Examples of org.apache.lucene.index.DocsAndPositionsEnum

org.apache.lucene.index.DocsAndPositionsEnum
Also iterates through positions.

    DocsQueue(List<DocsAndPositionsEnum> docsEnums) throws IOException {
      super(docsEnums.size());


      Iterator<DocsAndPositionsEnum> i = docsEnums.iterator();
      while (i.hasNext()) {
        DocsAndPositionsEnum postings = i.next();
        if (postings.nextDoc() != DocIdSetIterator.NO_MORE_DOCS) {
          add(postings);
        }
      }
    }

View Full Code Here

    final AtomicReader reader = getOnlySegmentReader(writer.getReader());
    final Fields fields = reader.getTermVectors(0);
    final Terms terms = fields.terms("f");
    final TermsEnum te = terms.iterator(null);
    assertEquals(new BytesRef("a"), te.next());
    final DocsAndPositionsEnum dpe = te.docsAndPositions(null, null);
    assertEquals(0, dpe.nextDoc());
    assertEquals(2, dpe.freq());
    assertEquals(0, dpe.nextPosition());
    assertEquals(0, dpe.startOffset());
    final int endOffset = dpe.endOffset();
    assertEquals(1 + positionGap, dpe.nextPosition());
    assertEquals(1 + endOffset + offsetGap, dpe.endOffset());
    assertEquals(null, te.next());
    reader.close();
    writer.close();
    writer.w.getDirectory().close();
  }

View Full Code Here

   * if deep is false, it does a 'shallow' test that doesnt go down to the docsenums
   */
  public void assertTermsEnum(TermsEnum leftTermsEnum, TermsEnum rightTermsEnum, boolean deep) throws Exception {
    BytesRef term;
    Bits randomBits = new RandomBits(MAXDOC, random().nextDouble(), random());
    DocsAndPositionsEnum leftPositions = null;
    DocsAndPositionsEnum rightPositions = null;
    DocsEnum leftDocs = null;
    DocsEnum rightDocs = null;
    
    while ((term = leftTermsEnum.next()) != null) {
      assertEquals(term, rightTermsEnum.next());

View Full Code Here

          final boolean offsets = iwTerms.hasOffsets() && memTerms.hasOffsets();
         
          while(iwTermsIter.next() != null) {
            assertNotNull(memTermsIter.next());
            assertEquals(iwTermsIter.term(), memTermsIter.term());
            DocsAndPositionsEnum iwDocsAndPos = iwTermsIter.docsAndPositions(null, null);
            DocsAndPositionsEnum memDocsAndPos = memTermsIter.docsAndPositions(null, null);
            while(iwDocsAndPos.nextDoc() != DocsAndPositionsEnum.NO_MORE_DOCS) {
              assertEquals(iwDocsAndPos.docID(), memDocsAndPos.nextDoc());
              assertEquals(iwDocsAndPos.freq(), memDocsAndPos.freq());
              for (int i = 0; i < iwDocsAndPos.freq(); i++) {
                assertEquals("term: " + iwTermsIter.term().utf8ToString(), iwDocsAndPos.nextPosition(), memDocsAndPos.nextPosition());
                if (offsets) {
                  assertEquals(iwDocsAndPos.startOffset(), memDocsAndPos.startOffset());
                  assertEquals(iwDocsAndPos.endOffset(), memDocsAndPos.endOffset());
                }
              }
              
            }
            
          }
        } else {
          while(iwTermsIter.next() != null) {
            assertEquals(iwTermsIter.term(), memTermsIter.term());
            DocsEnum iwDocsAndPos = iwTermsIter.docs(null, null);
            DocsEnum memDocsAndPos = memTermsIter.docs(null, null);
            while(iwDocsAndPos.nextDoc() != DocsAndPositionsEnum.NO_MORE_DOCS) {
              assertEquals(iwDocsAndPos.docID(), memDocsAndPos.nextDoc());
              assertEquals(iwDocsAndPos.freq(), memDocsAndPos.freq());
            }
          }
        }
      }

View Full Code Here

    MemoryIndex memory = new MemoryIndex(true,  random().nextInt(50) * 1024 * 1024);
    for (int i = 0; i < numIters; i++) { // check reuse
      memory.addField("foo", "bar", analyzer);
      AtomicReader reader = (AtomicReader) memory.createSearcher().getIndexReader();
      assertEquals(1, reader.terms("foo").getSumTotalTermFreq());
      DocsAndPositionsEnum disi = reader.termPositionsEnum(new Term("foo", "bar"));
      int docid = disi.docID();
      assertEquals(-1, docid);
      assertTrue(disi.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
      assertEquals(0, disi.nextPosition());
      assertEquals(0, disi.startOffset());
      assertEquals(3, disi.endOffset());
      
      // now reuse and check again
      TermsEnum te = reader.terms("foo").iterator(null);
      assertTrue(te.seekExact(new BytesRef("bar")));
      disi = te.docsAndPositions(null, disi);
      docid = disi.docID();
      assertEquals(-1, docid);
      assertTrue(disi.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
      reader.close();
      memory.reset();
    }
  }

View Full Code Here


    while (termEnum.next() != null) {
      assertNotNull(memTermEnum.next());
      assertThat(termEnum.totalTermFreq(), equalTo(memTermEnum.totalTermFreq()));


