Example usage for org.apache.hadoop.mapred JobConf setMapOutputKeyClass

List of usage examples for org.apache.hadoop.mapred JobConf setMapOutputKeyClass

  1. HOME
  2. Java
  3. org.apache.hadoop
  4. org.apache.hadoop.mapred.*
  5. JobConf
  6. setMapOutputKeyClass

Introduction

In this page you can find the example usage for org.apache.hadoop.mapred JobConf setMapOutputKeyClass.

Prototype

public void setMapOutputKeyClass(Class<?> theClass)

Source Link

Document

Set the key class for the map output data.

Usage

From source file:com.vsii.ttxvn.crawling.DeleteFailedDataJob.java

License:Apache License

public int run(String[] args) throws IOException {
    if (args.length < 1) {
        System.err.println("Usage: DeleteFailedDataJob <crawldb>");
        return 1;
    }/*from w  w  w.  j av a2 s .co  m*/

    String crawldb = args[0];

    SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
    long start = System.currentTimeMillis();
    LOG.info("DeleteFailedDataJob: starting at " + sdf.format(start));

    Path tempDir = new Path(getConf().get("mapred.temp.dir", ".") + "/dedup-temp-"
            + Integer.toString(new Random().nextInt(Integer.MAX_VALUE)));

    JobConf job = new NutchJob(getConf());

    job.setJobName("DeleteFailedData on " + crawldb);

    FileInputFormat.addInputPath(job, new Path(crawldb, CrawlDb.CURRENT_NAME));
    job.setInputFormat(SequenceFileInputFormat.class);

    FileOutputFormat.setOutputPath(job, tempDir);
    job.setOutputFormat(SequenceFileOutputFormat.class);

    job.setMapOutputKeyClass(BytesWritable.class);
    job.setMapOutputValueClass(CrawlDatum.class);

    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(CrawlDatum.class);

    job.setMapperClass(DBFilter.class);
    job.setReducerClass(DedupReducer.class);

    try {
        RunningJob rj = JobClient.runJob(job);
        Group g = rj.getCounters().getGroup("DeleteFailedDataJobStatus");
        if (g != null) {
            long dups = g.getCounter("Documents marked as duplicate");
            LOG.info("DeleteFailedData: " + (int) dups + " documents marked as duplicates");
        }
    } catch (final Exception e) {
        LOG.error("DeleteFailedDataJob: " + StringUtils.stringifyException(e));
        return -1;
    }

    // merge with existing crawl db
    if (LOG.isInfoEnabled()) {
        LOG.info("DeleteFailedData: Updating status of duplicate urls into crawl db.");
    }

    Path dbPath = new Path(crawldb);
    JobConf mergeJob = CrawlDb.createJob(getConf(), dbPath);
    FileInputFormat.addInputPath(mergeJob, tempDir);
    mergeJob.setReducerClass(StatusUpdateReducer.class);

    try {
        JobClient.runJob(mergeJob);
    } catch (final Exception e) {
        LOG.error("DeleteFailedDataMergeJob: " + StringUtils.stringifyException(e));
        return -1;
    }

    CrawlDb.install(mergeJob, dbPath);

    // clean up
    FileSystem fs = FileSystem.get(getConf());
    fs.delete(tempDir, true);

    long end = System.currentTimeMillis();
    LOG.info("DeleteFailedData finished at " + sdf.format(end) + ", elapsed: "
            + TimingUtil.elapsedTime(start, end));

    return 0;
}

From source file:com.yahoo.semsearch.fastlinking.io.ExtractWikipediaAnchorText.java

License:Apache License

/**
 * Extracts redirects and the target for each.
 *
 * @param inputPath/*from   w w  w .  j  a va  2  s .c o m*/
 * @param outputPath
 * @throws IOException
 */
private void task0(String inputPath, String outputPath) throws IOException {
    LOG.info("Extracting redirects (phase 0)...");
    LOG.info(" - input: " + inputPath);
    LOG.info(" - output: " + outputPath);

    JobConf conf = new JobConf(getConf(), ExtractWikipediaAnchorText.class);
    conf.setJobName(
            String.format("ExtractWikipediaAnchorText:phase0[input: %s, output: %s]", inputPath, outputPath));

    conf.setNumReduceTasks(1);

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(Text.class);

    conf.setMapperClass(MyMapper0.class);
    conf.setReducerClass(IdentityReducer.class);

