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

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

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

Introduction

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

Prototype

public void set(String name, String value)

Source Link

Document

Set the value of the name property.

Usage

From source file:com.davidgildeh.hadoop.utils.FileUtils.java

License:Apache License

/**
 * Given a path to a valid key/value properties file on HDFS, all values will be
 * loaded into the provided JobConf to provide configuration properties for the 
 * Hadoop Job/*from w ww . ja v  a 2 s . c  o  m*/
 * 
 * @param jobConf           The JobConf to load property values into
 * @param path              The path to the properties file
 * @return                  The JobConf with loaded properties values added     
 */
public static JobConf loadJobConf(JobConf jobConf, String path) throws IOException {

    Properties propFile = loadPropertiesFile(path);

    // Loop through all properties in properties file
    for (Object keyObject : propFile.keySet()) {
        String key = (String) keyObject;
        String value = propFile.getProperty(key);
        jobConf.set(key, value);
        if (LOG.isDebugEnabled()) {
            LOG.debug("Loaded Configuration Property " + key + ": " + value);
        }
    }

    return jobConf;
}

From source file:com.digitalpebble.behemoth.ClassifierJob.java

License:Apache License

public int run(String[] args) throws Exception {

    Options options = new Options();
    // automatically generate the help statement
    HelpFormatter formatter = new HelpFormatter();
    // create the parser
    CommandLineParser parser = new GnuParser();

    options.addOption("h", "help", false, "print this message");
    options.addOption("i", "input", true, "input Behemoth corpus");
    options.addOption("o", "output", true, "output Behemoth corpus");
    options.addOption("m", "model", true, "location of the model");

    // parse the command line arguments
    CommandLine line = null;// w  w  w  .j ava2 s  . c o  m
    try {
        line = parser.parse(options, args);
        String input = line.getOptionValue("i");
        String output = line.getOptionValue("o");
        String model = line.getOptionValue("m");
        if (line.hasOption("help")) {
            formatter.printHelp("ClassifierJob", options);
            return 0;
        }
        if (model == null | input == null | output == null) {
            formatter.printHelp("ClassifierJob", options);
            return -1;
        }
    } catch (ParseException e) {
        formatter.printHelp("ClassifierJob", options);
    }

    final FileSystem fs = FileSystem.get(getConf());

    Path inputPath = new Path(line.getOptionValue("i"));
    Path outputPath = new Path(line.getOptionValue("o"));
    String modelPath = line.getOptionValue("m");

    JobConf job = new JobConf(getConf());

    // push the model file to the DistributedCache
    DistributedCache.addCacheArchive(new URI(modelPath), job);

    job.setJarByClass(this.getClass());

    job.setJobName("ClassifierJob : " + inputPath.toString());

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

    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(BehemothDocument.class);
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(BehemothDocument.class);

    job.setMapperClass(TextClassifierMapper.class);
    job.setNumReduceTasks(0);

    FileInputFormat.addInputPath(job, inputPath);
    FileOutputFormat.setOutputPath(job, outputPath);

    job.set(modelNameParam, modelPath);

    try {
        JobClient.runJob(job);
    } catch (Exception e) {
        e.printStackTrace();
    } finally {
    }

    return 0;
}

From source file:com.digitalpebble.behemoth.gate.GATEDriver.java

License:Apache License

public int run(String[] args) throws Exception {

    final FileSystem fs = FileSystem.get(getConf());

    if (args.length < 3 | args.length > 4) {
        String syntax = "com.digitalpebble.behemoth.gate.GATEDriver in out path_gate_file [-XML]";
        System.err.println(syntax);
        return -1;
    }// w  ww .j  a  v  a 2s . co m

    boolean dumpGATEXML = false;

    for (String arg : args) {
        if (arg.equalsIgnoreCase("-xml"))
            dumpGATEXML = true;
    }

    Path inputPath = new Path(args[0]);
    Path outputPath = new Path(args[1]);
    String zip_application_path = args[2];

