util.Tool;
044import org.apache.hadoop.util.ToolRunner;
045
046/**
047 * A tool for the distributed downloading of data into Hadoop SequenceFiles.
048 * 
049 * @author Sina Samangooei (ss@ecs.soton.ac.uk)
050 * @author Jonathon Hare (jsh2@ecs.soton.ac.uk)
051 * 
052 */
053public class HadoopDownloader extends Configured implements Tool {
054        protected static final String ARGS_KEY = "hadoop.downloader.args";
055
056        @Override
057        public int run(String[] args) throws Exception {
058                final HadoopDownloaderOptions options = new HadoopDownloaderOptions(args);
059                options.prepare(true);
060
061                final Job job = new Job(getConf());
062
063                job.setJarByClass(HadoopDownloader.class);
064                job.setJobName("Hadoop Downloader Utility");
065
066                job.setOutputKeyClass(Text.class);
067                job.setOutputValueClass(BytesWritable.class);
068
069                if (options.getNumberOfThreads() <= 1) {
070                        job.setMapperClass(DownloadMapper.class);
071                } else {
072                        job.setMapperClass(MultithreadedMapper.class);
073                        MultithreadedMapper.setMapperClass(job, DownloadMapper.class);
074                        MultithreadedMapper.setNumberOfThreads(job, options.getNumberOfThreads());
075                }
076
077                job.setInputFormatClass(TextInputFormat.class);
078                job.setOutputFormatClass(SequenceFileOutputFormat.class);
079
080                job.setNumReduceTasks(options.getNumberOfReducers());
081
082                job.getConfiguration().setStrings(ARGS_KEY, args);
083
084                FileInputFormat.setInputPaths(job, options.getInputPaths());
085                SequenceFileOutputFormat.setOutputPath(job, options.getOutputPath());
086                SequenceFileOutputFormat.setCompressOutput(job, true);
087                SequenceFileOutputFormat.setOutputCompressorClass(job, DefaultCodec.class);
088                SequenceFileOutputFormat.setOutputCompressionType(job, CompressionType.BLOCK);
089
090                job.waitForCompletion(true);
091
092                return 0;
093        }
094
095        /**
096         * Main program entry point
097         * 
098         * @param args
099         *            command-line arguments
100         * @throws Exception
101         *             if an error occurs
102         */
103        public static void main(String[] args) throws Exception {
104                final int res = ToolRunner.run(new Configuration(), new HadoopDownloader(), args);
105
106                System.exit(res);
107        }
108}