    JobClient.runJob(conf);
}

From source file:com.yahoo.semsearch.fastlinking.io.ExtractWikipediaAnchorText.java

License:Apache License

/**
 * Maps from Wikipedia article to (srcID, (targetID, anchor).
 *
 * @param inputPath/*from   w  w  w.j  av a2s .  c om*/
 * @param outputPath
 * @throws IOException
 */
private void task1(String inputPath, String outputPath) throws IOException {
    LOG.info("Extracting anchor text (phase 1)...");
    LOG.info(" - input: " + inputPath);
    LOG.info(" - output: " + outputPath);

    JobConf conf = new JobConf(getConf(), ExtractWikipediaAnchorText.class);
    conf.setJobName(
            String.format("ExtractWikipediaAnchorText:phase1[input: %s, output: %s]", inputPath, outputPath));

    // 10 reducers is reasonable.
    conf.setNumReduceTasks(10);

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);

    conf.setMapOutputKeyClass(PairOfStringInt.class);
    conf.setMapOutputValueClass(PairOfStrings.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(PairOfStrings.class);

    conf.setMapperClass(MyMapper1.class);
    conf.setReducerClass(MyReducer1.class);
    conf.setPartitionerClass(MyPartitioner1.class);

    // Delete the output directory if it exists already.
    FileSystem.get(conf).delete(new Path(outputPath), true);

    JobClient.runJob(conf);
}

From source file:com.yahoo.semsearch.fastlinking.io.ExtractWikipediaAnchorText.java

License:Apache License

/**
 *
 * Maps from (srcID, (targetID, anchor) to (targetID, (anchor, count)).
 *
 * @param inputPath/*from   w  w w  .  java 2 s .  c  om*/
 * @param outputPath
 * @throws IOException
 */
private void task2(String inputPath, String outputPath, String redirPath) throws IOException {
    LOG.info("Extracting anchor text (phase 2)...");
    LOG.info(" - input: " + inputPath);
    LOG.info(" - output: " + outputPath);
    Random r = new Random();
    //String tmpOutput = "tmp-" + this.getClass().getCanonicalName() + "-" + r.nextInt(10000);
    //LOG.info( "intermediate folder for merge " + tmpOutput );

    JobConf conf = new JobConf(getConf(), ExtractWikipediaAnchorText.class);
    conf.setJobName(
            String.format("ExtractWikipediaAnchorText:phase2[input: %s, output: %s]", inputPath, outputPath));

    // Gathers everything together for convenience; feasible for Wikipedia.
    conf.setNumReduceTasks(1);

    try {
        DistributedCache.addCacheFile(new URI(redirPath + "/part-00000" + "#" + "redirs.dat"), conf);
        DistributedCache.createSymlink(conf);
    } catch (URISyntaxException e) {
        e.printStackTrace();
    }

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));
    //FileOutputFormat.setOutputPath(conf, new Path(tmpOutput));

    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(MapFileOutputFormat.class);
    // conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(HMapSIW.class);

    conf.setMapperClass(MyMapper2.class);
    conf.setReducerClass(MyReducer2.class);

    // Delete the output directory if it exists already.
    FileSystem.get(conf).delete(new Path(outputPath), true);

    JobClient.runJob(conf);
    // Clean up intermediate data.
    FileSystem.get(conf).delete(new Path(inputPath), true);

    /*
    //merge
    String finalO = outputPath+"/part-00000/data";
    FileSystem.get(conf).mkdirs( new Path( outputPath + "part-00000") );
    getMergeInHdfs( tmpOutput, finalO, conf );
    FileSystem.get(conf).delete(new Path(tmpOutput), true);
    */
}

From source file:com.yahoo.semsearch.fastlinking.io.ExtractWikipediaAnchorText.java

License:Apache License

/**
 * Extracts CF for each found anchor./*from ww  w  .jav a 2s.  c om*/
 *
 * @param inputPath
 * @param mapPath
 * @param outputPath
 * @throws IOException
 */
private void task3(String inputPath, String mapPath, String outputPath) throws IOException {
    LOG.info("Extracting anchor text (phase 3)...");
    LOG.info(" - input:   " + inputPath);
    LOG.info(" - output:  " + outputPath);
    LOG.info(" - mapping: " + mapPath);