    // check that the GATE application has been stored on HDFS
    Path zap = new Path(zip_application_path);
    if (fs.exists(zap) == false) {
        System.err
                .println("The GATE application " + zip_application_path + "can't be found on HDFS - aborting");
        return -1;
    }

    JobConf job = new JobConf(getConf());
    // MUST not forget the line below
    job.setJarByClass(this.getClass());

    job.setJobName("Processing " + args[0] + " with GATE application from " + zip_application_path);

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

    job.setOutputKeyClass(Text.class);

    if (dumpGATEXML) {
        job.setOutputValueClass(Text.class);
        job.setMapperClass(GATEXMLMapper.class);
    } else {
        job.setOutputValueClass(BehemothDocument.class);
        job.setMapperClass(GATEMapper.class);
    }

    // detect if any filters have been defined
    // and activate the reducer accordingly
    boolean isFilterRequired = BehemothReducer.isRequired(job);
    if (isFilterRequired)
        job.setReducerClass(BehemothReducer.class);
    else {
        job.setNumReduceTasks(0);
    }

    FileInputFormat.addInputPath(job, inputPath);
    FileOutputFormat.setOutputPath(job, outputPath);

    // push the zipped_gate_application onto the DistributedCache
    DistributedCache.addCacheArchive(new URI(zip_application_path), job);

    job.set("gate.application.path", zip_application_path.toString());

    try {
        long start = System.currentTimeMillis();
        JobClient.runJob(job);
        long finish = System.currentTimeMillis();
        if (LOG.isInfoEnabled()) {
            LOG.info("GATEDriver completed. Timing: " + (finish - start) + " ms");
        }
    } catch (Exception e) {
        LOG.error("Exception caught", e);
        // leave even partial output
        // fs.delete(outputPath, true);
    } finally {
    }

    return 0;
}

From source file:com.digitalpebble.behemoth.solr.LucidWorksIndexerJob.java

License:Apache License

public int run(String[] args) throws Exception {

    final FileSystem fs = FileSystem.get(getConf());

    if (args.length != 2) {
        String syntax = "com.digitalpebble.solr.LucidWorksIndexerJob in solrURL";
        System.err.println(syntax);
        return -1;
    }/*from   w  w  w.  ja v  a2  s  .  co m*/

    Path inputPath = new Path(args[0]);
    String solrURL = args[1];

    JobConf job = new JobConf(getConf());

    job.setJarByClass(this.getClass());

    job.setJobName("Indexing " + inputPath + " into LucidWorks");

    job.setInputFormat(SequenceFileInputFormat.class);
    job.setOutputFormat(LucidWorksOutputFormat.class);

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

    job.setMapperClass(IdentityMapper.class);
    // no reducer : send straight to SOLR at end of mapping
    job.setNumReduceTasks(0);

    FileInputFormat.addInputPath(job, inputPath);
    final Path tmp = new Path("tmp_" + System.currentTimeMillis() + "-" + new Random().nextInt());
    FileOutputFormat.setOutputPath(job, tmp);

    job.set("solr.server.url", solrURL);

    try {
        long start = System.currentTimeMillis();
        JobClient.runJob(job);
        long finish = System.currentTimeMillis();
        if (LOG.isInfoEnabled()) {
            LOG.info("LucidWorksIndexerJob completed. Time " + (finish - start) + " ms");
        }
    } catch (Exception e) {
        LOG.error(e);
    } finally {
        fs.delete(tmp, true);
    }

    return 0;
}

From source file:com.digitalpebble.behemoth.solr.SOLRIndexerJob.java

License:Apache License

public int run(String[] args) throws Exception {

    final FileSystem fs = FileSystem.get(getConf());

    if (args.length != 2) {
        String syntax = "com.digitalpebble.solr.SOLRIndexerJob in solrURL";
        System.err.println(syntax);
        return -1;
    }/*  w ww  .j  av a2s  . co m*/

    Path inputPath = new Path(args[0]);
    String solrURL = args[1];

    JobConf job = new JobConf(getConf());

    job.setJarByClass(this.getClass());

    job.setJobName("Indexing " + inputPath + " into SOLR");

    job.setInputFormat(SequenceFileInputFormat.class);
    job.setOutputFormat(SOLROutputFormat.class);

