Java Code Examples for org.apache.hadoop.mapred.JobConf#setOutputKeyClass()

public void testEmptyJoin() throws Exception {
  JobConf job = new JobConf();
  Path base = cluster.getFileSystem().makeQualified(new Path("/empty"));
  Path[] src = { new Path(base,"i0"), new Path("i1"), new Path("i2") };
  job.set("mapreduce.join.expr", CompositeInputFormat.compose("outer",
      Fake_IF.class, src));
  job.setInputFormat(CompositeInputFormat.class);
  FileOutputFormat.setOutputPath(job, new Path(base, "out"));

  job.setMapperClass(IdentityMapper.class);
  job.setReducerClass(IdentityReducer.class);
  job.setOutputKeyClass(IncomparableKey.class);
  job.setOutputValueClass(NullWritable.class);

  JobClient.runJob(job);
  base.getFileSystem(job).delete(base, true);
}
 

/**
 * Run the test
 * 
 * @throws IOException on error
 */
public static void runTests() throws IOException {
  config.setLong("io.bytes.per.checksum", bytesPerChecksum);
  
  JobConf job = new JobConf(config, NNBench.class);

  job.setJobName("NNBench-" + operation);
  FileInputFormat.setInputPaths(job, new Path(baseDir, CONTROL_DIR_NAME));
  job.setInputFormat(SequenceFileInputFormat.class);
  
  // Explicitly set number of max map attempts to 1.
  job.setMaxMapAttempts(1);
  
  // Explicitly turn off speculative execution
  job.setSpeculativeExecution(false);

  job.setMapperClass(NNBenchMapper.class);
  job.setReducerClass(NNBenchReducer.class);

  FileOutputFormat.setOutputPath(job, new Path(baseDir, OUTPUT_DIR_NAME));
  job.setOutputKeyClass(Text.class);
  job.setOutputValueClass(Text.class);
  job.setNumReduceTasks((int) numberOfReduces);
  JobClient.runJob(job);
}
 

/**
 * Creates a simple copy job.
 * 
 * @param indirs List of input directories.
 * @param outdir Output directory.
 * @return JobConf initialised for a simple copy job.
 * @throws Exception If an error occurs creating job configuration.
 */
static JobConf createCopyJob(List<Path> indirs, Path outdir) throws Exception {

  Configuration defaults = new Configuration();
  JobConf theJob = new JobConf(defaults, TestJobControl.class);
  theJob.setJobName("DataMoveJob");

  FileInputFormat.setInputPaths(theJob, indirs.toArray(new Path[0]));
  theJob.setMapperClass(DataCopy.class);
  FileOutputFormat.setOutputPath(theJob, outdir);
  theJob.setOutputKeyClass(Text.class);
  theJob.setOutputValueClass(Text.class);
  theJob.setReducerClass(DataCopy.class);
  theJob.setNumMapTasks(12);
  theJob.setNumReduceTasks(4);
  return theJob;
}
 

private void runIOTest(
        Class<? extends Mapper<Text, LongWritable, Text, Text>> mapperClass, 
        Path outputDir) throws IOException {
  JobConf job = new JobConf(config, TestDFSIO.class);

  FileInputFormat.setInputPaths(job, getControlDir(config));
  job.setInputFormat(SequenceFileInputFormat.class);

  job.setMapperClass(mapperClass);
  job.setReducerClass(AccumulatingReducer.class);

  FileOutputFormat.setOutputPath(job, outputDir);
  job.setOutputKeyClass(Text.class);
  job.setOutputValueClass(Text.class);
  job.setNumReduceTasks(1);
  JobClient.runJob(job);
}
 

public static void seekTest(FileSystem fs, boolean fastCheck)
  throws Exception {

  fs.delete(READ_DIR, true);

  JobConf job = new JobConf(conf, TestFileSystem.class);
  job.setBoolean("fs.test.fastCheck", fastCheck);

  FileInputFormat.setInputPaths(job,CONTROL_DIR);
  job.setInputFormat(SequenceFileInputFormat.class);

  job.setMapperClass(SeekMapper.class);
  job.setReducerClass(LongSumReducer.class);

