Java Code Examples for org.apache.lucene.index.DirectoryReader#maxDoc()

@Test
public void testMultipleWritersOpenOnSameDirectory() throws IOException {
  IndexWriterConfig config = new IndexWriterConfig(Version.LUCENE_43, new KeywordAnalyzer());
  FastHdfsKeyValueDirectory directory = new FastHdfsKeyValueDirectory(false, _timer, _configuration, new Path(_path,
      "test_multiple"));
  IndexWriter writer1 = new IndexWriter(directory, config.clone());
  addDoc(writer1, getDoc(1));
  IndexWriter writer2 = new IndexWriter(directory, config.clone());
  addDoc(writer2, getDoc(2));
  writer1.close();
  writer2.close();

  DirectoryReader reader = DirectoryReader.open(directory);
  int maxDoc = reader.maxDoc();
  assertEquals(1, maxDoc);
  Document document = reader.document(0);
  assertEquals("2", document.get("id"));
  reader.close();
}
 

public void testDocMakerThreadSafety() throws Exception {
  // 1. alg definition (required in every "logic" test)
  String algLines[] = {
      "# ----- properties ",
      "content.source=org.apache.lucene.benchmark.byTask.feeds.SortableSingleDocSource",
      "doc.term.vector=false",
      "log.step.AddDoc=10000",
      "content.source.forever=true",
      "directory=ByteBuffersDirectory",
      "doc.reuse.fields=false",
      "doc.stored=true",
      "doc.tokenized=false",
      "doc.index.props=true",
      "# ----- alg ",
      "CreateIndex",
      "[ { AddDoc > : 250 ] : 4",
      "CloseIndex",
  };
  
  // 2. we test this value later
  CountingSearchTestTask.numSearches = 0;
  
  // 3. execute the algorithm  (required in every "logic" test)
  Benchmark benchmark = execBenchmark(algLines);

  DirectoryReader r = DirectoryReader.open(benchmark.getRunData().getDirectory());
  
  final int maxDoc = r.maxDoc();
  assertEquals(1000, maxDoc);
  for(int i=0;i<1000;i++) {
    assertNotNull("doc " + i + " has null country", r.document(i).getField("country"));
  }
  r.close();
}
 

private void verifyResults(Directory indexDir, Directory taxoDir) throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  DirectoryTaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
  IndexSearcher searcher = newSearcher(indexReader);
  
  FacetsCollector collector = new FacetsCollector();
  FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, collector);

  // tag facets
  Facets tagFacets = new FastTaxonomyFacetCounts("$tags", taxoReader, facetConfig, collector);
  FacetResult result = tagFacets.getTopChildren(10, "tag");
  for (LabelAndValue lv: result.labelValues) {
    if (VERBOSE) {
      System.out.println(lv);
    }
    assertEquals(NUM_DOCS, lv.value.intValue());
  }
  
  // id facets
  Facets idFacets = new FastTaxonomyFacetCounts(taxoReader, facetConfig, collector);
  FacetResult idResult = idFacets.getTopChildren(10, "id");
  assertEquals(NUM_DOCS, idResult.childCount);
  assertEquals(NUM_DOCS * 2, idResult.value); // each "id" appears twice
  
  BinaryDocValues bdv = MultiDocValues.getBinaryValues(indexReader, "bdv");
  BinaryDocValues cbdv = MultiDocValues.getBinaryValues(indexReader, "cbdv");
  for (int i = 0; i < indexReader.maxDoc(); i++) {
    assertEquals(i, bdv.nextDoc());
    assertEquals(i, cbdv.nextDoc());
    assertEquals(Integer.parseInt(cbdv.binaryValue().utf8ToString()), Integer.parseInt(bdv.binaryValue().utf8ToString())*2);
  }
  IOUtils.close(indexReader, taxoReader);
}
 

private int getDocumentCount(Directory directory) throws IOException {
  if (DirectoryReader.indexExists(directory)) {
    DirectoryReader reader = DirectoryReader.open(directory);
    int maxDoc = reader.maxDoc();
    reader.close();
    return maxDoc;
  }
  return 0;
}
 

public static void main(String[] args) throws CorruptIndexException, IOException, SolrServerException {

        if (args.length < 3) {
            System.err.println("Usage: java -Dfile.encoding=UTF8 -Dclient.encoding.override=UTF-8 -Xmx256m -Xms256m -server " + IndexLoader.class.getName()
                    + " </path/to/index> <AutoCompleteSolrUrl> <indexField1,acField1> [indexField2,acField2 ... ]");
            System.exit(0);
        }
        Map<String,String> fieldMap = getFieldMapping(args, 2);
        DirectoryReader reader = DirectoryReader.open(FSDirectory.open(Paths.get(args[0])));
        int docs = reader.maxDoc();
        SolrClient solr = new ConcurrentUpdateSolrClient.Builder(args[1]).withQueueSize(10000).withThreadCount(2).build();
        Set<SolrInputDocument> batch = new HashSet<SolrInputDocument>(1000);
        
        Bits liveDocs = MultiFields.getLiveDocs(reader);
        
        // go through all docs in the index
        for (int i = 0; i < docs; i++) {
            // process doc only if not deleted
            if (liveDocs == null || liveDocs.get(i)) {
                // loop through all fields to be looked at
                SolrInputDocument doc = new SolrInputDocument();
                Iterator<String> iter = fieldMap.keySet().iterator();
                
                boolean phraseFieldEmpty = false;
                
                while (iter.hasNext()) {
                    String indexField = iter.next();
                    String acField = fieldMap.get(indexField);
                    IndexableField field = reader.document(i).getField(indexField);
                    String value = field != null ? reader.document(i).getField(indexField).stringValue() : null;
                    
                    if (field != null && value != null && !value.isEmpty()) {
                      doc.addField(acField, value);
                    } else {
                      // not very relevant piece of info
                      // System.err.println("Field is null or empty, skipping: " + indexField);
                      
                      if (acField.equalsIgnoreCase("phrase")) {
                        System.err.println("Since AC phrase field would be null, this doc will not be created: " + reader.document(i));
                        phraseFieldEmpty = true;
                        break;
                      }
                    }
                }

                if (!phraseFieldEmpty) {
                  solr.add(doc);
                  if (docs % 1000 == 0) {
                    System.out.println("Docs: " + docs);
                  }
                }
            }
        }
        if (!batch.isEmpty())
            solr.add(batch);
        reader.close();
        System.out.println("Optimizing...");
        solr.optimize();
        solr.close();
    }