Example usage for org.apache.lucene.index IndexReader maxDoc

Introduction

In this page you can find the example usage for org.apache.lucene.index IndexReader maxDoc.

Prototype

public abstract int maxDoc();

Source Link

Document

Returns one greater than the largest possible document number.

Usage

From source file:org.apache.jena.larq.TestLuceneNRT.java

License:Apache License

@Test
public void indexReaderOpenDirectoryNewInstance() throws Exception {
    IndexReader reader = IndexReader.open(directory);
    writer.addDocument(doc1);/*from  ww w . ja  v a2s  . c o m*/
    writer.commit();

    assertFalse(reader.isCurrent());
    assertEquals(0, reader.maxDoc());
    reader = IndexReader.open(directory);
    assertTrue(reader.isCurrent());
    assertEquals(1, reader.maxDoc());
}

---

From source file:org.apache.jena.larq.TestLuceneNRT.java

License:Apache License

@Test
public void indexReaderOpenWriter() throws Exception {
    IndexReader reader = IndexReader.open(writer, true);
    assertNotNull(reader);//from ww  w.jav a2s .  co m
    assertSame(directory, reader.directory());
    assertEquals(1, reader.getRefCount());
    assertTrue(reader.isCurrent());
    assertEquals(0, reader.maxDoc());
    close(reader);
    assertEquals(0, reader.getRefCount());
}

---

From source file:org.apache.jena.larq.TestLuceneNRT.java

License:Apache License

@Test
public void indexReaderOpenWriterAddDocument() throws Exception {
    IndexReader reader = IndexReader.open(writer, true);
    writer.addDocument(doc1);/*from w  w w  .ja  v  a  2 s . c o  m*/
    assertFalse(reader.isCurrent());
    assertEquals(1, writer.maxDoc());
    assertEquals(0, reader.maxDoc());
}

---

From source file:org.apache.jena.larq.TestLuceneNRT.java

License:Apache License

@Test
public void indexReaderOpenWriterAddDocumentAndCommit() throws Exception {
    IndexReader reader = IndexReader.open(writer, true);
    writer.addDocument(doc1);//from www.  j ava  2 s . co m
    writer.commit();
    assertFalse(reader.isCurrent());
    assertEquals(1, writer.maxDoc());
    assertEquals(0, reader.maxDoc());
}

---

From source file:org.apache.jena.larq.TestLuceneNRT.java

License:Apache License

@Test
public void indexReaderOpenWriterReopen() throws Exception {
    IndexReader reader = IndexReader.open(writer, true);
    writer.addDocument(doc1);/*from   ww w  .j  av  a2  s .com*/
    writer.commit();

    assertFalse(reader.isCurrent());
    assertEquals(0, reader.maxDoc());
    reader = TestLARQUtils.openIfChanged(reader);
    assertTrue(reader.isCurrent());
    assertEquals(1, reader.maxDoc());
}

---

From source file:org.apache.jena.larq.TestLuceneNRT.java

License:Apache License

@Test
public void indexReaderOpenWriterNewInstance() throws Exception {
    IndexReader reader = IndexReader.open(writer, true);
    writer.addDocument(doc1);//from  w w  w .  j  av a  2s  .com
    writer.commit();

    assertFalse(reader.isCurrent());
    assertEquals(0, reader.maxDoc());
    reader = IndexReader.open(writer, true);
    assertTrue(reader.isCurrent());
    assertEquals(1, reader.maxDoc());
}

---

From source file:org.apache.jena.larq.TestLuceneNRT.java

License:Apache License

@Test
public void indexReaderOpenWriterDeleteDocumentAndCommit() throws Exception {
    Term term = new Term("foo", "bar1");
    IndexReader reader = IndexReader.open(writer, false);
    writer.addDocument(doc1);//from   ww w. j  a  v  a 2 s.  c  om
    writer.addDocument(doc2);
    writer.commit();

    assertEquals(2, writer.maxDoc());
    assertFalse(reader.isCurrent());
    assertEquals(0, reader.maxDoc());
    assertEquals(0, count(reader, term));

    reader = TestLARQUtils.openIfChanged(reader);
    assertTrue(reader.isCurrent());
    assertEquals(1, count(reader, term));

    writer.deleteDocuments(term);
    assertEquals(2, writer.maxDoc());
    assertEquals(1, count(reader, term));

    reader = TestLARQUtils.openIfChanged(reader);
    assertEquals(1, count(reader, term));

    writer.commit();
    reader = TestLARQUtils.openIfChanged(reader);
    assertEquals(2, writer.maxDoc());
    assertEquals(0, count(reader, term));

    reader = TestLARQUtils.openIfChanged(reader);
    assertEquals(0, count(reader, term));

    writer.forceMergeDeletes();
    assertEquals(1, writer.maxDoc());
    assertEquals(0, count(reader, term));
}