  FileOutputFormat.setOutputPath(job, READ_DIR);
  job.setOutputKeyClass(UTF8.class);
  job.setOutputValueClass(LongWritable.class);
  job.setNumReduceTasks(1);
  JobClient.runJob(job);
}
 

/**
 * Creates a simple copy job.
 * 
 * @param indirs List of input directories.
 * @param outdir Output directory.
 * @return JobConf initialised for a simple copy job.
 * @throws Exception If an error occurs creating job configuration.
 */
static JobConf createCopyJob(List<Path> indirs, Path outdir) throws Exception {

  Configuration defaults = new Configuration();
  JobConf theJob = new JobConf(defaults, TestJobControl.class);
  theJob.setJobName("DataMoveJob");

  FileInputFormat.setInputPaths(theJob, indirs.toArray(new Path[0]));
  theJob.setMapperClass(DataCopy.class);
  FileOutputFormat.setOutputPath(theJob, outdir);
  theJob.setOutputKeyClass(Text.class);
  theJob.setOutputValueClass(Text.class);
  theJob.setReducerClass(DataCopy.class);
  theJob.setNumMapTasks(12);
  theJob.setNumReduceTasks(4);
  return theJob;
}
 

public void processDumpJob(String crawlDb, String output, Configuration config, String format, String regex, String status, Integer retry) throws IOException {
  if (LOG.isInfoEnabled()) {
    LOG.info("CrawlDb dump: starting");
    LOG.info("CrawlDb db: " + crawlDb);
  }

  Path outFolder = new Path(output);

  JobConf job = new NutchJob(config);
  job.setJobName("dump " + crawlDb);

  FileInputFormat.addInputPath(job, new Path(crawlDb, CrawlDb.CURRENT_NAME));
  job.setInputFormat(SequenceFileInputFormat.class);
  FileOutputFormat.setOutputPath(job, outFolder);

  if (format.equals("csv")) {
    job.setOutputFormat(CrawlDatumCsvOutputFormat.class);
  }
  else if (format.equals("crawldb")) {
    job.setOutputFormat(MapFileOutputFormat.class);
  } else {
    job.setOutputFormat(TextOutputFormat.class);
  }

  if (status != null) job.set("status", status);
  if (regex != null) job.set("regex", regex);
  if (retry != null) job.setInt("retry", retry);
  
  job.setMapperClass(CrawlDbDumpMapper.class);
  job.setOutputKeyClass(Text.class);
  job.setOutputValueClass(CrawlDatum.class);

  JobClient.runJob(job);
  if (LOG.isInfoEnabled()) { LOG.info("CrawlDb dump: done"); }
}
 

/**
 * Runs the inverter job. The inverter job flips outlinks to inlinks to be
 * passed into the analysis job.
 * 
 * The inverter job takes a link loops database if it exists. It is an
 * optional componenet of link analysis due to its extreme computational and
 * space requirements but it can be very useful is weeding out and eliminating
 * link farms and other spam pages.
 * 
 * @param nodeDb The node database to use.
 * @param outlinkDb The outlink database to use.
 * @param loopDb The loop database to use if it exists.
 * @param output The output directory.
 * 
 * @throws IOException If an error occurs while running the inverter job.
 */
private void runInverter(Path nodeDb, Path outlinkDb, Path loopDb, Path output)
  throws IOException {

  // configure the inverter
  JobConf inverter = new NutchJob(getConf());
  inverter.setJobName("LinkAnalysis Inverter");
  FileInputFormat.addInputPath(inverter, nodeDb);
  FileInputFormat.addInputPath(inverter, outlinkDb);

  // add the loop database if it exists, isn't null
  if (loopDb != null) {
    FileInputFormat.addInputPath(inverter, loopDb);
  }
  FileOutputFormat.setOutputPath(inverter, output);
  inverter.setInputFormat(SequenceFileInputFormat.class);
  inverter.setMapperClass(Inverter.class);
  inverter.setReducerClass(Inverter.class);
  inverter.setMapOutputKeyClass(Text.class);
  inverter.setMapOutputValueClass(ObjectWritable.class);
  inverter.setOutputKeyClass(Text.class);
  inverter.setOutputValueClass(LinkDatum.class);
  inverter.setOutputFormat(SequenceFileOutputFormat.class);
  inverter.setBoolean("mapreduce.fileoutputcommitter.marksuccessfuljobs", false);

