Java IndexWriterConfig类代码示例

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testRamDirectory() throws IOException {
    long start = System.currentTimeMillis();
    IndexWriterConfig indexWriterConfig = new IndexWriterConfig(new WhitespaceAnalyzer()).setOpenMode(IndexWriterConfig
            .OpenMode.CREATE);
    RAMDirectory ramDirectory = new RAMDirectory();
    IndexWriter indexWriter = new IndexWriter(ramDirectory, indexWriterConfig);
    for (int i = 0; i < 10000000; i++) {
        indexWriter.addDocument(addDocument(i));
    }
    indexWriter.commit();
    indexWriter.close();
    long end = System.currentTimeMillis();
    log.error("RamDirectory consumes {}s!", (end - start) / 1000);
    start = System.currentTimeMillis();
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(ramDirectory));
    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("RamDirectory search consumes {}ms!", (end - start));
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testMMapDirectory() throws IOException {
    long start = System.currentTimeMillis();
    IndexWriterConfig indexWriterConfig = new IndexWriterConfig(new WhitespaceAnalyzer()).setOpenMode(IndexWriterConfig
            .OpenMode.CREATE);
    FSDirectory open = FSDirectory.open(Paths.get("E:/testlucene"));
    IndexWriter indexWriter = new IndexWriter(open, indexWriterConfig);
    for (int i = 0; i < 10000000; i++) {
        indexWriter.addDocument(addDocument(i));
    }
    indexWriter.commit();
    indexWriter.close();
    long end = System.currentTimeMillis();
    log.error("MMapDirectory consumes {}s!", (end - start) / 1000);
    start = System.currentTimeMillis();
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(open));
    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("MMapDirectory search consumes {}ms!", (end - start));
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testSimpleNumericOps() throws Exception {
    Directory dir = new RAMDirectory();
    IndexWriter indexWriter = new IndexWriter(dir, new IndexWriterConfig(Lucene.STANDARD_ANALYZER));

    Document document = new Document();
    document.add(new TextField("_id", "1", Field.Store.YES));
    document.add(new LegacyIntField("test", 2, LegacyIntField.TYPE_STORED));
    indexWriter.addDocument(document);

    IndexReader reader = DirectoryReader.open(indexWriter);
    IndexSearcher searcher = new IndexSearcher(reader);
    TopDocs topDocs = searcher.search(new TermQuery(new Term("_id", "1")), 1);
    Document doc = searcher.doc(topDocs.scoreDocs[0].doc);
    IndexableField f = doc.getField("test");
    assertThat(f.stringValue(), equalTo("2"));

    BytesRefBuilder bytes = new BytesRefBuilder();
    LegacyNumericUtils.intToPrefixCoded(2, 0, bytes);
    topDocs = searcher.search(new TermQuery(new Term("test", bytes.get())), 1);
    doc = searcher.doc(topDocs.scoreDocs[0].doc);
    f = doc.getField("test");
    assertThat(f.stringValue(), equalTo("2"));

