Java DirectoryReader.open方法代码示例

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
private void testCase(Query query, String field, int precision, CheckedConsumer<RandomIndexWriter, IOException> buildIndex,
                      Consumer<InternalGeoHashGrid> verify) throws IOException {
    Directory directory = newDirectory();
    RandomIndexWriter indexWriter = new RandomIndexWriter(random(), directory);
    buildIndex.accept(indexWriter);
    indexWriter.close();

    IndexReader indexReader = DirectoryReader.open(directory);
    IndexSearcher indexSearcher = newSearcher(indexReader, true, true);

    GeoGridAggregationBuilder aggregationBuilder = new GeoGridAggregationBuilder("_name").field(field);
    aggregationBuilder.precision(precision);
    MappedFieldType fieldType = new GeoPointFieldMapper.GeoPointFieldType();
    fieldType.setHasDocValues(true);
    fieldType.setName(FIELD_NAME);
    try (Aggregator aggregator = createAggregator(aggregationBuilder, indexSearcher, fieldType)) {
        aggregator.preCollection();
        indexSearcher.search(query, aggregator);
        aggregator.postCollection();
        verify.accept((InternalGeoHashGrid) aggregator.buildAggregation(0L));
    }
    indexReader.close();
    directory.close();
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
/** User runs a query and counts facets. */
private List<FacetResult> search() throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  IndexSearcher searcher = new IndexSearcher(indexReader);
  SortedSetDocValuesReaderState state = new DefaultSortedSetDocValuesReaderState(indexReader);

  // Aggregatses the facet counts
  FacetsCollector fc = new FacetsCollector();

  // MatchAllDocsQuery is for "browsing" (counts facets
  // for all non-deleted docs in the index); normally
  // you'd use a "normal" query:
  FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, fc);

  // Retrieve results
  Facets facets = new SortedSetDocValuesFacetCounts(state, fc);

  List<FacetResult> results = new ArrayList<FacetResult>();
  results.add(facets.getTopChildren(10, "Author"));
  results.add(facets.getTopChildren(10, "Publish Year"));
  indexReader.close();
  
  return results;
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
/**
 * Search all books of a given author. 
 * 
 * @throws Exception never, otherwise the test fails.
 */
@Test
public void findByAuthorSurname() throws Exception {
	IndexSearcher searcher = new IndexSearcher(DirectoryReader.open(directory));
	
	Query query = new QueryParser("author", new StandardAnalyzer()).parse("Gazzarini");
	TopDocs matches = searcher.search(query, 10);
	
	assertEquals(1, matches.totalHits);
			
	final String id = Arrays.stream(matches.scoreDocs)
		.map(scoreDoc -> luceneDoc(scoreDoc.doc, searcher))
		.map(doc -> doc.get("id"))
		.findFirst()
		.get();
	