  // run the inverter job
  LOG.info("Starting inverter job");
  try {
    JobClient.runJob(inverter);
  }
  catch (IOException e) {
    LOG.error(StringUtils.stringifyException(e));
    throw e;
  }
  LOG.info("Finished inverter job.");
}
 

/**
 * Runs the initializer job. The initializer job sets up the nodes with a
 * default starting score for link analysis.
 * 
 * @param nodeDb The node database to use.
 * @param output The job output directory.
 * 
 * @throws IOException If an error occurs while running the initializer job.
 */
private void runInitializer(Path nodeDb, Path output)
  throws IOException {

  // configure the initializer
  JobConf initializer = new NutchJob(getConf());
  initializer.setJobName("LinkAnalysis Initializer");
  FileInputFormat.addInputPath(initializer, nodeDb);
  FileOutputFormat.setOutputPath(initializer, output);
  initializer.setInputFormat(SequenceFileInputFormat.class);
  initializer.setMapperClass(Initializer.class);
  initializer.setMapOutputKeyClass(Text.class);
  initializer.setMapOutputValueClass(Node.class);
  initializer.setOutputKeyClass(Text.class);
  initializer.setOutputValueClass(Node.class);
  initializer.setOutputFormat(MapFileOutputFormat.class);
  initializer.setBoolean("mapreduce.fileoutputcommitter.marksuccessfuljobs", false);

  // run the initializer
  LOG.info("Starting initialization job");
  try {
    JobClient.runJob(initializer);
  }
  catch (IOException e) {
    LOG.error(StringUtils.stringifyException(e));
    throw e;
  }
  LOG.info("Finished initialization job.");
}
 

/**
 * Runs the inverter job. The inverter job flips outlinks to inlinks to be
 * passed into the analysis job.
 * 
 * The inverter job takes a link loops database if it exists. It is an
 * optional componenet of link analysis due to its extreme computational and
 * space requirements but it can be very useful is weeding out and eliminating
 * link farms and other spam pages.
 * 
 * @param nodeDb The node database to use.
 * @param outlinkDb The outlink database to use.
 * @param loopDb The loop database to use if it exists.
 * @param output The output directory.
 * 
 * @throws IOException If an error occurs while running the inverter job.
 */
private void runInverter(Path nodeDb, Path outlinkDb, Path loopDb, Path output)
  throws IOException {

  // configure the inverter
  JobConf inverter = new NutchJob(getConf());
  inverter.setJobName("LinkAnalysis Inverter");
  FileInputFormat.addInputPath(inverter, nodeDb);
  FileInputFormat.addInputPath(inverter, outlinkDb);

  // add the loop database if it exists, isn't null
  if (loopDb != null) {
    FileInputFormat.addInputPath(inverter, loopDb);
  }
  FileOutputFormat.setOutputPath(inverter, output);
  inverter.setInputFormat(SequenceFileInputFormat.class);
  inverter.setMapperClass(Inverter.class);
  inverter.setReducerClass(Inverter.class);
  inverter.setMapOutputKeyClass(Text.class);
  inverter.setMapOutputValueClass(ObjectWritable.class);
  inverter.setOutputKeyClass(Text.class);
  inverter.setOutputValueClass(LinkDatum.class);
  inverter.setOutputFormat(SequenceFileOutputFormat.class);
  inverter.setBoolean("mapreduce.fileoutputcommitter.marksuccessfuljobs", false);

  // run the inverter job
  LOG.info("Starting inverter job");
  try {
    JobClient.runJob(inverter);
  }
  catch (IOException e) {
    LOG.error(StringUtils.stringifyException(e));
    throw e;
  }
  LOG.info("Finished inverter job.");
}
 