    indexWriter.close();
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
private IndexWriterConfig getIndexWriterConfig(boolean create) {
    final IndexWriterConfig iwc = new IndexWriterConfig(engineConfig.getAnalyzer());
    iwc.setCommitOnClose(false); // we by default don't commit on close
    iwc.setOpenMode(create ? IndexWriterConfig.OpenMode.CREATE : IndexWriterConfig.OpenMode.APPEND);
    iwc.setIndexDeletionPolicy(deletionPolicy);
    // with tests.verbose, lucene sets this up: plumb to align with filesystem stream
    boolean verbose = false;
    try {
        verbose = Boolean.parseBoolean(System.getProperty("tests.verbose"));
    } catch (Exception ignore) {
    }
    iwc.setInfoStream(verbose ? InfoStream.getDefault() : new LoggerInfoStream(logger));
    iwc.setMergeScheduler(mergeScheduler);
    MergePolicy mergePolicy = config().getMergePolicy();
    // Give us the opportunity to upgrade old segments while performing
    // background merges
    mergePolicy = new ElasticsearchMergePolicy(mergePolicy);
    iwc.setMergePolicy(mergePolicy);
    iwc.setSimilarity(engineConfig.getSimilarity());
    iwc.setRAMBufferSizeMB(engineConfig.getIndexingBufferSize().getMbFrac());
    iwc.setCodec(engineConfig.getCodec());
    iwc.setUseCompoundFile(true); // always use compound on flush - reduces # of file-handles on refresh
    return iwc;
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
/**
 * This method removes all lucene files from the given directory. It will first try to delete all commit points / segments
 * files to ensure broken commits or corrupted indices will not be opened in the future. If any of the segment files can't be deleted
 * this operation fails.
 */
public static void cleanLuceneIndex(Directory directory) throws IOException {
    try (Lock writeLock = directory.obtainLock(IndexWriter.WRITE_LOCK_NAME)) {
        for (final String file : directory.listAll()) {
            if (file.startsWith(IndexFileNames.SEGMENTS) || file.equals(IndexFileNames.OLD_SEGMENTS_GEN)) {
                directory.deleteFile(file); // remove all segment_N files
            }
        }
    }
    try (IndexWriter writer = new IndexWriter(directory, new IndexWriterConfig(Lucene.STANDARD_ANALYZER)
            .setMergePolicy(NoMergePolicy.INSTANCE) // no merges
            .setCommitOnClose(false) // no commits
            .setOpenMode(IndexWriterConfig.OpenMode.CREATE))) // force creation - don't append...
    {
        // do nothing and close this will kick of IndexFileDeleter which will remove all pending files
    }
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
private IndexReader indexOneDoc(Directory dir, String field, String value, Analyzer analyzer) throws IOException {
    IndexWriterConfig iwc = newIndexWriterConfig(analyzer);
    iwc.setMergePolicy(newLogMergePolicy());
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, iwc);

    FieldType ft = new FieldType(TextField.TYPE_STORED);
    ft.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS);
    Field textField = new Field(field, "", ft);
    Document doc = new Document();
    doc.add(textField);

    textField.setStringValue(value);
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    return ir;
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
/**
 * Test the WordScorer emitted by the smoothing model
 */
public void testBuildWordScorer() throws IOException {
    SmoothingModel testModel = createTestModel();
    Map<String, Analyzer> mapping = new HashMap<>();
    mapping.put("field", new WhitespaceAnalyzer());
    PerFieldAnalyzerWrapper wrapper = new PerFieldAnalyzerWrapper(new WhitespaceAnalyzer(), mapping);
    IndexWriter writer = new IndexWriter(new RAMDirectory(), new IndexWriterConfig(wrapper));
    Document doc = new Document();
    doc.add(new Field("field", "someText", TextField.TYPE_NOT_STORED));
    writer.addDocument(doc);
    DirectoryReader ir = DirectoryReader.open(writer);