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

    job.setMapperClass(IdentityMapper.class);
    // no reducer : send straight to SOLR at end of mapping
    job.setNumReduceTasks(0);

    FileInputFormat.addInputPath(job, inputPath);
    final Path tmp = new Path("tmp_" + System.currentTimeMillis() + "-" + new Random().nextInt());
    FileOutputFormat.setOutputPath(job, tmp);

    job.set("solr.server.url", solrURL);

    try {
        long start = System.currentTimeMillis();
        JobClient.runJob(job);
        long finish = System.currentTimeMillis();
        if (LOG.isInfoEnabled()) {
            LOG.info("SOLRIndexerJob completed. Timing: " + (finish - start) + " ms");
        }
    } catch (Exception e) {
        LOG.error(e);
    } finally {
        fs.delete(tmp, true);
    }

    return 0;
}

From source file:com.digitalpebble.behemoth.solr.TestSOLRWriter.java

License:Apache License

@Test
public void testFieldMappings() throws IOException {
    JobConf conf = new JobConf();
    conf.set("solr.server.url", "http://example.org");
    conf.set("solr.f.person", "Person.string");
    conf.set("solr.f.personTitle", "Person.title");
    conf.set("solr.f.location", "Location");

    Progressable progress = new Progressable() {
        @Override//from  w w w  . ja va  2  s .c om
        public void progress() {

        }
    };
    SOLRWriter writer = new SOLRWriter(progress);
    writer.open(conf, "test");

    assertEquals(writer.getFieldMapping().size(), 2);
    assertNotNull(writer.getFieldMapping().get("Person"));
    assertEquals(writer.getFieldMapping().get("Person").size(), 2);
    assertEquals(writer.getFieldMapping().get("Person").get("string"), "person");
    assertEquals(writer.getFieldMapping().get("Person").get("title"), "personTitle");
    assertNotNull(writer.getFieldMapping().get("Location"));
    assertEquals(writer.getFieldMapping().get("Location").size(), 1);
    assertEquals(writer.getFieldMapping().get("Location").get("*"), "location");
}

From source file:com.digitalpebble.behemoth.tika.TikaDriver.java

License:Apache License

public int run(String[] args) throws Exception {

    final FileSystem fs = FileSystem.get(getConf());
    GroupBuilder gBuilder = new GroupBuilder().withName("Options:");
    List<Option> options = new ArrayList<Option>();
    Option inputOpt = buildOption("input", "i", "The input path", true, true, null);
    options.add(inputOpt);//from   w  w w  .ja  v a2  s  . c  om
    Option outOpt = buildOption("output", "o", "The output path", true, true, null);
    options.add(outOpt);
    Option tikaOpt = buildOption("tikaProcessor", "t",
            "The fully qualified name of a TikaProcessor class that handles the extraction (optional)", true,
            false, null);
    options.add(tikaOpt);
    Option mimeTypeOpt = buildOption("mimeType", "m", "The mime type to use (optional)", true, false, "");
    options.add(mimeTypeOpt);
    for (Option opt : options) {
        gBuilder = gBuilder.withOption(opt);
    }

    Group group = gBuilder.create();

    try {
        Parser parser = new Parser();
        parser.setGroup(group);
        // TODO catch exceptions with parsing of opts
        CommandLine cmdLine = parser.parse(args);
        Path inputPath = new Path(cmdLine.getValue(inputOpt).toString());
        Path outputPath = new Path(cmdLine.getValue(outOpt).toString());
        String handlerName = null;
        if (cmdLine.hasOption(tikaOpt)) {
            handlerName = cmdLine.getValue(tikaOpt).toString();
        }

        JobConf job = new JobConf(getConf());
        job.setJarByClass(this.getClass());

        if (cmdLine.hasOption(mimeTypeOpt)) {
            String mimeType = cmdLine.getValue(mimeTypeOpt).toString();
            job.set(TikaConstants.TIKA_MIME_TYPE_KEY, mimeType);
        }

        if (handlerName != null && handlerName.equals("") == false) {
            job.set(TIKA_PROCESSOR_KEY, handlerName);
        }

        job.setJobName("Tika : " + inputPath.toString());