/**
 * Runs the link analysis job. The link analysis job applies the link rank
 * formula to create a score per url and stores that score in the NodeDb.
 * 
 * Typically the link analysis job is run a number of times to allow the link
 * rank scores to converge.
 * 
 * @param nodeDb The node database from which we are getting previous link
 * rank scores.
 * @param inverted The inverted inlinks
 * @param output The link analysis output.
 * @param iteration The current iteration number.
 * @param numIterations The total number of link analysis iterations
 * 
 * @throws IOException If an error occurs during link analysis.
 */
private void runAnalysis(Path nodeDb, Path inverted, Path output,
  int iteration, int numIterations, float rankOne)
  throws IOException {

  JobConf analyzer = new NutchJob(getConf());
  analyzer.set("link.analyze.iteration", String.valueOf(iteration + 1));
  analyzer.setJobName("LinkAnalysis Analyzer, iteration " + (iteration + 1)
    + " of " + numIterations);
  FileInputFormat.addInputPath(analyzer, nodeDb);
  FileInputFormat.addInputPath(analyzer, inverted);
  FileOutputFormat.setOutputPath(analyzer, output);
  analyzer.set("link.analyze.rank.one", String.valueOf(rankOne));
  analyzer.setMapOutputKeyClass(Text.class);
  analyzer.setMapOutputValueClass(ObjectWritable.class);
  analyzer.setInputFormat(SequenceFileInputFormat.class);
  analyzer.setMapperClass(Analyzer.class);
  analyzer.setReducerClass(Analyzer.class);
  analyzer.setOutputKeyClass(Text.class);
  analyzer.setOutputValueClass(Node.class);
  analyzer.setOutputFormat(MapFileOutputFormat.class);
  analyzer.setBoolean("mapreduce.fileoutputcommitter.marksuccessfuljobs", false);

  LOG.info("Starting analysis job");
  try {
    JobClient.runJob(analyzer);
  }
  catch (IOException e) {
    LOG.error(StringUtils.stringifyException(e));
    throw e;
  }
  LOG.info("Finished analysis job.");
}
 

/**
 * <p>Creates the arc files to segments job.</p>
 * 
 * @param arcFiles The path to the directory holding the arc files
 * @param segmentsOutDir The output directory for writing the segments
 * 
 * @throws IOException If an IO error occurs while running the job.
 */
public void createSegments(Path arcFiles, Path segmentsOutDir)
  throws IOException {

  SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
  long start = System.currentTimeMillis();
  if (LOG.isInfoEnabled()) {
    LOG.info("ArcSegmentCreator: starting at " + sdf.format(start));
    LOG.info("ArcSegmentCreator: arc files dir: " + arcFiles);
  }

  JobConf job = new NutchJob(getConf());
  job.setJobName("ArcSegmentCreator " + arcFiles);
  String segName = generateSegmentName();
  job.set(Nutch.SEGMENT_NAME_KEY, segName);
  FileInputFormat.addInputPath(job, arcFiles);
  job.setInputFormat(ArcInputFormat.class);
  job.setMapperClass(ArcSegmentCreator.class);
  FileOutputFormat.setOutputPath(job, new Path(segmentsOutDir, segName));
  job.setOutputFormat(FetcherOutputFormat.class);
  job.setOutputKeyClass(Text.class);
  job.setOutputValueClass(NutchWritable.class);

  JobClient.runJob(job);

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

public static JobConf createDataJoinJob(String args[]) throws IOException {

    String inputDir = args[0];
    String outputDir = args[1];
    Class inputFormat = SequenceFileInputFormat.class;
    if (args[2].compareToIgnoreCase("text") != 0) {
      System.out.println("Using SequenceFileInputFormat: " + args[2]);
    } else {
      System.out.println("Using TextInputFormat: " + args[2]);
      inputFormat = TextInputFormat.class;
    }
    int numOfReducers = Integer.parseInt(args[3]);
    Class mapper = getClassByName(args[4]);
    Class reducer = getClassByName(args[5]);
    Class mapoutputValueClass = getClassByName(args[6]);
    Class outputFormat = TextOutputFormat.class;
    Class outputValueClass = Text.class;
    if (args[7].compareToIgnoreCase("text") != 0) {
      System.out.println("Using SequenceFileOutputFormat: " + args[7]);
      outputFormat = SequenceFileOutputFormat.class;
      outputValueClass = getClassByName(args[7]);
    } else {
      System.out.println("Using TextOutputFormat: " + args[7]);
    }
    long maxNumOfValuesPerGroup = 100;
    String jobName = "";
    if (args.length > 8) {
      maxNumOfValuesPerGroup = Long.parseLong(args[8]);
    }
    if (args.length > 9) {
      jobName = args[9];
    }
    Configuration defaults = new Configuration();
    JobConf job = new JobConf(defaults, DataJoinJob.class);
    job.setJobName("DataJoinJob: " + jobName);

