Package org.apache.lucene.codecs

Examples of org.apache.lucene.codecs.BlockTermState

325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
      // 3rd pass: write the freqs as byte[] blob
      // TODO: cutover to better intblock codec.  simple64?
      // write prefix, suffix first:
      for(int termCount=0;termCount<pendingCount;termCount++) {
        final BlockTermState state = pendingTerms[termCount].state;
        assert state != null;
        bytesWriter.writeVInt(state.docFreq);
        if (fieldInfo.getIndexOptions() != IndexOptions.DOCS_ONLY) {
          bytesWriter.writeVLong(state.totalTermFreq-state.docFreq);
        }
      }
      out.writeVInt((int) bytesWriter.getFilePointer());
      bytesWriter.writeTo(out);
      bytesWriter.reset();

      // 4th pass: write the metadata
      long[] longs = new long[longsSize];
      boolean absolute = true;
      for(int termCount=0;termCount<pendingCount;termCount++) {
        final BlockTermState state = pendingTerms[termCount].state;
        postingsWriter.encodeTerm(longs, bufferWriter, fieldInfo, state, absolute);
        for (int i = 0; i < longsSize; i++) {
          bytesWriter.writeVLong(longs[i]);
        }
        bufferWriter.writeTo(bytesWriter);
View Full Code Here
240
241
242
243
244
245
246
247
248
249
250
    }

    @Override
    public void finishTerm(BytesRef text, TermStats stats) throws IOException {
      // write term meta data into fst
      final BlockTermState state = postingsWriter.newTermState();
      final FSTTermOutputs.TermData meta = new FSTTermOutputs.TermData();
      meta.longs = new long[longsSize];
      meta.bytes = null;
      meta.docFreq = state.docFreq = stats.docFreq;
      meta.totalTermFreq = state.totalTermFreq = stats.totalTermFreq;
View Full Code Here
313
314
315
316
317
318
319
320
321
322
323
          statsOut.writeVLong(stats.totalTermFreq-stats.docFreq);
        }
      } else {
        statsOut.writeVInt(stats.docFreq);
      }
      BlockTermState state = postingsWriter.newTermState();
      state.docFreq = stats.docFreq;
      state.totalTermFreq = stats.totalTermFreq;
      postingsWriter.finishTerm(state);
      postingsWriter.encodeTerm(longs, metaBytesOut, fieldInfo, state, true);
      for (int i = 0; i < longsSize; i++) {
View Full Code Here
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
      if (isLeafBlock) {
        subIndices = null;
        for (PendingEntry ent : slice) {
          assert ent.isTerm;
          PendingTerm term = (PendingTerm) ent;
          BlockTermState state = term.state;
          final int suffix = term.term.length - prefixLength;
          // if (DEBUG) {
          //   BytesRef suffixBytes = new BytesRef(suffix);
          //   System.arraycopy(term.term.bytes, prefixLength, suffixBytes.bytes, 0, suffix);
          //   suffixBytes.length = suffix;
          //   System.out.println("    write term suffix=" + suffixBytes);
          // }
          // For leaf block we write suffix straight
          suffixWriter.writeVInt(suffix);
          suffixWriter.writeBytes(term.term.bytes, prefixLength, suffix);

          // Write term stats, to separate byte[] blob:
          statsWriter.writeVInt(state.docFreq);
          if (fieldInfo.getIndexOptions() != IndexOptions.DOCS_ONLY) {
            assert state.totalTermFreq >= state.docFreq: state.totalTermFreq + " vs " + state.docFreq;
            statsWriter.writeVLong(state.totalTermFreq - state.docFreq);
          }

          // Write term meta data
          postingsWriter.encodeTerm(longs, bytesWriter, fieldInfo, state, absolute);
          for (int pos = 0; pos < longsSize; pos++) {
            assert longs[pos] >= 0;
            metaWriter.writeVLong(longs[pos]);
          }
          bytesWriter.writeTo(metaWriter);
          bytesWriter.reset();
          absolute = false;
        }
        termCount = length;
      } else {
        subIndices = new ArrayList<>();
        termCount = 0;
        for (PendingEntry ent : slice) {
          if (ent.isTerm) {
            PendingTerm term = (PendingTerm) ent;
            BlockTermState state = term.state;
            final int suffix = term.term.length - prefixLength;
            // if (DEBUG) {
            //   BytesRef suffixBytes = new BytesRef(suffix);
            //   System.arraycopy(term.term.bytes, prefixLength, suffixBytes.bytes, 0, suffix);
            //   suffixBytes.length = suffix;
View Full Code Here
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
      assert stats.docFreq > 0;
      //if (DEBUG) System.out.println("BTTW.finishTerm term=" + fieldInfo.name + ":" + toString(text) + " seg=" + segment + " df=" + stats.docFreq);

      blockBuilder.add(Util.toIntsRef(text, scratchIntsRef), noOutputs.getNoOutput());
      BlockTermState state = postingsWriter.newTermState();
      state.docFreq = stats.docFreq;
      state.totalTermFreq = stats.totalTermFreq;
      postingsWriter.finishTerm(state);

