Package edu.umd.cloud9.util.map

Examples of edu.umd.cloud9.util.map.HMapID$Entry


  public DocnoMapping getDocnoMapping() throws IOException {
    return loadDocnoMapping(indexPath, fs);
  }

  public static DocnoMapping loadDocnoMapping(String indexPath, FileSystem fs) throws IOException {
    DocnoMapping mDocMapping = null;
    // load the docid to docno mappings
    try {
      LOG.info("Loading DocnoMapping file...");
      RetrievalEnvironment env = new RetrievalEnvironment(indexPath, fs);

      String className = env.readDocnoMappingClass();
      LOG.info(" - Class name: " + className);
      mDocMapping = (DocnoMapping) Class.forName(className).newInstance();

      Path mappingFile = env.getDocnoMappingData();
      LOG.info(" - File name: " + mappingFile);
      mDocMapping.loadMapping(mappingFile, fs);
      LOG.info("Done!");
    } catch (Exception e) {
      throw new IOException("Error initializing DocnoMapping!");
    }
    return mDocMapping;
View Full Code Here


      return -1;
    }

    RetrievalEnvironment env = new RetrievalEnvironment(indexPath, fs);
    Path mappingFile = env.getDocnoMappingData();
    new ClueWarcDocnoMappingBuilder().build(new Path(collection), mappingFile, conf);

    conf.set(Constants.CollectionName, "ClueWeb:English:Segment" + segment);
    conf.set(Constants.CollectionPath, collection);
    conf.set(Constants.IndexPath, indexPath);
    conf.set(Constants.InputFormat, SequenceFileInputFormat.class.getCanonicalName());
View Full Code Here

    if (!fs.exists(p)) {
      sLogger.info("docno-mapping.dat doesn't exist, creating...");
      String[] arr = new String[] { collection, mappingDir.toString(), mappingFile.toString(),
          new Integer(numMappers).toString() };
      NumberTextDocuments tool = new NumberTextDocuments();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(mappingDir, true);
    }

    // Now we're ready to start the preprocessing pipeline... set
View Full Code Here

    // (sequentially-number integer). If it doesn't exist create it.
    Path mappingFile = env.getDocnoMappingData();
    if (!fs.exists(mappingFile)) {
      sLogger.info(mappingFile + " doesn't exist, creating...");
      String[] arr = new String[] { collection, indexPath + "/medline-docid-tmp",  mappingFile.toString(), new Integer(numMappers).toString() };
      NumberMedlineCitations tool = new NumberMedlineCitations();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(new Path(indexPath + "/medline-docid-tmp"), true);
    }

    // Now we're ready to start the preprocessing pipeline... set
View Full Code Here

    if (!fs.exists(mappingFile)) {
      LOG.info("docno-mapping.dat doesn't exist, creating...");
      String[] arr = new String[] { collection, mappingDir.toString(),
              mappingFile.toString() };
      NumberTrecDocuments2 tool = new NumberTrecDocuments2();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(mappingDir, true);
    }

    conf.set(Constants.CollectionName, "TREC_vol45");
View Full Code Here

    if (!fs.exists(mappingFile)) {
      LOG.info("docno-mapping.dat doesn't exist, creating...");
      String[] arr = new String[] { collection, mappingDir.toString(),
          mappingFile.toString(), "100" };
      NumberTrecWebDocuments tool = new NumberTrecWebDocuments();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(mappingDir, true);
    }

    conf.set(Constants.CollectionName, "Wt10g");
View Full Code Here

    if (!fs.exists(mappingFile)) {
      LOG.info("docno-mapping.dat doesn't exist, creating...");
      String[] arr = new String[] { collection, mappingDir.toString(),
          mappingFile.toString(), "100" };
      NumberTrecWebDocuments tool = new NumberTrecWebDocuments();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(mappingDir, true);
    }

    conf.set(Constants.CollectionName, "Gov2");
View Full Code Here

          "-output_path=" + indexRootPath + "/wiki-docid-tmp",
          "-output_file=" + mappingFile.toString(),
          "-lang=" + collectionLang };
      LOG.info("Running BuildWikipediaDocnoMapping with args " + Arrays.toString(arr));