    FileSystem fs = FileSystem.get(defaults);
    fs.delete(new Path(outputDir));
    FileInputFormat.setInputPaths(job, inputDir);

    job.setInputFormat(inputFormat);

    job.setMapperClass(mapper);
    FileOutputFormat.setOutputPath(job, new Path(outputDir));
    job.setOutputFormat(outputFormat);
    SequenceFileOutputFormat.setOutputCompressionType(job,
            SequenceFile.CompressionType.BLOCK);
    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(mapoutputValueClass);
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(outputValueClass);
    job.setReducerClass(reducer);

    job.setNumMapTasks(1);
    job.setNumReduceTasks(numOfReducers);
    job.setLong("datajoin.maxNumOfValuesPerGroup", maxNumOfValuesPerGroup);
    return job;
  }
 

public static FileSystem setupJob(String indir,
                            String outdir, String cacheDir,
                            JobConf conf, String input,
                            boolean withSymlink)
throws IOException {
  final Path inDir = new Path(indir);
  final Path outDir = new Path(outdir);
  FileSystem fs = FileSystem.get(conf);
  fs.delete(outDir, true);
  if (!fs.mkdirs(inDir)) {
    throw new IOException("Mkdirs failed to create " + inDir.toString());
  }
  {
    DataOutputStream file = fs.create(new Path(inDir, "part-0"));
    file.writeBytes(input);
    file.close();
  }
  conf.setJobName("sharedcachetest");

  // the keys are words (strings)
  conf.setOutputKeyClass(Text.class);
  // the values are counts (ints)
  conf.setOutputValueClass(IntWritable.class);

  conf.setCombinerClass(MRSharedCaching.ReduceClass.class);
  conf.setReducerClass(MRSharedCaching.ReduceClass.class);
  FileInputFormat.setInputPaths(conf, inDir);
  FileOutputFormat.setOutputPath(conf, outDir);
  conf.setNumMapTasks(1);
  conf.setNumReduceTasks(1);
  conf.setSpeculativeExecution(false);

  if (!withSymlink) {
    conf.setMapperClass(MRSharedCaching.MapClass.class);
  } else {
    conf.setMapperClass(MRSharedCaching.MapClass2.class);
  }
  
  // Turn on sharing
  conf.set("mapred.cache.shared.enabled", "true");

  return fs;
}
 

/**
 * This is the main routine for launching a distributed random write job.
 * It runs 10 maps/node and each node writes 1 gig of data to a DFS file.
 * The reduce doesn't do anything.
 * 
 * @throws IOException 
 */
public int run(String[] args) throws Exception {    
  if (args.length == 0) {
    System.out.println("Usage: writer <out-dir>");
    ToolRunner.printGenericCommandUsage(System.out);
    return -1;
  }
  
  Path outDir = new Path(args[0]);
  JobConf job = new JobConf(getConf());
  
  job.setJarByClass(RandomWriter.class);
  job.setJobName("random-writer");
  FileOutputFormat.setOutputPath(job, outDir);
  
  job.setOutputKeyClass(BytesWritable.class);
  job.setOutputValueClass(BytesWritable.class);
  
  job.setInputFormat(RandomInputFormat.class);
  job.setMapperClass(Map.class);        
  job.setReducerClass(IdentityReducer.class);
  job.setOutputFormat(SequenceFileOutputFormat.class);
  