---

From source file:org.apache.mahout.utils.vectors.lucene.ClusterLabels.java

License:Apache License

/**
 * Get the list of labels, sorted by best score.
 *//*  w  w w. ja  va  2s  . c o  m*/
protected List<TermInfoClusterInOut> getClusterLabels(Integer integer,
        Collection<WeightedPropertyVectorWritable> wpvws) throws IOException {

    if (wpvws.size() < minNumIds) {
        log.info("Skipping small cluster {} with size: {}", integer, wpvws.size());
        return null;
    }

    log.info("Processing Cluster {} with {} documents", integer, wpvws.size());
    Directory dir = FSDirectory.open(new File(this.indexDir));
    IndexReader reader = DirectoryReader.open(dir);

    log.info("# of documents in the index {}", reader.numDocs());

    Collection<String> idSet = Sets.newHashSet();
    for (WeightedPropertyVectorWritable wpvw : wpvws) {
        Vector vector = wpvw.getVector();
        if (vector instanceof NamedVector) {
            idSet.add(((NamedVector) vector).getName());
        }
    }

    int numDocs = reader.numDocs();

    OpenBitSet clusterDocBitset = getClusterDocBitset(reader, idSet, this.idField);

    log.info("Populating term infos from the index");

    /**
     * This code is as that of CachedTermInfo, with one major change, which is to get the document frequency.
     * 
     * Since we have deleted the documents out of the cluster, the document frequency for a term should only
     * include the in-cluster documents. The document frequency obtained from TermEnum reflects the frequency
     * in the entire index. To get the in-cluster frequency, we need to query the index to get the term
     * frequencies in each document. The number of results of this call will be the in-cluster document
     * frequency.
     */
    Terms t = MultiFields.getTerms(reader, contentField);
    TermsEnum te = t.iterator(null);
    Map<String, TermEntry> termEntryMap = new LinkedHashMap<String, TermEntry>();
    Bits liveDocs = MultiFields.getLiveDocs(reader); //WARNING: returns null if there are no deletions

    int count = 0;
    BytesRef term;
    while ((term = te.next()) != null) {
        OpenBitSet termBitset = new OpenBitSet(reader.maxDoc());
        DocsEnum docsEnum = MultiFields.getTermDocsEnum(reader, null, contentField, term);
        int docID;
        while ((docID = docsEnum.nextDoc()) != DocIdSetIterator.NO_MORE_DOCS) {
            //check to see if we don't have an deletions (null) or if document is live
            if (liveDocs != null && !liveDocs.get(docID)) {
                // document is deleted...
                termBitset.set(docsEnum.docID());
            }
        }
        // AND the term's bitset with cluster doc bitset to get the term's in-cluster frequency.
        // This modifies the termBitset, but that's fine as we are not using it anywhere else.
        termBitset.and(clusterDocBitset);
        int inclusterDF = (int) termBitset.cardinality();

        TermEntry entry = new TermEntry(term.utf8ToString(), count++, inclusterDF);
        termEntryMap.put(entry.getTerm(), entry);

    }

    List<TermInfoClusterInOut> clusteredTermInfo = Lists.newLinkedList();

    int clusterSize = wpvws.size();

    for (TermEntry termEntry : termEntryMap.values()) {

        int corpusDF = reader.docFreq(new Term(this.contentField, termEntry.getTerm()));
        int outDF = corpusDF - termEntry.getDocFreq();
        int inDF = termEntry.getDocFreq();
        double logLikelihoodRatio = scoreDocumentFrequencies(inDF, outDF, clusterSize, numDocs);
        TermInfoClusterInOut termInfoCluster = new TermInfoClusterInOut(termEntry.getTerm(), inDF, outDF,
                logLikelihoodRatio);
        clusteredTermInfo.add(termInfoCluster);
    }

    Collections.sort(clusteredTermInfo);
    // Cleanup
    Closeables.close(reader, true);
    termEntryMap.clear();

    return clusteredTermInfo.subList(0, Math.min(clusteredTermInfo.size(), maxLabels));
}