      BuildWikipediaDocnoMapping tool = new BuildWikipediaDocnoMapping();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(new Path(indexRootPath + "/wiki-docid-tmp"), true);
    }

    // Repack Wikipedia into sequential compressed block
    p = new Path(seqCollection);
    LOG.info(seqCollection + " doesn't exist, creating...");
    String[] arr = new String[] { "-input=" + rawCollection,
        "-output=" + seqCollection,
        "-mapping_file=" + mappingFile.toString(),
        "-compression_type=block",
        "-wiki_language=" + collectionLang };
    LOG.info("Running RepackWikipedia with args " + Arrays.toString(arr));

    RepackWikipedia tool = new RepackWikipedia();
    tool.setConf(conf);
    tool.run(arr);

    conf.set(Constants.CollectionName, "Wikipedia-"+collectionLang);
    conf.setInt(Constants.NumMapTasks, numMappers);
    conf.setInt(Constants.NumReduceTasks, numReducers);
    conf.set(Constants.CollectionPath, seqCollection);
View Full Code Here

    Path mappingFile = env.getDocnoMappingData();
    if (!fs.exists(mappingFile)) {
      LOG.info(mappingFile + " doesn't exist, creating...");
      String[] arr = new String[] { rawCollection, indexRootPath + "/wiki-docid-tmp",
          mappingFile.toString(), new Integer(numMappers).toString() };
      BuildWikipediaDocnoMapping tool = new BuildWikipediaDocnoMapping();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(new Path(indexRootPath + "/wiki-docid-tmp"), true);
    }else{
      LOG.info(p+" exists");
    }

    // Repack Wikipedia into sequential compressed block
    p = new Path(seqCollection);
    if (!fs.exists(p)) {
      LOG.info(seqCollection + " doesn't exist, creating...");
      String[] arr = new String[] { rawCollection, seqCollection, mappingFile.toString(), "block"};
      RepackWikipedia tool = new RepackWikipedia();
      tool.setConf(conf);
      tool.run(arr);
    }

    conf.set("Ivory.CollectionName", "Wikipedia-"+collectionLang);
    conf.setInt("Ivory.NumMapTasks", numMappers);
    conf.setInt("Ivory.NumReduceTasks", numReducers);
View Full Code Here

          "-output_path=" + indexRootPath + "/wiki-docid-tmp",
          "-output_file=" + mappingFile.toString(),
          "-wiki_language=" + collectionLang };
      LOG.info("Running BuildWikipediaDocnoMapping with args " + Arrays.toString(arr));

      BuildWikipediaDocnoMapping tool = new BuildWikipediaDocnoMapping();
      tool.setConf(conf);
      tool.run(arr);

      fs.delete(new Path(indexRootPath + "/wiki-docid-tmp"), true);
    } else {
      LOG.info("Docno mapping already exists at: " + mappingFile);
    }

    // Repack Wikipedia into sequential compressed block
    if (!fs.exists(new Path(seqCollection + "/part-00000"))) {
      LOG.info(seqCollection + " doesn't exist, creating...");
      String[] arr = new String[] { "-input=" + rawCollection,
          "-output=" + seqCollection,
          "-mapping_file=" + mappingFile.toString(),
          "-compression_type=block",
          "-wiki_language=" + collectionLang };
      LOG.info("Running RepackWikipedia with args " + Arrays.toString(arr));

      RepackWikipedia tool = new RepackWikipedia();
      tool.setConf(conf);
      tool.run(arr);
    } else {
      LOG.info("Repacked collection already exists at: " + seqCollection);     
    }

    conf.set(Constants.CollectionName, "Wikipedia-"+collectionLang);
View Full Code Here

TOP

Related Classes of edu.umd.cloud9.util.map.HMapID$Entry

Copyright © 2018 www.massapicom. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.