  JobClient client = new JobClient(job);
  ClusterStatus cluster = client.getClusterStatus();
  int numMapsPerHost = job.getInt("test.randomwriter.maps_per_host", 10);
  long numBytesToWritePerMap = job.getLong("test.randomwrite.bytes_per_map",
                                           1*1024*1024*1024);
  if (numBytesToWritePerMap == 0) {
    System.err.println("Cannot have test.randomwrite.bytes_per_map set to 0");
    return -2;
  }
  long totalBytesToWrite = job.getLong("test.randomwrite.total_bytes", 
       numMapsPerHost*numBytesToWritePerMap*cluster.getTaskTrackers());
  int numMaps = (int) (totalBytesToWrite / numBytesToWritePerMap);
  if (numMaps == 0 && totalBytesToWrite > 0) {
    numMaps = 1;
    job.setLong("test.randomwrite.bytes_per_map", totalBytesToWrite);
  }
  
  job.setNumMapTasks(numMaps);
  System.out.println("Running " + numMaps + " maps.");
  
  // reducer NONE
  job.setNumReduceTasks(0);
  
  Date startTime = new Date();
  System.out.println("Job started: " + startTime);
  JobClient.runJob(job);
  Date endTime = new Date();
  System.out.println("Job ended: " + endTime);
  System.out.println("The job took " + 
                     (endTime.getTime() - startTime.getTime()) /1000 + 
                     " seconds.");
  
  return 0;
}
 

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

    GfxdDataSerializable.initTypes();

    JobConf conf = new JobConf(getConf());
    conf.setJobName("TradeSellOrdersHdfsDataVerifier");

    String hdfsHomeDir = args[0];
    String url         = args[1];
    String tableName   = args[2];

    System.out.println("TradeSellOrdersHdfsDataVerifier.run() invoked with " 
                       + " hdfsHomeDir = " + hdfsHomeDir 
                       + " url = " + url
                       + " tableName = " + tableName);

    // Job-specific params
    conf.set(RowInputFormat.HOME_DIR, hdfsHomeDir);
    conf.set(RowInputFormat.INPUT_TABLE, tableName);
    conf.setBoolean(RowInputFormat.CHECKPOINT_MODE, false);
    
    conf.setInputFormat(RowInputFormat.class);
    conf.setMapperClass(HdfsDataMapper.class);
    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(TradeSellOrdersRow.class);
    
    conf.setReducerClass(HdfsDataReducer.class);
    conf.set(RowOutputFormat.OUTPUT_TABLE, tableName + "_HDFS");
    //conf.set(GfxdOutputFormat.OUTPUT_SCHEMA, "APP");
    conf.set(RowOutputFormat.OUTPUT_URL, url);
    conf.setOutputFormat(RowOutputFormat.class);
    conf.setOutputKeyClass(Key.class);
    conf.setOutputValueClass(TradeSellOrdersOutputObject.class);

    StringBuffer aStr = new StringBuffer();
    aStr.append("HOME_DIR = " + conf.get(RowInputFormat.HOME_DIR) + " ");
    aStr.append("INPUT_TABLE = " + conf.get(RowInputFormat.INPUT_TABLE) + " ");
    aStr.append("OUTPUT_TABLE = " + conf.get(RowOutputFormat.OUTPUT_TABLE) + " ");
    aStr.append("OUTPUT_URL = " + conf.get(RowOutputFormat.OUTPUT_URL) + " ");
    System.out.println("VerifyHdfsData running with the following conf: " + aStr.toString());

    
    FileOutputFormat.setOutputPath(conf, new Path("" + System.currentTimeMillis()));
    
    JobClient.runJob(conf);
    return 0;
  }
 

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

    GfxdDataSerializable.initTypes();

    JobConf conf = new JobConf(getConf());
    conf.setJobName("TradeBuyOrdersHdfsDataVerifier");

    String hdfsHomeDir = args[0];
    String url         = args[1];
    String tableName   = args[2];

    System.out.println("TradeBuyOrdersHdfsDataVerifier.run() invoked with " 
                       + " hdfsHomeDir = " + hdfsHomeDir 
                       + " url = " + url
                       + " tableName = " + tableName);