	assertEquals("1", id);
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
/**
 * without combine script, the "_aggs" map should contain a list of the size of the number of documents matched
 */
@SuppressWarnings("unchecked")
public void testScriptedMetricWithoutCombine() throws IOException {
    try (Directory directory = newDirectory()) {
        int numDocs = randomInt(100);
        try (RandomIndexWriter indexWriter = new RandomIndexWriter(random(), directory)) {
            for (int i = 0; i < numDocs; i++) {
                indexWriter.addDocument(singleton(new SortedNumericDocValuesField("number", i)));
            }
        }
        try (IndexReader indexReader = DirectoryReader.open(directory)) {
            ScriptedMetricAggregationBuilder aggregationBuilder = new ScriptedMetricAggregationBuilder(AGG_NAME);
            aggregationBuilder.initScript(INIT_SCRIPT).mapScript(MAP_SCRIPT);
            ScriptedMetric scriptedMetric = search(newSearcher(indexReader, true, true), new MatchAllDocsQuery(), aggregationBuilder);
            assertEquals(AGG_NAME, scriptedMetric.getName());
            assertNotNull(scriptedMetric.aggregation());
            Map<String, Object> agg = (Map<String, Object>) scriptedMetric.aggregation();
            assertEquals(numDocs, ((List<Integer>) agg.get("collector")).size());
        }
    }
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
public void doTestDocValueRangeQueries(NumberType type, Supplier<Number> valueSupplier) throws Exception {
    Directory dir = newDirectory();
    IndexWriter w = new IndexWriter(dir, newIndexWriterConfig());
    final int numDocs = TestUtil.nextInt(random(), 100, 500);
    for (int i = 0; i < numDocs; ++i) {
        w.addDocument(type.createFields("foo", valueSupplier.get(), true, true, false));
    }
    DirectoryReader reader = DirectoryReader.open(w);
    IndexSearcher searcher = newSearcher(reader);
    w.close();
    final int iters = 10;
    for (int iter = 0; iter < iters; ++iter) {
        Query query = type.rangeQuery("foo",
                random().nextBoolean() ? null : valueSupplier.get(),
                random().nextBoolean() ? null : valueSupplier.get(),
                randomBoolean(), randomBoolean(), true);
        assertThat(query, Matchers.instanceOf(IndexOrDocValuesQuery.class));
        IndexOrDocValuesQuery indexOrDvQuery = (IndexOrDocValuesQuery) query;
        assertEquals(
                searcher.count(indexOrDvQuery.getIndexQuery()),
                searcher.count(indexOrDvQuery.getRandomAccessQuery()));
    }
    reader.close();
    dir.close();
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
/**
 * Search sample. 
 * 
 * @param directory the index directory.
 * @throws IOException in case of I/O failure.
 * @throws ParseException in case of Query parse exception.
 */	
public static void search(Directory directory) throws IOException, ParseException {
	IndexSearcher searcher = new IndexSearcher(DirectoryReader.open(directory));
	
	Query query = new QueryParser("title", new StandardAnalyzer()).parse("title:Solr");
	TopDocs matches = searcher.search(query, 10);
	
	System.out.println("Search returned " + matches.totalHits + " matches.");
	Arrays.stream(matches.scoreDocs)
		.map(scoreDoc -> luceneDoc(scoreDoc, searcher))
		.forEach(doc -> {
			System.out.println("-------------------------------------");				
			System.out.println("ID:\t" + doc.get("id"));
			System.out.println("TITLE:\t" + doc.get("title"));
			System.out.println("AUTHOR:\t" + doc.get("author"));
			System.out.println("SCORE:\t" + doc.get("score"));
			
		});
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
/** User drills down on 'Publish Year/2010'. */
private FacetResult drillDown() throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  IndexSearcher searcher = new IndexSearcher(indexReader);
  SortedSetDocValuesReaderState state = new DefaultSortedSetDocValuesReaderState(indexReader);

  // Now user drills down on Publish Year/2010:
  DrillDownQuery q = new DrillDownQuery(config);
  q.add("Publish Year", "2010");
  FacetsCollector fc = new FacetsCollector();
  FacetsCollector.search(searcher, q, 10, fc);

  // Retrieve results
  Facets facets = new SortedSetDocValuesFacetCounts(state, fc);
  FacetResult result = facets.getTopChildren(10, "Author");
  indexReader.close();
  