      PendingTerm term = new PendingTerm(BytesRef.deepCopyOf(text), state);
View Full Code Here
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
      if (isLeafBlock) {
        subIndices = null;
        for (PendingEntry ent : slice) {
          assert ent.isTerm;
          PendingTerm term = (PendingTerm) ent;
          BlockTermState state = term.state;
          maxVersionInBlock = Math.max(maxVersionInBlock, ((IDVersionTermState) state).idVersion);
          final int suffix = term.term.length - prefixLength;
          // if (DEBUG) {
          //    BytesRef suffixBytes = new BytesRef(suffix);
          //    System.arraycopy(term.term.bytes, prefixLength, suffixBytes.bytes, 0, suffix);
          //    suffixBytes.length = suffix;
          //    System.out.println("    " + (countx++) + ": write term suffix=" + toString(suffixBytes));
          // }
          // For leaf block we write suffix straight
          suffixWriter.writeVInt(suffix);
          suffixWriter.writeBytes(term.term.bytes, prefixLength, suffix);

          // Write term meta data
          postingsWriter.encodeTerm(longs, bytesWriter, fieldInfo, state, absolute);
          for (int pos = 0; pos < longsSize; pos++) {
            assert longs[pos] >= 0;
            metaWriter.writeVLong(longs[pos]);
          }
          bytesWriter.writeTo(metaWriter);
          bytesWriter.reset();
          absolute = false;
        }
        termCount = length;
      } else {
        subIndices = new ArrayList<>();
        termCount = 0;
        for (PendingEntry ent : slice) {
          if (ent.isTerm) {
            PendingTerm term = (PendingTerm) ent;
            BlockTermState state = term.state;
            maxVersionInBlock = Math.max(maxVersionInBlock, ((IDVersionTermState) state).idVersion);
            final int suffix = term.term.length - prefixLength;
            // if (DEBUG) {
            //    BytesRef suffixBytes = new BytesRef(suffix);
            //    System.arraycopy(term.term.bytes, prefixLength, suffixBytes.bytes, 0, suffix);
View Full Code Here
938
939
940
941
942
943
944
945
946
947
948
      assert stats.docFreq > 0;
      //if (DEBUG) System.out.println("BTTW.finishTerm term=" + fieldInfo.name + ":" + toString(text) + " seg=" + segment + " df=" + stats.docFreq);
      if (((IDVersionPostingsWriter) postingsWriter).lastDocID != -1) {
        blockBuilder.add(Util.toIntsRef(text, scratchIntsRef), noOutputs.getNoOutput());
        BlockTermState state = postingsWriter.newTermState();
        state.docFreq = stats.docFreq;
        state.totalTermFreq = stats.totalTermFreq;
        postingsWriter.finishTerm(state);

        PendingTerm term = new PendingTerm(BytesRef.deepCopyOf(text), state);
View Full Code Here
323
324
325
326
327
328
329
330
331
332
333
          statsOut.writeVLong(stats.totalTermFreq-stats.docFreq);
        }
      } else {
        statsOut.writeVInt(stats.docFreq);
      }
      BlockTermState state = postingsWriter.newTermState();
      state.docFreq = stats.docFreq;
      state.totalTermFreq = stats.totalTermFreq;
      postingsWriter.finishTerm(state);
      postingsWriter.encodeTerm(longs, metaBytesOut, fieldInfo, state, true);
      for (int i = 0; i < longsSize; i++) {
View Full Code Here
248
249
250
251
252
253
254
255
256
257
258
    }

    @Override
    public void finishTerm(BytesRef text, TermStats stats) throws IOException {
      // write term meta data into fst
      final BlockTermState state = postingsWriter.newTermState();
      final FSTTermOutputs.TermData meta = new FSTTermOutputs.TermData();
      meta.longs = new long[longsSize];
      meta.bytes = null;
      meta.docFreq = state.docFreq = stats.docFreq;
      meta.totalTermFreq = state.totalTermFreq = stats.totalTermFreq;
View Full Code Here
245
246
247
248
249
250
251
252
253
254
255
    }

    @Override
    public void finishTerm(BytesRef text, TermStats stats) throws IOException {
      // write term meta data into fst
      final BlockTermState state = postingsWriter.newTermState();
      final FSTTermOutputs.TermData meta = new FSTTermOutputs.TermData();
      meta.longs = new long[longsSize];
      meta.bytes = null;
      meta.docFreq = state.docFreq = stats.docFreq;
      meta.totalTermFreq = state.totalTermFreq = stats.totalTermFreq;
View Full Code Here
TOP

Related Classes of org.apache.lucene.codecs.BlockTermState

  • org.apache.lucene.codecs.blockterms.BlockTermsWriter$TermsWriter
  • org.apache.lucene.codecs.blocktree.BlockTreeTermsWriter$TermsWriter
  • org.apache.lucene.codecs.idversion.VersionBlockTreeTermsWriter$TermsWriter
  • org.apache.lucene.codecs.memory.FSTOrdTermsWriter$TermsWriter
  • org.apache.lucene.codecs.memory.FSTTermsWriter$TermsWriter
    • 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.