    // Job-specific params
    conf.set(RowInputFormat.HOME_DIR, hdfsHomeDir);
    conf.set(RowInputFormat.INPUT_TABLE, tableName);
    conf.setBoolean(RowInputFormat.CHECKPOINT_MODE, false);
    
    conf.setInputFormat(RowInputFormat.class);
    conf.setMapperClass(HdfsDataMapper.class);
    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(TradeBuyOrdersRow.class);
    
    conf.setReducerClass(HdfsDataReducer.class);
    conf.set(RowOutputFormat.OUTPUT_TABLE, tableName + "_HDFS");
    //conf.set(GfxdOutputFormat.OUTPUT_SCHEMA, "APP");
    conf.set(RowOutputFormat.OUTPUT_URL, url);
    conf.setOutputFormat(RowOutputFormat.class);
    conf.setOutputKeyClass(Key.class);
    conf.setOutputValueClass(TradeBuyOrdersOutputObject.class);

    StringBuffer aStr = new StringBuffer();
    aStr.append("HOME_DIR = " + conf.get(RowInputFormat.HOME_DIR) + " ");
    aStr.append("INPUT_TABLE = " + conf.get(RowInputFormat.INPUT_TABLE) + " ");
    aStr.append("OUTPUT_TABLE = " + conf.get(RowOutputFormat.OUTPUT_TABLE) + " ");
    aStr.append("OUTPUT_URL = " + conf.get(RowOutputFormat.OUTPUT_URL) + " ");
    System.out.println("VerifyHdfsData running with the following conf: " + aStr.toString());

    
    FileOutputFormat.setOutputPath(conf, new Path("" + System.currentTimeMillis()));
    
    JobClient.runJob(conf);
    return 0;
  }
 

public void runParseTest(String fieldTerminator, String lineTerminator,
    String encloser, String escape, boolean encloseRequired)
    throws IOException {

  ClassLoader prevClassLoader = null;

  String [] argv = getArgv(true, fieldTerminator, lineTerminator,
      encloser, escape, encloseRequired);
  runImport(argv);
  try {
    String tableClassName = getTableName();

    argv = getArgv(false, fieldTerminator, lineTerminator, encloser, escape,
        encloseRequired);
    SqoopOptions opts = new ImportTool().parseArguments(argv, null, null,
        true);

    CompilationManager compileMgr = new CompilationManager(opts);
    String jarFileName = compileMgr.getJarFilename();

    // Make sure the user's class is loaded into our address space.
    prevClassLoader = ClassLoaderStack.addJarFile(jarFileName,
        tableClassName);

    JobConf job = new JobConf();
    job.setJar(jarFileName);

    // Tell the job what class we're testing.
    job.set(ReparseMapper.USER_TYPE_NAME_KEY, tableClassName);

    // use local mode in the same JVM.
    ConfigurationHelper.setJobtrackerAddr(job, "local");
    if (!BaseSqoopTestCase.isOnPhysicalCluster()) {
      job.set(CommonArgs.FS_DEFAULT_NAME, CommonArgs.LOCAL_FS);
    }
    String warehouseDir = getWarehouseDir();
    Path warehousePath = new Path(warehouseDir);
    Path inputPath = new Path(warehousePath, getTableName());
    Path outputPath = new Path(warehousePath, getTableName() + "-out");

    job.setMapperClass(ReparseMapper.class);
    job.setNumReduceTasks(0);
    FileInputFormat.addInputPath(job, inputPath);
    FileOutputFormat.setOutputPath(job, outputPath);

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

    JobClient.runJob(job);
  } catch (InvalidOptionsException ioe) {
    fail(ioe.toString());
  } catch (ParseException pe) {
    fail(pe.toString());
  } finally {
    if (null != prevClassLoader) {
      ClassLoaderStack.setCurrentClassLoader(prevClassLoader);
    }
  }
}
 

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

    // todo@lhughes -- why do we need this?
    GfxdDataSerializable.initTypes();

    JobConf conf = new JobConf(getConf());
    conf.setJobName("hdfsMapReduce");

    String hdfsHomeDir = args[0];
    String url         = args[1];
    String tableName   = args[2];