  return result;
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
public void testRedisDirectoryWithJedisPool() throws IOException {
    long start = System.currentTimeMillis();
    IndexWriterConfig indexWriterConfig = new IndexWriterConfig(new WhitespaceAnalyzer()).setOpenMode(IndexWriterConfig
            .OpenMode.CREATE);
    //indexWriterConfig.setInfoStream(System.out);
    //indexWriterConfig.setRAMBufferSizeMB(2048);
    //LogByteSizeMergePolicy logByteSizeMergePolicy = new LogByteSizeMergePolicy();
    //logByteSizeMergePolicy.setMinMergeMB(1);
    //logByteSizeMergePolicy.setMaxMergeMB(64);
    //logByteSizeMergePolicy.setMaxCFSSegmentSizeMB(64);
    //indexWriterConfig.setRAMBufferSizeMB(1024).setMergePolicy(logByteSizeMergePolicy).setUseCompoundFile(false);
    //GenericObjectPoolConfig genericObjectPoolConfig = new GenericObjectPoolConfig();
    //获取连接等待时间
    //genericObjectPoolConfig.setMaxWaitMillis(3000);
    //10s超时时间
    JedisPool jedisPool = new JedisPool(new JedisPoolConfig(), "localhost", 6379, Constants.TIME_OUT);
    RedisDirectory redisDirectory = new RedisDirectory(new JedisPoolStream(jedisPool));
    IndexWriter indexWriter = new IndexWriter(redisDirectory, indexWriterConfig);
    for (int i = 0; i < 10000000; i++) {
        indexWriter.addDocument(addDocument(i));
    }
    indexWriter.commit();
    indexWriter.close();
    redisDirectory.close();
    long end = System.currentTimeMillis();
    log.error("RedisDirectoryWithJedisPool consumes {}s!", (end - start) / 1000);
    start = System.currentTimeMillis();
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(new RedisDirectory(new JedisStream("localhost",
            6379))));
    int total = 0;
    for (int i = 0; i < 10000000; i++) {
        TermQuery key1 = new TermQuery(new Term("key1", "key" + i));
        TopDocs search = indexSearcher.search(key1, 10);
        total += search.totalHits;
    }
    System.out.println(total);
    end = System.currentTimeMillis();
    log.error("RedisDirectoryWithJedisPool search consumes {}ms!", (end - start));
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
@Override
public void indexingDone() {
	try {
		spellChecker = new DirectSpellChecker();
		spellChecker.setMaxEdits(2);
		spellChecker.setAccuracy(0.1f);
		spellChecker.setMinPrefix(0);
		reader = DirectoryReader.open(writer);

		fuzzySuggester = new FuzzySuggester(directory, "", writer.getAnalyzer());
		Dictionary dict = new DocumentValueSourceDictionary(reader, WORD_FIELD, new LongValuesSource() {
			
			@Override
			public boolean needsScores() {
				return false;
			}
			
			@Override
			public LongValues getValues(LeafReaderContext ctx, DoubleValues scores) throws IOException {
				return null;
			}
		});
		fuzzySuggester.build(dict);
		
		writer.close();
		searcher = new IndexSearcher(DirectoryReader.open(directory));
	} catch (IOException e) {
		throw new RuntimeException(e);
	}
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
/** User runs a query and counts facets. */
private List<FacetResult> search() throws IOException {
  DirectoryReader indexReader = DirectoryReader.open(indexDir);
  IndexSearcher searcher = new IndexSearcher(indexReader);
  TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);

  FacetsCollector fc = new FacetsCollector();

  // MatchAllDocsQuery is for "browsing" (counts facets
  // for all non-deleted docs in the index); normally
  // you'd use a "normal" query:
  FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, fc);

  // Retrieve results
  List<FacetResult> results = new ArrayList<FacetResult>();

  // Count both "Publish Date" and "Author" dimensions
  Facets author = new FastTaxonomyFacetCounts("author", taxoReader, config, fc);
  results.add(author.getTopChildren(10, "Author"));

  Facets pubDate = new FastTaxonomyFacetCounts("pubdate", taxoReader, config, fc);
  results.add(pubDate.getTopChildren(10, "Publish Date"));
  
  indexReader.close();
  taxoReader.close();
  
  return results;
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
private boolean openIndex(String foldername, String folderpath) {
    try {
        Analyzer analyzer = new StandardAnalyzer();

        //diskConfig = new IndexWriterConfig(analyzer);
        //diskConfig.setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND);
        //diskConfig.setRAMBufferSizeMB(256.0);
        //init director

        FSDirectory diskDirectory = FSDirectory.open(Paths.get(folderpath));

        //init reader
        DirectoryReader diskReader = DirectoryReader.open(diskDirectory);

        //every thing is ok
        if (analyzer != null && diskDirectory != null && diskReader != null) {
            analyzerList.put(foldername, analyzer);
            directorList.put(foldername, diskDirectory);
            readerList.put(foldername, diskReader);

            log.info("Load Index Success:" + foldername + " path:" + folderpath);
            return true;
        } else {
            log.error("Load Index Fail:" + foldername);
            return false;
        }

