Package org.apache.nutch.io

Examples of org.apache.nutch.io.LongWritable

241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
    // XXX One should check for this later, when actually
    // XXX reading the entries.
    FetcherOutput fo = new FetcherOutput();
    fetcherReader.next(fo);
    started = fo.getFetchDate();
    LongWritable w = new LongWritable(-1);
    try {
      fetcherReader.finalKey(w);
    } catch (Throwable eof) {
      // the file is truncated - probably due to a crashed fetcher.
      // Use just the part that we can...
      LOG.warning(" - data in segment " + dir + " is corrupt, using only " + w.get() + " entries.");
    }
    // go back until you get a good entry
    size = w.get()+1;
    boolean ok = false;
    int back = 0;
    do {
      try {
        fetcherReader.seek(size - 2 - back);
View Full Code Here
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
      nfs.delete(unsortedFile);
      nfs.delete(sortedFile);
      SequenceFile.Writer seqWriter = new SequenceFile.Writer(nfs,
              unsortedFile.toString(), UTF8.class, LongWritable.class);
      FetchListEntry fle;
      LongWritable rec = new LongWritable();
      UTF8 url = new UTF8();
      String urlString;
      while (fetcherReader.next(fo) != null) {
        fle = fo.getFetchListEntry();
        urlString = fle.getPage().getURL().toString();
        rec.set(recNo);
        url.set(urlString);
        seqWriter.append(url, rec);
        recNo++;
      }
      seqWriter.close();
      // sort the SequenceFile
      long start = System.currentTimeMillis();

      SequenceFile.Sorter sorter = new SequenceFile.Sorter(nfs,
              new UTF8.Comparator(), LongWritable.class);

      sorter.sort(unsortedFile.toString(), sortedFile.toString());

      float localSecs = (System.currentTimeMillis() - start) / 1000.0f;
      LOG.info(" - sorted: " + recNo + " entries in " + localSecs + "s, "
        + (recNo/localSecs) + " entries/s");

      nfs.delete(unsortedFile);
      SequenceFile.Reader seqReader = new SequenceFile.Reader(nfs, sortedFile.toString());
      while (seqReader.next(url, rec)) {
        recNo = rec.get();
        get(recNo, fo, co, pt, pd);
        output.println("Recno:: " + recNo++);
        output.println("FetcherOutput::\n" + fo.toString());
        if (contentReader != null)
          output.println("Content::\n" + co.toString());
View Full Code Here
79
80
81
82
83
84
85
86
87
88
89
90
    }

    public void run() {
      for (int i = 0; i < count; i++) {
        try {
          LongWritable param = new LongWritable(RANDOM.nextLong());
          LongWritable value =
            (LongWritable)client.call(param, new InetSocketAddress(PORT));
          if (!param.equals(value)) {
            LOG.severe("Call failed!");
            failed = true;
            break;
View Full Code Here
114
115
116
117
118
119
120
121
122
123
124
    public void run() {
      for (int i = 0; i < count; i++) {
        try {
          Writable[] params = new Writable[addresses.length];
          for (int j = 0; j < addresses.length; j++)
            params[j] = new LongWritable(RANDOM.nextLong());
          Writable[] values = client.call(params, addresses);
          for (int j = 0; j < addresses.length; j++) {
            if (!params[j].equals(values[j])) {
              LOG.severe("Call failed!");
              failed = true;
View Full Code Here
42
43
44
45
46
47
48
49
    // tokenize the value
    StringTokenizer st = new StringTokenizer(text);
    while (st.hasMoreTokens()) {
      // output <token,1> pairs
      output.collect(new UTF8(st.nextToken()), new LongWritable(1));
    }
  }
View Full Code Here
45
46
47
48
49
50
51
52
  public void map(WritableComparable key, Writable value,
                  OutputCollector output) throws IOException {
    String text = ((UTF8)value).toString();
    Matcher matcher = pattern.matcher(text);
    while (matcher.find()) {
      output.collect(new UTF8(matcher.group(group)), new LongWritable(1));
    }
  }
View Full Code Here
41
42
43
44
45
46
47
    while (values.hasNext()) {
      sum += ((LongWritable)values.next()).get();
    }

    // output sum
    output.collect(key, new LongWritable(sum));
  }
View Full Code Here
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
    // XXX One should check for this later, when actually
    // XXX reading the entries.
    FetcherOutput fo = new FetcherOutput();
    fetcherReader.next(fo);
    started = fo.getFetchDate();
    LongWritable w = new LongWritable(-1);
    try {
      fetcherReader.finalKey(w);
    } catch (Throwable eof) {
      // the file is truncated - probably due to a crashed fetcher.
      // Use just the part that we can...
      LOG.warning(" - data in segment " + dir + " is corrupt, using only " + w.get() + " entries.");
    }
    // go back until you get a good entry
    size = w.get()+1;
    boolean ok = false;
    int back = 0;
    do {
      try {
        fetcherReader.seek(size - 2 - back);
View Full Code Here
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
      nfs.delete(unsortedFile);
      nfs.delete(sortedFile);
      SequenceFile.Writer seqWriter = new SequenceFile.Writer(nfs,
              unsortedFile.toString(), UTF8.class, LongWritable.class);
      FetchListEntry fle;
      LongWritable rec = new LongWritable();
      UTF8 url = new UTF8();
      String urlString;
      while (fetcherReader.next(fo) != null) {
        fle = fo.getFetchListEntry();
        urlString = fle.getPage().getURL().toString();
        rec.set(recNo);
        url.set(urlString);
        seqWriter.append(url, rec);
        recNo++;
      }
      seqWriter.close();
      // sort the SequenceFile
      long start = System.currentTimeMillis();

      SequenceFile.Sorter sorter = new SequenceFile.Sorter(nfs,
              new UTF8.Comparator(), LongWritable.class);

      sorter.sort(unsortedFile.toString(), sortedFile.toString());

      float localSecs = (System.currentTimeMillis() - start) / 1000.0f;
      LOG.info(" - sorted: " + recNo + " entries in " + localSecs + "s, "
        + (recNo/localSecs) + " entries/s");

      nfs.delete(unsortedFile);
      SequenceFile.Reader seqReader = new SequenceFile.Reader(nfs, sortedFile.toString());
      while (seqReader.next(url, rec)) {
        recNo = rec.get();
        get(recNo, fo, co, pt, pd);
        output.println("Recno:: " + recNo++);
        output.println("FetcherOutput::\n" + fo.toString());
        if (contentReader != null)
          output.println("Content::\n" + co.toString());
View Full Code Here
TOP

Related Classes of org.apache.nutch.io.LongWritable

  • org.apache.nutch.ipc.TestIPC$ParallelCaller
  • org.apache.nutch.ipc.TestIPC$SerialCaller
  • org.apache.nutch.mapReduce.lib.LongSumReducer
  • org.apache.nutch.mapReduce.lib.RegexMapper
  • org.apache.nutch.mapReduce.lib.TokenCountMapper
  • org.apache.nutch.segment.SegmentReader
    • 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.