    System.out.println("VerifyHdfsData.run() invoked with " 
                       + " hdfsHomeDir = " + hdfsHomeDir 
                       + " url = " + url
                       + " tableName = " + tableName);

    // Job-specific params
    conf.set(RowInputFormat.HOME_DIR, hdfsHomeDir);
    conf.set(RowInputFormat.INPUT_TABLE, tableName);
    conf.setBoolean(RowInputFormat.CHECKPOINT_MODE, false);
    
    conf.setInputFormat(RowInputFormat.class);
    conf.setMapperClass(HdfsDataMapper.class);
    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(MyRow.class);
    
    conf.setReducerClass(HdfsDataReducer.class);
    conf.set(RowOutputFormat.OUTPUT_TABLE, "TRADE.HDFS_CUSTOMERS");
    //conf.set(GfxdOutputFormat.OUTPUT_SCHEMA, "APP");
    conf.set(RowOutputFormat.OUTPUT_URL, url);
    conf.setOutputFormat(RowOutputFormat.class);
    conf.setOutputKeyClass(Key.class);
    conf.setOutputValueClass(DataObject.class);

    StringBuffer aStr = new StringBuffer();
    aStr.append("HOME_DIR = " + conf.get(RowInputFormat.HOME_DIR) + " ");
    aStr.append("INPUT_TABLE = " + conf.get(RowInputFormat.INPUT_TABLE) + " ");
    aStr.append("OUTPUT_TABLE = " + conf.get(RowOutputFormat.OUTPUT_TABLE) + " ");
    aStr.append("OUTPUT_URL = " + conf.get(RowOutputFormat.OUTPUT_URL) + " ");
    System.out.println("VerifyHdfsData running with the following conf: " + aStr.toString());

    // not planning to use this, but I get an NPE without it
    FileOutputFormat.setOutputPath(conf, new Path("" + System.currentTimeMillis()));
    
    JobClient.runJob(conf);
    return 0;
  }
 

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

    GfxdDataSerializable.initTypes();

    JobConf conf = new JobConf(getConf());
    conf.setJobName("TradesHdfsDataVerifier");

    String hdfsHomeDir = args[0];
    String url         = args[1];
    String tableName   = args[2];

    System.out.println("TradesHdfsDataVerifier.run() invoked with " 
                       + " hdfsHomeDir = " + hdfsHomeDir 
                       + " url = " + url
                       + " tableName = " + tableName);

    // Job-specific params
    conf.set(RowInputFormat.HOME_DIR, hdfsHomeDir);
    conf.set(RowInputFormat.INPUT_TABLE, tableName);
    conf.setBoolean(RowInputFormat.CHECKPOINT_MODE, false);
    
    conf.setInputFormat(RowInputFormat.class);
    conf.setMapperClass(HdfsDataMapper.class);
    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(TradesRow.class);
    
    conf.setReducerClass(HdfsDataReducer.class);
    conf.set(RowOutputFormat.OUTPUT_TABLE, tableName + "_HDFS");
    //conf.set(GfxdOutputFormat.OUTPUT_SCHEMA, "APP");
    conf.set(RowOutputFormat.OUTPUT_URL, url);
    conf.setOutputFormat(RowOutputFormat.class);
    conf.setOutputKeyClass(Key.class);
    conf.setOutputValueClass(TradeOutputObject.class);

    StringBuffer aStr = new StringBuffer();
    aStr.append("HOME_DIR = " + conf.get(RowInputFormat.HOME_DIR) + " ");
    aStr.append("INPUT_TABLE = " + conf.get(RowInputFormat.INPUT_TABLE) + " ");
    aStr.append("OUTPUT_TABLE = " + conf.get(RowOutputFormat.OUTPUT_TABLE) + " ");
    aStr.append("OUTPUT_URL = " + conf.get(RowOutputFormat.OUTPUT_URL) + " ");
    System.out.println("VerifyHdfsData running with the following conf: " + aStr.toString());

    
    FileOutputFormat.setOutputPath(conf, new Path("" + System.currentTimeMillis()));
    
    JobClient.runJob(conf);
    return 0;
  }