Examples of org.apache.lucene.index.DocsEnum

org.apache.lucene.index.DocsEnum
Iterates through the documents and term freqs. NOTE: you must first call {@link #nextDoc} before usingany of the per-doc methods.

    }


    // If we're still here, we have a cache miss. We need to fetch the
    // value from disk, and then also put it in the cache:
    int ret = TaxonomyReader.INVALID_ORDINAL;
    DocsEnum docs = MultiFields.getTermDocsEnum(indexReader, null, Consts.FULL, new BytesRef(FacetsConfig.pathToString(cp.components, cp.length)), 0);
    if (docs != null && docs.nextDoc() != DocIdSetIterator.NO_MORE_DOCS) {
      ret = docs.docID();
      
      // we only store the fact that a category exists, not its inexistence.
      // This is required because the caches are shared with new DTR instances
      // that are allocated from doOpenIfChanged. Therefore, if we only store
      // information about found categories, we cannot accidently tell a new

View Full Code Here

    int doc = -1;
    DirectoryReader reader = readerManager.acquire();
    try {
      final BytesRef catTerm = new BytesRef(FacetsConfig.pathToString(categoryPath.components, categoryPath.length));
      TermsEnum termsEnum = null; // reuse
      DocsEnum docs = null; // reuse
      for (AtomicReaderContext ctx : reader.leaves()) {
        Terms terms = ctx.reader().terms(Consts.FULL);
        if (terms != null) {
          termsEnum = terms.iterator(termsEnum);
          if (termsEnum.seekExact(catTerm)) {
            // liveDocs=null because the taxonomy has no deletes
            docs = termsEnum.docs(null, docs, 0 /* freqs not required */);
            // if the term was found, we know it has exactly one document.
            doc = docs.nextDoc() + ctx.docBase;
            break;
          }
        }
      }
    } finally {

View Full Code Here


    boolean aborted = false;
    DirectoryReader reader = readerManager.acquire();
    try {
      TermsEnum termsEnum = null;
      DocsEnum docsEnum = null;
      for (AtomicReaderContext ctx : reader.leaves()) {
        Terms terms = ctx.reader().terms(Consts.FULL);
        if (terms != null) { // cannot really happen, but be on the safe side
          termsEnum = terms.iterator(termsEnum);
          while (termsEnum.next() != null) {
            if (!cache.isFull()) {
              BytesRef t = termsEnum.term();
              // Since we guarantee uniqueness of categories, each term has exactly
              // one document. Also, since we do not allow removing categories (and
              // hence documents), there are no deletions in the index. Therefore, it
              // is sufficient to call next(), and then doc(), exactly once with no
              // 'validation' checks.
              FacetLabel cp = new FacetLabel(FacetsConfig.stringToPath(t.utf8ToString()));
              docsEnum = termsEnum.docs(null, docsEnum, DocsEnum.FLAG_NONE);
              boolean res = cache.put(cp, docsEnum.nextDoc() + ctx.docBase);
              assert !res : "entries should not have been evicted from the cache";
            } else {
              // the cache is full and the next put() will evict entries from it, therefore abort the iteration.
              aborted = true;
              break;

View Full Code Here

      final int size = r.numDocs();
      final OrdinalMap ordinalMap = map;
      ordinalMap.setSize(size);
      int base = 0;
      TermsEnum te = null;
      DocsEnum docs = null;
      for (final AtomicReaderContext ctx : r.leaves()) {
        final AtomicReader ar = ctx.reader();
        final Terms terms = ar.terms(Consts.FULL);
        te = terms.iterator(te);
        while (te.next() != null) {
          FacetLabel cp = new FacetLabel(FacetsConfig.stringToPath(te.term().utf8ToString()));
          final int ordinal = addCategory(cp);
          docs = te.docs(null, docs, DocsEnum.FLAG_NONE);
          ordinalMap.addMapping(docs.nextDoc() + base, ordinal);
        }
        base += ar.maxDoc(); // no deletions, so we're ok
      }
      ordinalMap.addDone();
    } finally {

View Full Code Here

      if (multipleValuesPerDocument) {
        if (scoreDocsInOrder) {
          AtomicReader slowCompositeReader = SlowCompositeReaderWrapper.wrap(toSearcher.getIndexReader());
          Terms terms = slowCompositeReader.terms(toField);
          if (terms != null) {
            DocsEnum docsEnum = null;
            TermsEnum termsEnum = null;
            SortedSet<BytesRef> joinValues = new TreeSet<BytesRef>(BytesRef.getUTF8SortedAsUnicodeComparator());
            joinValues.addAll(joinValueToJoinScores.keySet());
            for (BytesRef joinValue : joinValues) {
              termsEnum = terms.iterator(termsEnum);
              if (termsEnum.seekExact(joinValue)) {
                docsEnum = termsEnum.docs(slowCompositeReader.getLiveDocs(), docsEnum, DocsEnum.FLAG_NONE);
                JoinScore joinScore = joinValueToJoinScores.get(joinValue);


                for (int doc = docsEnum.nextDoc(); doc != DocIdSetIterator.NO_MORE_DOCS; doc = docsEnum.nextDoc()) {
                  // First encountered join value determines the score.
                  // Something to keep in mind for many-to-many relations.
                  if (!docToJoinScore.containsKey(doc)) {
                    docToJoinScore.put(doc, joinScore);
                  }

