Examples of DocsAndPositionsEnum

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

Examples of org.apache.lucene.index.DocsAndPositionsEnum

    // waste CPU merging them:
    _posList.clear();
    _doc = _queue.top().docID();


    // merge sort all positions together
    DocsAndPositionsEnum postings;
    do {
      postings = _queue.top();


      final int freq = postings.freq();
      for (int i = 0; i < freq; i++) {
        _posList.add(postings.nextPosition());
      }


      if (postings.nextDoc() != NO_MORE_DOCS) {
        _queue.updateTop();
      } else {
        _queue.pop();
      }
    } while (_queue.size() > 0 && _queue.top().docID() == _doc);

View Full Code Here

Examples of org.apache.lucene.index.DocsAndPositionsEnum

  }


  @Override
  public final int advance(int target) throws IOException {
    while (_queue.top() != null && target > _queue.top().docID()) {
      DocsAndPositionsEnum postings = _queue.pop();
      if (postings.advance(target) != NO_MORE_DOCS) {
        _queue.add(postings);
      }
    }
    return nextDoc();
  }

View Full Code Here

Examples of org.apache.lucene.index.DocsAndPositionsEnum

    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

Examples of org.apache.lucene.index.DocsAndPositionsEnum

      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

Examples of org.apache.lucene.index.DocsAndPositionsEnum

   * 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

Examples of org.apache.lucene.index.DocsAndPositionsEnum

    }
    
    assertEquals(2, allEnums.size());
    
    allEnums.clear();
    DocsAndPositionsEnum posReuse = null;
    te = segment.terms("foo").iterator(null);
    while (te.next() != null) {
      posReuse = te.docsAndPositions(null, posReuse);
      allEnums.put(posReuse, true);
    }

View Full Code Here

Examples of org.apache.lucene.index.DocsAndPositionsEnum

    }
    
    assertEquals(4, allEnums.size());
    
    allEnums.clear();
    DocsAndPositionsEnum posReuse = null;
    te = segment.terms("foo").iterator(null);
    while (te.next() != null) {
      posReuse = te.docsAndPositions(null, posReuse);
      allEnums.put(posReuse, true);
    }

View Full Code Here

Examples of org.apache.lucene.index.DocsAndPositionsEnum

    payloadAttribute = addAttribute(PayloadAttribute.class);
    final boolean hasOffsets = vector.hasOffsets();
    final boolean hasPayloads = vector.hasPayloads();
    final TermsEnum termsEnum = vector.iterator(null);
    BytesRef text;
    DocsAndPositionsEnum dpEnum = null;
    while((text = termsEnum.next()) != null) {
      dpEnum = termsEnum.docsAndPositions(null, dpEnum);
      assert dpEnum != null; // presumably checked by TokenSources.hasPositions earlier
      dpEnum.nextDoc();
      final int freq = dpEnum.freq();
      for (int j = 0; j < freq; j++) {
        int pos = dpEnum.nextPosition();
        Token token;
        if (hasOffsets) {
          token = new Token(text.utf8ToString(),
                            dpEnum.startOffset(),
                            dpEnum.endOffset());
        } else {
          token = new Token();
          token.setEmpty().append(text.utf8ToString());
        }
        if (hasPayloads) {
          // Must make a deep copy of the returned payload,
          // since D&PEnum API is allowed to re-use on every
          // call:
          token.setPayload(BytesRef.deepCopyOf(dpEnum.getPayload()));
        }


        // Yes - this is the position, not the increment! This is for
        // sorting. This value
        // will be corrected before use.

View Full Code Here

Examples of org.apache.lucene.index.DocsAndPositionsEnum

      return;
    }


    final CharsRef spare = new CharsRef();
    final TermsEnum termsEnum = vector.iterator(null);
    DocsAndPositionsEnum dpEnum = null;
    BytesRef text;
    
    int numDocs = reader.maxDoc();
    
    while ((text = termsEnum.next()) != null) {
      UnicodeUtil.UTF8toUTF16(text, spare);
      final String term = spare.toString();
      if (!termSet.contains(term)) {
        continue;
      }
      dpEnum = termsEnum.docsAndPositions(null, dpEnum);
      if (dpEnum == null) {
        // null snippet
        return;
      }


      dpEnum.nextDoc();
      
      // For weight look here: http://lucene.apache.org/core/3_6_0/api/core/org/apache/lucene/search/DefaultSimilarity.html
      final float weight = ( float ) ( Math.log( numDocs / ( double ) ( reader.docFreq( new Term(fieldName, text) ) + 1 ) ) + 1.0 );


      final int freq = dpEnum.freq();
      
      for(int i = 0;i < freq;i++) {
        int pos = dpEnum.nextPosition();
        if (dpEnum.startOffset() < 0) {
          return; // no offsets, null snippet
        }
        termList.add( new TermInfo( term, dpEnum.startOffset(), dpEnum.endOffset(), pos, weight ) );
      }
    }
    
    // sort by position
    Collections.sort(termList);

View Full Code Here

Examples of org.apache.lucene.index.DocsAndPositionsEnum

  
  @Test
  public void testDocsAndPositionsEnum() throws Exception {
    TermsEnum termsEnum = reader.terms(DOC_POSITIONS_FIELD).iterator(null);
    assertEquals(SeekStatus.FOUND, termsEnum.seekCeil(new BytesRef(DOC_POSITIONS_TERM)));
    DocsAndPositionsEnum sortedPositions = termsEnum.docsAndPositions(null, null);
    int doc;
    
    // test nextDoc()
    while ((doc = sortedPositions.nextDoc()) != DocIdSetIterator.NO_MORE_DOCS) {
      int freq = sortedPositions.freq();
      assertEquals("incorrect freq for doc=" + doc, sortedValues[doc].intValue() / 10 + 1, freq);
      for (int i = 0; i < freq; i++) {
        assertEquals("incorrect position for doc=" + doc, i, sortedPositions.nextPosition());
        if (!doesntSupportOffsets.contains(_TestUtil.getPostingsFormat(DOC_POSITIONS_FIELD))) {
          assertEquals("incorrect startOffset for doc=" + doc, i, sortedPositions.startOffset());
          assertEquals("incorrect endOffset for doc=" + doc, i, sortedPositions.endOffset());
        }
        assertEquals("incorrect payload for doc=" + doc, freq - i, Integer.parseInt(sortedPositions.getPayload().utf8ToString()));
      }
    }
    
    // test advance()
    final DocsAndPositionsEnum reuse = sortedPositions;
    sortedPositions = termsEnum.docsAndPositions(null, reuse);
    if (sortedPositions instanceof SortingDocsAndPositionsEnum) {
      assertTrue(((SortingDocsAndPositionsEnum) sortedPositions).reused(reuse)); // make sure reuse worked
    }
    doc = 0;

View Full Code Here

0 1 2 3 4 5

TOP

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.