    JobConf conf = new JobConf(getConf(), ExtractWikipediaAnchorText.class);
    conf.setJobName(
            String.format("ExtractWikipediaAnchorText:phase3[input: %s, output: %s]", inputPath, outputPath));

    conf.setNumReduceTasks(1);
    String location = "map.dat";

    try {
        DistributedCache.addCacheFile(new URI(mapPath + "/part-00000/data" + "#" + location), conf);
        //DistributedCache.addCacheFile(new URI(mapPath + "/singleentitymap.data" + "#" + location), conf);
        DistributedCache.createSymlink(conf);
    } catch (URISyntaxException e) {
        e.printStackTrace();
    }

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(MapFileOutputFormat.class);
    // conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(IntWritable.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(IntWritable.class);

    conf.setMapperClass(MyMapper3.class);
    conf.setCombinerClass(MyReducer3.class);
    conf.setReducerClass(MyReducer3.class);

    JobClient.runJob(conf);
}

From source file:com.yahoo.semsearch.fastlinking.io.ExtractWikipediaAnchorText.java

License:Apache License

/**
 * Maps from (targetID, (anchor, count)) to (anchor, (targetID, count)).
 *
 * @param inputPath//from ww  w  .  ja va 2  s  .com
 * @param outputPath
 * @throws IOException
 */
private void task4(String inputPath, String outputPath) throws IOException {
    LOG.info("Extracting anchor text (phase 4)...");
    LOG.info(" - input:   " + inputPath);
    LOG.info(" - output:  " + outputPath);

    JobConf conf = new JobConf(getConf(), ExtractWikipediaAnchorText.class);
    conf.setJobName(
            String.format("ExtractWikipediaAnchorText:phase4[input: %s, output: %s]", inputPath, outputPath));

    conf.setNumReduceTasks(1);

    //FileInputFormat.addInputPath(conf, new Path(inputPath + "/part-00000/data"));
    FileInputFormat.addInputPath(conf, new Path(inputPath + "/part-*/data"));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(MapFileOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(HMapSIW.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(HMapSIW.class);

    conf.setMapperClass(MyMapper4.class);
    conf.setReducerClass(MyReducer4.class);

    JobClient.runJob(conf);
}

From source file:com.zjy.mongo.util.MongoTool.java

License:Apache License

private int runMapredJob(final Configuration conf) {
    final JobConf job = new JobConf(conf, getClass());
    /**/*  ww  w . ja  v  a 2 s  .  c om*/
     * Any arguments specified with -D <property>=<value>
     * on the CLI will be picked up and set here
     * They override any XML level values
     * Note that -D<space> is important - no space will
     * not work as it gets picked up by Java itself
     */
    // TODO - Do we need to set job name somehow more specifically?
    // This may or may not be correct/sane
    job.setJarByClass(getClass());
    final Class<? extends org.apache.hadoop.mapred.Mapper> mapper = MapredMongoConfigUtil.getMapper(conf);

    if (LOG.isDebugEnabled()) {
        LOG.debug("Mapper Class: " + mapper);
        LOG.debug("Input URI: " + conf.get(MapredMongoConfigUtil.INPUT_URI));
    }
    job.setMapperClass(mapper);
    Class<? extends org.apache.hadoop.mapred.Reducer> combiner = MapredMongoConfigUtil.getCombiner(conf);
    if (combiner != null) {
        job.setCombinerClass(combiner);
    }
    job.setReducerClass(MapredMongoConfigUtil.getReducer(conf));

    job.setOutputFormat(MapredMongoConfigUtil.getOutputFormat(conf));
    job.setOutputKeyClass(MapredMongoConfigUtil.getOutputKey(conf));
    job.setOutputValueClass(MapredMongoConfigUtil.getOutputValue(conf));
    job.setInputFormat(MapredMongoConfigUtil.getInputFormat(conf));
    Class mapOutputKeyClass = MapredMongoConfigUtil.getMapperOutputKey(conf);
    Class mapOutputValueClass = MapredMongoConfigUtil.getMapperOutputValue(conf);

    if (mapOutputKeyClass != null) {
        job.setMapOutputKeyClass(mapOutputKeyClass);
    }
    if (mapOutputValueClass != null) {
        job.setMapOutputValueClass(mapOutputValueClass);
    }