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

        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(BehemothDocument.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(BehemothDocument.class);

        job.setMapperClass(TikaMapper.class);

        boolean isFilterRequired = BehemothReducer.isRequired(job);
        if (isFilterRequired)
            job.setReducerClass(BehemothReducer.class);
        else {
            job.setNumReduceTasks(0);
        }

        FileInputFormat.addInputPath(job, inputPath);
        FileOutputFormat.setOutputPath(job, outputPath);

        try {
            long start = System.currentTimeMillis();
            JobClient.runJob(job);
            long finish = System.currentTimeMillis();
            if (log.isInfoEnabled()) {
                log.info("TikaDriver completed. Timing: " + (finish - start) + " ms");
            }
        } catch (Exception e) {
            log.error("Exception", e);
            return -1;
            // don't delete the output as some of it could be used
            // fs.delete(outputPath, true);
        } finally {
        }

    } catch (OptionException e) {
        log.error("OptionException", e.getMessage());
        HelpFormatter formatter = new HelpFormatter();
        formatter.setGroup(group);
        formatter.print();
        return -1;
    }

    return 0;
}

From source file:com.digitalpebble.behemoth.uima.UIMADriver.java

License:Apache License

public int run(String[] args) throws Exception {

    final FileSystem fs = FileSystem.get(getConf());

    if (args.length != 3) {
        String syntax = "com.digitalpebble.behemoth.uima.UIMADriver in out path_pear_file";
        System.err.println(syntax);
        return -1;
    }/*from w ww  .  j av a2s.  c om*/

    Path inputPath = new Path(args[0]);
    Path outputPath = new Path(args[1]);
    String pearPath = args[2];

    // check that the GATE application has been stored on HDFS
    Path zap = new Path(pearPath);
    if (fs.exists(zap) == false) {
        System.err.println("The UIMA application " + pearPath + "can't be found on HDFS - aborting");
        return -1;
    }

    JobConf job = new JobConf(getConf());
    job.setJarByClass(this.getClass());
    job.setJobName("Processing with UIMA application : " + pearPath);

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

    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(BehemothDocument.class);
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(BehemothDocument.class);

    job.setMapperClass(UIMAMapper.class);

    job.setNumReduceTasks(0);

    FileInputFormat.addInputPath(job, inputPath);
    FileOutputFormat.setOutputPath(job, outputPath);

    // push the UIMA pear onto the DistributedCache
    DistributedCache.addCacheFile(new URI(pearPath), job);

    job.set("uima.pear.path", pearPath);

    try {
        long start = System.currentTimeMillis();
        JobClient.runJob(job);
        long finish = System.currentTimeMillis();
        if (LOG.isInfoEnabled()) {
            LOG.info("UIMADriver completed. Timing: " + (finish - start) + " ms");
        }
    } catch (Exception e) {
        LOG.error("Exception", e);
        fs.delete(outputPath, true);
    } finally {
    }

    return 0;
}

From source file:com.ebay.erl.mobius.core.builder.TSVDataset.java

License:Apache License

@Override
public JobConf createJobConf(byte jobSequenceNumber) throws IOException {
    JobConf conf = super.createJobConf(jobSequenceNumber);
    if (!this.delimiter.equals("\t")) {
        conf.set(this.getID() + ".delimiter", SerializableUtil.serializeToBase64(delimiter));
    }/* www. j a  va 2 s.  co m*/
    return conf;
}

From source file:com.ebay.erl.mobius.core.datajoin.EvenlyPartitioner.java

License:Apache License

/**
 * Set the path to the SequenceFile storing the sorted partition keyset.
 * It must be the case that for <tt>R</tt> reduces, there are <tt>R-1</tt>
 * keys in the SequenceFile./*from w  w  w.ja v a 2  s  .co m*/
 */
public static void setPartitionFile(JobConf job, Path p) {
    job.set("total.order.partitioner.path", p.toString());
}