    } catch (org.apache.lucene.index.IndexNotFoundException xe) {
        log.error("Load Index Not Found:" + foldername);
        //throw new Exception(e.getMessage());
        //e.printStackTrace();
    } catch (Exception xxe) {
        //do nothing
        xxe.printStackTrace();
        log.error("load index Exception:" + xxe.getMessage());
        //throw new Exception(e.getMessage());
    }
    return false;
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
private void testCase(Query query,
                      CheckedConsumer<RandomIndexWriter, IOException> indexer,
                      Consumer<Sum> verify) throws IOException {

    try (Directory directory = newDirectory()) {
        try (RandomIndexWriter indexWriter = new RandomIndexWriter(random(), directory)) {
            indexer.accept(indexWriter);
        }

        try (IndexReader indexReader = DirectoryReader.open(directory)) {
            IndexSearcher indexSearcher = newSearcher(indexReader, true, true);

            MappedFieldType fieldType = new NumberFieldMapper.NumberFieldType(NumberFieldMapper.NumberType.LONG);
            fieldType.setName(FIELD_NAME);
            fieldType.setHasDocValues(true);

            SumAggregationBuilder aggregationBuilder = new SumAggregationBuilder("_name");
            aggregationBuilder.field(FIELD_NAME);

            try (SumAggregator aggregator = createAggregator(aggregationBuilder, indexSearcher, fieldType)) {
                aggregator.preCollection();
                indexSearcher.search(query, aggregator);
                aggregator.postCollection();

                verify.accept((Sum) aggregator.buildAggregation(0L));
            }
        }
    }
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
public static IndexSearcher getSearcher() throws IOException {
    Directory directory = FSDirectory.open(Paths.get(CommonConstants.INDEX_FILE_PATH));
    IndexReader reader = DirectoryReader.open(directory);
    return new IndexSearcher(reader);
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
/**
 * Searches gene IDs, affected by variations in specified VCF files in a specified project
 *
 * @param projectId an ID of a project to search genes
 * @param gene a prefix of a gene ID to search
 * @param vcfFileIds a {@code List} of IDs of VCF files in project to search for gene IDs
 * @return a {@code Set} of gene IDs, that are affected by some variations in specified VCf files
 * @throws IOException
 */
public Set<String> searchGenesInVcfFilesInProject(long projectId, String gene, List<Long> vcfFileIds) throws
        IOException {
    if (vcfFileIds == null || vcfFileIds.isEmpty()) {
        return Collections.emptySet();
    }

    BooleanQuery.Builder builder = new BooleanQuery.Builder();

    PrefixQuery geneIdPrefixQuery = new PrefixQuery(new Term(FeatureIndexFields.GENE_ID.getFieldName(),
            gene.toLowerCase()));
    PrefixQuery geneNamePrefixQuery = new PrefixQuery(new Term(FeatureIndexFields.GENE_NAME.getFieldName(),
                                                               gene.toLowerCase()));
    BooleanQuery.Builder geneIdOrNameQuery = new BooleanQuery.Builder();
    geneIdOrNameQuery.add(geneIdPrefixQuery, BooleanClause.Occur.SHOULD);
    geneIdOrNameQuery.add(geneNamePrefixQuery, BooleanClause.Occur.SHOULD);

    builder.add(geneIdOrNameQuery.build(), BooleanClause.Occur.MUST);

    List<Term> terms = vcfFileIds.stream()
            .map(vcfFileId -> new Term(FeatureIndexFields.FILE_ID.getFieldName(), vcfFileId.toString()))
            .collect(Collectors.toList());
    TermsQuery termsQuery = new TermsQuery(terms);
    builder.add(termsQuery, BooleanClause.Occur.MUST);
    BooleanQuery query =  builder.build();

    Set<String> geneIds;

    try (
        Directory index = fileManager.getIndexForProject(projectId);
        IndexReader reader = DirectoryReader.open(index)
    ) {
        if (reader.numDocs() == 0) {
            return Collections.emptySet();
        }