    /**
     * Determines if the job will run verbosely e.g. print debug output
     * Only works with foreground jobs
     */
    final boolean verbose = MapredMongoConfigUtil.isJobVerbose(conf);
    /**
     * Run job in foreground aka wait for completion or background?
     */
    final boolean background = MapredMongoConfigUtil.isJobBackground(conf);
    try {
        RunningJob runningJob = JobClient.runJob(job);
        if (background) {
            LOG.info("Setting up and running MapReduce job in background.");
            return 0;
        } else {
            LOG.info("Setting up and running MapReduce job in foreground, will wait for results.  {Verbose? "
                    + verbose + "}");
            runningJob.waitForCompletion();
            return 0;
        }
    } catch (final Exception e) {
        LOG.error("Exception while executing job... ", e);
        return 1;
    }

}

From source file:contrail.stages.GraphToFasta.java

License:Open Source License

@Override
public RunningJob runJob() throws Exception {
    String inputPath = (String) stage_options.get("inputpath");
    String outputPath = (String) stage_options.get("outputpath");

    sLogger.info(" - inputpath: " + inputPath);
    sLogger.info(" - outputpath: " + outputPath);

    JobConf conf = new JobConf(GraphToFasta.class);

    AvroJob.setInputSchema(conf, GraphNodeData.SCHEMA$);

    initializeJobConfiguration(conf);/* w ww  .  jav a  2s . co m*/

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));
    AvroInputFormat<GraphNodeData> input_format = new AvroInputFormat<GraphNodeData>();
    conf.setInputFormat(input_format.getClass());
    conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);

    // Make it mapper only.
    conf.setNumReduceTasks(0);
    conf.setMapperClass(GraphToFastqMapper.class);

    if (stage_options.containsKey("writeconfig")) {
        writeJobConfig(conf);
    } else {
        // Delete the output directory if it exists already
        Path out_path = new Path(outputPath);
        if (FileSystem.get(conf).exists(out_path)) {
            // TODO(jlewi): We should only delete an existing directory
            // if explicitly told to do so.
            sLogger.info("Deleting output path: " + out_path.toString() + " " + "because it already exists.");
            FileSystem.get(conf).delete(out_path, true);
        }

        long starttime = System.currentTimeMillis();
        RunningJob result = JobClient.runJob(conf);
        long endtime = System.currentTimeMillis();

        float diff = (float) ((endtime - starttime) / 1000.0);

        System.out.println("Runtime: " + diff + " s");
        return result;
    }
    return null;
}

From source file:Corrector.Correction.java

License:Apache License

public RunningJob run(String inputPath, String outputPath) throws Exception {
    sLogger.info("Tool name: Correction [0/7]");
    sLogger.info(" - input: " + inputPath);
    sLogger.info(" - output: " + outputPath);

    JobConf conf = new JobConf(Correction.class);
    conf.setJobName("Correction " + inputPath + " " + Config.K);

    Config.initializeConfiguration(conf);

    FileInputFormat.addInputPaths(conf, inputPath);
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(TextInputFormat.class);
    conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);
    //conf.setBoolean("mapred.output.compress", true);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(Text.class);

    conf.setMapperClass(CorrectionMapper.class);
    conf.setReducerClass(CorrectionReducer.class);

    //delete the output directory if it exists already
    FileSystem.get(conf).delete(new Path(outputPath), true);

    return JobClient.runJob(conf);
}

From source file:Corrector.FindError.java

License:Apache License

public RunningJob run(String inputPath, String outputPath, int idx, String hkmerlist) throws Exception {
    sLogger.info("Tool name: FindError");
    sLogger.info(" - input: " + inputPath);
    sLogger.info(" - output: " + outputPath);

    JobConf conf = new JobConf(FindError.class);
    conf.setJobName("FindError " + inputPath + " " + Config.K);
    conf.setLong("IDX", idx);
    //\\/*w  w  w .  j a  va 2 s . c  om*/
    DistributedCache.addCacheFile(new URI(hkmerlist), conf);
    //\\

    Config.initializeConfiguration(conf);

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(TextInputFormat.class);
    conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(Text.class);

    conf.setMapperClass(FindErrorMapper.class);
    conf.setReducerClass(FindErrorReducer.class);

    //delete the output directory if it exists already
    FileSystem.get(conf).delete(new Path(outputPath), true);

    return JobClient.runJob(conf);
}