---

From source file:org.apache.maven.index.AbstractRepoNexusIndexerTest.java

License:Apache License

public void testPackaging() throws Exception {
    IndexReader reader = context.acquireIndexSearcher().getIndexReader();

    Bits liveDocs = MultiFields.getLiveDocs(reader);
    for (int i = 0; i < reader.maxDoc(); i++) {
        if (liveDocs == null || liveDocs.get(i)) {
            Document document = reader.document(i);

            String uinfo = document.get(ArtifactInfo.UINFO);

            if (uinfo != null) {
                String info = document.get(ArtifactInfo.INFO);
                assertFalse("Bad:" + info, info.startsWith("null"));
            }/*ww w . j ava2s  .c  om*/
        }
    }

    // {
    // Query query = new TermQuery( new Term( MAVEN.PACKAGING, "jar" ) );
    // FlatSearchResponse response = nexusIndexer.searchFlat(new FlatSearchRequest(query));
    // assertEquals(response.getResults().toString(), 22, response.getTotalHits());
    // }
    {
        Query query = nexusIndexer.constructQuery(MAVEN.PACKAGING, "tar.gz", SearchType.EXACT);
        FlatSearchResponse response = nexusIndexer.searchFlat(new FlatSearchRequest(query));
        assertEquals(response.getResults().toString(), 1, response.getTotalHits());

        ArtifactInfo ai = response.getResults().iterator().next();
        assertEquals("tar.gz", ai.getPackaging());
        assertEquals("tar.gz", ai.getFileExtension());
    }
    {
        Query query = nexusIndexer.constructQuery(MAVEN.PACKAGING, "zip", SearchType.EXACT);
        FlatSearchResponse response = nexusIndexer.searchFlat(new FlatSearchRequest(query));
        assertEquals(response.getResults().toString(), 1, response.getTotalHits());

        ArtifactInfo ai = response.getResults().iterator().next();
        assertEquals("zip", ai.getPackaging());
        assertEquals("zip", ai.getFileExtension());
    }
}

---

From source file:org.apache.maven.index.context.DefaultIndexingContext.java

License:Apache License

public synchronized void merge(Directory directory, DocumentFilter filter) throws IOException {
    final IndexSearcher s = acquireIndexSearcher();
    try {//from  w  w  w  .j a  v  a2s .c om
        final IndexWriter w = getIndexWriter();
        final IndexReader directoryReader = DirectoryReader.open(directory);
        TopScoreDocCollector collector = null;
        try {
            int numDocs = directoryReader.maxDoc();

            Bits liveDocs = MultiFields.getLiveDocs(directoryReader);
            for (int i = 0; i < numDocs; i++) {
                if (liveDocs != null && !liveDocs.get(i)) {
                    continue;
                }

                Document d = directoryReader.document(i);
                if (filter != null && !filter.accept(d)) {
                    continue;
                }

                String uinfo = d.get(ArtifactInfo.UINFO);
                if (uinfo != null) {
                    collector = TopScoreDocCollector.create(1);
                    s.search(new TermQuery(new Term(ArtifactInfo.UINFO, uinfo)), collector);
                    if (collector.getTotalHits() == 0) {
                        w.addDocument(IndexUtils.updateDocument(d, this, false));
                    }
                } else {
                    String deleted = d.get(ArtifactInfo.DELETED);

                    if (deleted != null) {
                        // Deleting the document loses history that it was delete,
                        // so incrementals wont work. Therefore, put the delete
                        // document in as well
                        w.deleteDocuments(new Term(ArtifactInfo.UINFO, deleted));
                        w.addDocument(d);
                    }
                }
            }

        } finally {
            directoryReader.close();
            commit();
        }

        rebuildGroups();
        Date mergedTimestamp = IndexUtils.getTimestamp(directory);

        if (getTimestamp() != null && mergedTimestamp != null && mergedTimestamp.after(getTimestamp())) {
            // we have both, keep the newest
            updateTimestamp(true, mergedTimestamp);
        } else {
            updateTimestamp(true);
        }
        optimize();
    } finally {
        releaseIndexSearcher(s);
    }
}

---

• «
• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• 13
• 14
• 15
• 16
• 17
• 18
• 19
• 20
• 21
• 22
• 23
• 24
• 25
• 26
• 27
• 28
• 29
• 30
• 31
• 32
• 33
• 34
• »