View Full Code Here

        if (otherMatchingDocs == null) {
          continue;
        }


        for (RandomDoc otherSideDoc : otherMatchingDocs) {
          DocsEnum docsEnum = MultiFields.getTermDocsEnum(topLevelReader, MultiFields.getLiveDocs(topLevelReader), "id", new BytesRef(otherSideDoc.id), 0);
          assert docsEnum != null;
          int doc = docsEnum.nextDoc();
          expectedResult.set(doc);
        }
      }
    }
    return expectedResult;

View Full Code Here

          @Override
          public DocIdSet getDocIdSet(AtomicReaderContext context,
              Bits acceptDocs) throws IOException {
            final boolean nullBitset = random().nextInt(10) == 5;
            final AtomicReader reader = context.reader();
            DocsEnum termDocsEnum = reader.termDocsEnum(new Term("field", "0"));
            if (termDocsEnum == null) {
              return null; // no docs -- return null
            }
            final BitSet bitSet = new BitSet(reader.maxDoc());
            int d;
            while ((d = termDocsEnum.nextDoc()) != DocsEnum.NO_MORE_DOCS) {
              bitSet.set(d, true);
            }
            return new DocIdSet() {
              
              @Override

View Full Code Here

          public Bits bits() throws IOException {
             return null;
          }
          @Override
          public DocIdSetIterator iterator() throws IOException {
            final DocsEnum termDocsEnum = context.reader().termDocsEnum(new Term("field", "0"));
            if (termDocsEnum == null) {
              return null;
            }
            return new DocIdSetIterator() {
              boolean nextCalled;
              boolean advanceCalled;
              @Override
              public int nextDoc() throws IOException {
                assertTrue("queryFirst: "+ queryFirst + " advanced: " + advanceCalled + " next: "+ nextCalled, nextCalled || advanceCalled ^ !queryFirst);  
                nextCalled = true;
                return termDocsEnum.nextDoc();
              }
              
              @Override
              public int docID() {
                return termDocsEnum.docID();
              }
              
              @Override
              public int advance(int target) throws IOException {
                assertTrue("queryFirst: "+ queryFirst + " advanced: " + advanceCalled + " next: "+ nextCalled, advanceCalled || nextCalled ^ queryFirst);  
                advanceCalled = true;
                return termDocsEnum.advance(target);
              }
              
              @Override
              public long cost() {
                return termDocsEnum.cost();
              } 
            };
          }
          
        };

View Full Code Here

            
          }
        } 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

  public void testDocsEnumStart() throws Exception {
    Analyzer analyzer = new MockAnalyzer(random());
    MemoryIndex memory = new MemoryIndex(random().nextBoolean(),  random().nextInt(50) * 1024 * 1024);
    memory.addField("foo", "bar", analyzer);
    AtomicReader reader = (AtomicReader) memory.createSearcher().getIndexReader();
    DocsEnum disi = _TestUtil.docs(random(), reader, "foo", new BytesRef("bar"), null, null, DocsEnum.FLAG_NONE);
    int docid = disi.docID();
    assertEquals(-1, docid);
    assertTrue(disi.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
    
    // now reuse and check again
    TermsEnum te = reader.terms("foo").iterator(null);
    assertTrue(te.seekExact(new BytesRef("bar")));
    disi = te.docs(null, disi, DocsEnum.FLAG_NONE);
    docid = disi.docID();
    assertEquals(-1, docid);
    assertTrue(disi.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
    reader.close();
  }

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

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

com.basho.yokozuna.handler.EntropyData

com.browseengine.bobo.facets.data.FacetDataCache

com.browseengine.bobo.facets.data.MultiValueFacetDataCache

com.browseengine.bobo.facets.data.MultiValueWithWeightFacetDataCache

com.browseengine.bobo.facets.filter.AdaptiveFacetFilter$TermListRandomAccessDocIdSet$TermDocIdSet

com.browseengine.bobo.facets.impl.CompactMultiValueFacetHandler

com.pearson.entech.elasticsearch.search.facet.approx.termlist.TermListFacetExecutor$PostExecutor

org.apache.blur.lucene.search.PrimeDocCache

org.apache.blur.lucene.warmup.IndexTracer

org.apache.blur.utils.BlurUtil

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.