Method run

src/java/org/apache/nutch/indexer/CleaningJob.java:136–152

(Map<String, Object> args)

  }

  @Override
  public Map<String, Object> run(Map<String, Object> args) throws Exception {
    getConf().setBoolean(ARG_COMMIT, (Boolean) args.get(ARG_COMMIT));
    currentJob = NutchJob.getInstance(getConf(), "CleaningJob");
    currentJob.getConfiguration().setClass(
        "mapreduce.job.output.key.comparator.class", StringComparator.class,
        RawComparator.class);

    Collection<WebPage.Field> fields = getFields(currentJob);
    StorageUtils.initMapperJob(currentJob, fields, String.class, WebPage.class,
        CleanMapper.class);
    currentJob.setReducerClass(CleanReducer.class);
    currentJob.setOutputFormatClass(NullOutputFormat.class);
    currentJob.waitForCompletion(true);
    ToolUtil.recordJobStatus(null, currentJob, results);
    return results;
  }

  public int delete(boolean commit) throws Exception {
    LOG.info("CleaningJob: starting");