Package edu.umd.cloud9.collection

Examples of edu.umd.cloud9.collection.Indexable

658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
  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
359
360
361
362
363
364
365
366
367
368
369
    if (list.getLength() > 0) {  conf.set(Constants.StemmedStopwordListQ, list.item(0).getTextContent());  } 
  }

  static float eval(QueryEngine qe, Configuration conf, String setting){
    Qrels qrels = new Qrels(conf.get(Constants.QrelsPath));
    DocnoMapping mapping = qe.getDocnoMapping();
    float apSum = 0, p10Sum = 0;
    Map<String, Accumulator[]> results = qe.getResults();
    for (String qid : results.keySet()) {
      float ap = (float) RankedListEvaluator.computeAP(results.get(qid), mapping,
          qrels.getReldocsForQid(qid));
View Full Code Here
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
      int docno = 0;
      try {
        if (req.getParameterValues(DOCNO) != null)
          docno = Integer.parseInt(req.getParameterValues(DOCNO)[0]);

        Indexable doc = mForwardIndex.getDocument(docno);

        if (doc != null) {
          sLogger.info("fetched: " + doc.getDocid() + " = docno " + docno);
          res.setContentType(doc.getDisplayContentType());

          PrintWriter out = res.getWriter();
          out.print(doc.getDisplayContent());
          out.close();
        } else {
          throw new Exception();
        }
      } catch (Exception e) {
View Full Code Here
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
        if (req.getParameterValues("docno") != null)
          docno = Integer.parseInt(req.getParameterValues("docno")[0]);
        else if (req.getParameterValues("docid") != null)
          docno = sForwardIndex.getDocno(req.getParameterValues("docid")[0]);

        Indexable doc = null;
        int i = 0;
        for(i = 0; i < lastDocs.length; i++)
          if(docno <= lastDocs[i]) {
            doc = docForwardIndex[i].getDocument(docno);
            break;
          }

        if (doc != null) {
          LOG.info("fetched: " + doc.getDocid() + " = docno " + docno);
          res.setContentType(doc.getDisplayContentType());

          PrintWriter out = res.getWriter();
          out.print(doc.getContent().replaceAll("<\\s*/\\s*[bB][oO][dD][Yy]\\s*>", "<br><br><a href=\"/fetch_docno?docno=" + docno +
                "\"> Fetch anchor text for docno: " + docno + "</a></body>"));
          out.close();
        } else {
          throw new Exception();
        }
View Full Code Here
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
      try {
       
        if (req.getParameterValues("docid") != null)
          docid = req.getParameterValues("docid")[0];

        Indexable doc = sForwardIndex.getDocument(docid);
       
        if (doc != null) {
          LOG.info("fetched: " + doc.getDocid());
          res.setContentType(doc.getDisplayContentType());

          PrintWriter out = res.getWriter();
          out.print(doc.getContent().replace("<body>", "<body><a href=\"/fetch_content?docid=" + docid +
              "\"> Fetch content for docid: " + docid + "</a><br><br>"));
          out.close();
        } else {
          throw new Exception();
        }
View Full Code Here
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
      try {
       
        if (req.getParameterValues("docno") != null)
          docno = Integer.parseInt(req.getParameterValues("docno")[0]);

        Indexable doc = sForwardIndex.getDocument(docno);
 
        if (doc != null) {
          LOG.info("fetched: " + doc.getDocid() + " = docno " + docno);
          res.setContentType(doc.getDisplayContentType());
         
          PrintWriter out = res.getWriter();
          out.print(doc.getContent().replace("<body>", "<body><a href=\"/fetch_content?docno=" + docno +
              "\"> Fetch content for docno: " + docno + "</a><br><br>"));
          out.close();
        } else {
          throw new Exception();
        }
View Full Code Here
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
      int docno = 0;
      try {
        if (req.getParameterValues(DOCNO) != null)
          docno = Integer.parseInt(req.getParameterValues(DOCNO)[0]);

        Indexable doc = mForwardIndex.getDocument(docno);

        if (doc != null) {
          sLogger.info("fetched: " + doc.getDocid() + " = docno " + docno);
          res.setContentType(doc.getDisplayContentType());

          PrintWriter out = res.getWriter();
          out.print(doc.getDisplayContent());
          out.close();
        } else {
          throw new Exception();
        }
      } catch (Exception e) {
View Full Code Here
115
116
117
118
119
120
121
122
123
124
125
      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
82
83
84
85
86
87
88
89
90
91
92
93
94
    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
79
80
81
82
83
84
85
86
87
88
89
90
91
    // (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
TOP

Related Classes of edu.umd.cloud9.collection.Indexable

  • bak.pcj.IntIterator
  • edu.umd.cloud9.example.bfs.BfsNodeTest
  • edu.umd.cloud9.example.pagerank.PageRankNodeTest
  • edu.umd.cloud9.io.array.ArrayListOfIntsWritable
  • edu.umd.cloud9.io.fastutil.Int2FloatOpenHashMapWritableTest
  • edu.umd.cloud9.io.fastutil.Int2IntOpenHashMapWritableTest
  • edu.umd.cloud9.io.fastutil.String2FloatOpenHashMapWritableTest
  • edu.umd.cloud9.io.fastutil.String2IntOpenHashMapWritableTest
  • edu.umd.cloud9.io.map.HMapIDW
  • edu.umd.cloud9.io.map.HMapIDWTest
    • 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.