        IndexSearcher searcher = new IndexSearcher(reader);
        final TopDocs docs = searcher.search(query, reader.numDocs());
        final ScoreDoc[] hits = docs.scoreDocs;

        geneIds = fetchGeneIds(hits, searcher);
    } catch (IOException e) {
        LOGGER.error(MessageHelper.getMessage(MessagesConstants.ERROR_FEATURE_INDEX_SEARCH_FAILED), e);
        return Collections.emptySet();
    }

    return geneIds;
}
 

import org.apache.lucene.index.DirectoryReader; //导入方法依赖的package包/类
public void testConcurrentWritesAndCommits() throws Exception {
    try (Store store = createStore();
         InternalEngine engine = new InternalEngine(config(defaultSettings, store, createTempDir(), newMergePolicy(),
                                                                 new SnapshotDeletionPolicy(NoDeletionPolicy.INSTANCE),
                                                                 IndexRequest.UNSET_AUTO_GENERATED_TIMESTAMP, null))) {

        final int numIndexingThreads = scaledRandomIntBetween(3, 6);
        final int numDocsPerThread = randomIntBetween(500, 1000);
        final CyclicBarrier barrier = new CyclicBarrier(numIndexingThreads + 1);
        final List<Thread> indexingThreads = new ArrayList<>();
        // create N indexing threads to index documents simultaneously
        for (int threadNum = 0; threadNum < numIndexingThreads; threadNum++) {
            final int threadIdx = threadNum;
            Thread indexingThread = new Thread(() -> {
                try {
                    barrier.await(); // wait for all threads to start at the same time
                    // index random number of docs
                    for (int i = 0; i < numDocsPerThread; i++) {
                        final String id = "thread" + threadIdx + "#" + i;
                        ParsedDocument doc = testParsedDocument(id, "test", null, testDocument(), B_1, null);
                        engine.index(indexForDoc(doc));
                    }
                } catch (Exception e) {
                    throw new RuntimeException(e);
                }
            });
            indexingThreads.add(indexingThread);
        }

        // start the indexing threads
        for (Thread thread : indexingThreads) {
            thread.start();
        }
        barrier.await(); // wait for indexing threads to all be ready to start

        // create random commit points
        boolean doneIndexing;
        do {
            doneIndexing = indexingThreads.stream().filter(Thread::isAlive).count() == 0;
            //engine.flush(); // flush and commit
        } while (doneIndexing == false);

        // now, verify all the commits have the correct docs according to the user commit data
        long prevLocalCheckpoint = SequenceNumbersService.NO_OPS_PERFORMED;
        long prevMaxSeqNo = SequenceNumbersService.NO_OPS_PERFORMED;
        for (IndexCommit commit : DirectoryReader.listCommits(store.directory())) {
            Map<String, String> userData = commit.getUserData();
            long localCheckpoint = userData.containsKey(SequenceNumbers.LOCAL_CHECKPOINT_KEY) ?
                                       Long.parseLong(userData.get(SequenceNumbers.LOCAL_CHECKPOINT_KEY)) :
                                       SequenceNumbersService.NO_OPS_PERFORMED;
            long maxSeqNo = userData.containsKey(SequenceNumbers.MAX_SEQ_NO) ?
                                Long.parseLong(userData.get(SequenceNumbers.MAX_SEQ_NO)) :
                                SequenceNumbersService.UNASSIGNED_SEQ_NO;
            // local checkpoint and max seq no shouldn't go backwards
            assertThat(localCheckpoint, greaterThanOrEqualTo(prevLocalCheckpoint));
            assertThat(maxSeqNo, greaterThanOrEqualTo(prevMaxSeqNo));
            try (IndexReader reader = DirectoryReader.open(commit)) {
                FieldStats stats = SeqNoFieldMapper.SeqNoDefaults.FIELD_TYPE.stats(reader);
                final long highestSeqNo;
                if (stats != null) {
                    highestSeqNo = (long) stats.getMaxValue();
                } else {
                    highestSeqNo = SequenceNumbersService.NO_OPS_PERFORMED;
                }
                // make sure localCheckpoint <= highest seq no found <= maxSeqNo
                assertThat(highestSeqNo, greaterThanOrEqualTo(localCheckpoint));
                assertThat(highestSeqNo, lessThanOrEqualTo(maxSeqNo));
                // make sure all sequence numbers up to and including the local checkpoint are in the index
                FixedBitSet seqNosBitSet = getSeqNosSet(reader, highestSeqNo);
                for (int i = 0; i <= localCheckpoint; i++) {
                    assertTrue("local checkpoint [" + localCheckpoint + "], _seq_no [" + i + "] should be indexed",
                               seqNosBitSet.get(i));
                }
            }
            prevLocalCheckpoint = localCheckpoint;
            prevMaxSeqNo = maxSeqNo;
        }
    }
}