      DocsAndPositionsEnum docsPosEnum = termEnum.docsAndPositions(null, null, 0);
      DocsAndPositionsEnum memDocsPosEnum = memTermEnum.docsAndPositions(null, null, 0);
      String currentTerm = termEnum.term().utf8ToString();


      assertThat("Token mismatch for field: " + field_name, currentTerm, equalTo(memTermEnum.term().utf8ToString()));


      docsPosEnum.nextDoc();
      memDocsPosEnum.nextDoc();


      int freq = docsPosEnum.freq();
      assertThat(freq, equalTo(memDocsPosEnum.freq()));
      for (int i = 0; i < freq; i++) {
        String failDesc = " (field:" + field_name + " term:" + currentTerm + ")";
        int memPos = memDocsPosEnum.nextPosition();
        int pos = docsPosEnum.nextPosition();
        assertThat("Position test failed" + failDesc, memPos, equalTo(pos));
        assertThat("Start offset test failed" + failDesc, memDocsPosEnum.startOffset(), equalTo(docsPosEnum.startOffset()));
        assertThat("End offset test failed" + failDesc, memDocsPosEnum.endOffset(), equalTo(docsPosEnum.endOffset()));
        assertThat("Missing payload test failed" + failDesc, docsPosEnum.getPayload(), equalTo(null));
      }
    }
    assertNull("Still some tokens not processed", memTermEnum.next());
  }

View Full Code Here

          case 1: posFlags = DocsAndPositionsEnum.FLAG_OFFSETS; break;
          case 2: posFlags = DocsAndPositionsEnum.FLAG_PAYLOADS; break;
          default: posFlags = DocsAndPositionsEnum.FLAG_OFFSETS | DocsAndPositionsEnum.FLAG_PAYLOADS; break;
        }
        // TODO: cast to DocsAndPositionsEnum?
        DocsAndPositionsEnum docsAndPositions = termsEnum.docsAndPositions(liveDocs, null, posFlags);
        if (docsAndPositions != null) {
          return docsAndPositions;
        }
      }
      flags |= DocsEnum.FLAG_FREQS;

View Full Code Here

   * if deep is false, it does a 'shallow' test that doesnt go down to the docsenums
   */
  public void assertTermsEnumEquals(String info, IndexReader leftReader, TermsEnum leftTermsEnum, TermsEnum rightTermsEnum, boolean deep) throws IOException {
    BytesRef term;
    Bits randomBits = new RandomBits(leftReader.maxDoc(), random().nextDouble(), random());
    DocsAndPositionsEnum leftPositions = null;
    DocsAndPositionsEnum rightPositions = null;
    DocsEnum leftDocs = null;
    DocsEnum rightDocs = null;
    
    while ((term = leftTermsEnum.next()) != null) {
      assertEquals(info, term, rightTermsEnum.next());

View Full Code Here

    // TODO: we could use CachingWrapperFilter, (or consume twice) to allow us to have a true freq()
    // but this would have a performance cost for likely little gain in the user experience, it
    // would only serve to make this method less bogus.
    // instead, we always return freq() = Integer.MAX_VALUE and let PH terminate based on offset...
    
    return new DocsAndPositionsEnum() {
      int currentDoc = -1;
      int currentMatch = -1;
      int currentStartOffset = -1;
      int currentEndOffset = -1;
      TokenStream stream = ts;

View Full Code Here

      return new SortingDocsEnum(docMap.size(), wrapReuse, inDocs, withFreqs, docMap);
    }


    @Override
    public DocsAndPositionsEnum docsAndPositions(Bits liveDocs, DocsAndPositionsEnum reuse, final int flags) throws IOException {
      final DocsAndPositionsEnum inReuse;
      final SortingDocsAndPositionsEnum wrapReuse;
      if (reuse != null && reuse instanceof SortingDocsAndPositionsEnum) {
        // if we're asked to reuse the given DocsEnum and it is Sorting, return
        // the wrapped one, since some Codecs expect it.
        wrapReuse = (SortingDocsAndPositionsEnum) reuse;
        inReuse = wrapReuse.getWrapped();
      } else {
        wrapReuse = null;
        inReuse = reuse;
      }


      final DocsAndPositionsEnum inDocsAndPositions = in.docsAndPositions(newToOld(liveDocs), inReuse, flags);
      if (inDocsAndPositions == null) {
        return null;
      }


      // we ignore the fact that offsets may be stored but not asked for,

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.lucene.index.DocsAndPositionsEnum

com.browseengine.bobo.facets.data.MultiValueFacetDataCache$AllocOnlyLoader

com.browseengine.bobo.search.section.IntMetaDataCache

com.browseengine.bobo.sort.SortCollectorImpl

org.apache.blur.lucene.warmup.IndexTracer

org.apache.lucene.analysis.core.TestClassicAnalyzer

org.apache.lucene.analysis.sinks.TestTeeSinkTokenFilter

org.apache.lucene.analysis.TestCachingTokenFilter

org.apache.lucene.analysis.TestMockAnalyzer

org.apache.lucene.codecs.lucene41.TestBlockPostingsFormat3

org.apache.lucene.codecs.memory.DirectPostingsFormat$DirectField

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.