    WordScorer wordScorer = testModel.buildWordScorerFactory().newScorer(ir, MultiFields.getTerms(ir, "field"), "field", 0.9d,
            BytesRefs.toBytesRef(" "));
    assertWordScorer(wordScorer, testModel);
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testSingleValued() throws IOException {
    Directory dir = newDirectory();
    // we need the default codec to check for singletons
    IndexWriter w = new IndexWriter(dir, new IndexWriterConfig(null).setCodec(TestUtil.getDefaultCodec()));
    Document doc = new Document();
    for (IndexableField f : NumberFieldMapper.NumberType.HALF_FLOAT.createFields("half_float", 3f, false, true, false)) {
        doc.add(f);
    }
    w.addDocument(doc);
    final DirectoryReader dirReader = DirectoryReader.open(w);
    LeafReader reader = getOnlyLeafReader(dirReader);
    SortedNumericDoubleValues values = new SortedNumericDVIndexFieldData.SortedNumericHalfFloatFieldData(
            reader, "half_float").getDoubleValues();
    assertNotNull(FieldData.unwrapSingleton(values));
    values.setDocument(0);
    assertEquals(1, values.count());
    assertEquals(3f, values.valueAt(0), 0f);
    IOUtils.close(dirReader, w, dir);
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public static InternalEngine createInternalEngine(@Nullable final IndexWriterFactory indexWriterFactory,
                                                  @Nullable final Supplier<SequenceNumbersService> sequenceNumbersServiceSupplier,
                                                  final EngineConfig config) {
    return new InternalEngine(config) {
            @Override
            IndexWriter createWriter(Directory directory, IndexWriterConfig iwc) throws IOException {
                return (indexWriterFactory != null) ?
                    indexWriterFactory.createWriter(directory, iwc) :
                    super.createWriter(directory, iwc);
            }

            @Override
            public SequenceNumbersService seqNoService() {
                return (sequenceNumbersServiceSupplier != null) ? sequenceNumbersServiceSupplier.get() : super.seqNoService();
            }
        };
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
private void assertCompressionEquals(Mode expected, Codec actual) throws Exception {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig(null);
    iwc.setCodec(actual);
    IndexWriter iw = new IndexWriter(dir, iwc);
    iw.addDocument(new Document());
    iw.commit();
    iw.close();
    DirectoryReader ir = DirectoryReader.open(dir);
    SegmentReader sr = (SegmentReader) ir.leaves().get(0).reader();
    String v = sr.getSegmentInfo().info.getAttribute(Lucene50StoredFieldsFormat.MODE_KEY);
    assertNotNull(v);
    assertEquals(expected, Mode.valueOf(v));
    ir.close();
    dir.close();
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testCanOpenIndex() throws IOException {
    final ShardId shardId = new ShardId("index", "_na_", 1);
    IndexWriterConfig iwc = newIndexWriterConfig();
    Path tempDir = createTempDir();
    final BaseDirectoryWrapper dir = newFSDirectory(tempDir);
    assertFalse(Store.canOpenIndex(logger, tempDir, shardId, (id, l) -> new DummyShardLock(id)));
    IndexWriter writer = new IndexWriter(dir, iwc);
    Document doc = new Document();
    doc.add(new StringField("id", "1", random().nextBoolean() ? Field.Store.YES : Field.Store.NO));
    writer.addDocument(doc);
    writer.commit();
    writer.close();
    assertTrue(Store.canOpenIndex(logger, tempDir, shardId, (id, l) -> new DummyShardLock(id)));

    DirectoryService directoryService = new DirectoryService(shardId, INDEX_SETTINGS) {

        @Override
        public Directory newDirectory() throws IOException {
            return dir;
        }
    };
    Store store = new Store(shardId, INDEX_SETTINGS, directoryService, new DummyShardLock(shardId));
    store.markStoreCorrupted(new CorruptIndexException("foo", "bar"));
    assertFalse(Store.canOpenIndex(logger, tempDir, shardId, (id, l) -> new DummyShardLock(id)));
    store.close();
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testVectorHighlighter() throws Exception {
    Directory dir = new RAMDirectory();
    IndexWriter indexWriter = new IndexWriter(dir, new IndexWriterConfig(Lucene.STANDARD_ANALYZER));

    Document document = new Document();
    document.add(new TextField("_id", "1", Field.Store.YES));
    FieldType vectorsType = new FieldType(TextField.TYPE_STORED);
    vectorsType.setStoreTermVectors(true);
    vectorsType.setStoreTermVectorPositions(true);
    vectorsType.setStoreTermVectorOffsets(true);
    document.add(new Field("content", "the big bad dog", vectorsType));
    indexWriter.addDocument(document);

    IndexReader reader = DirectoryReader.open(indexWriter);
    IndexSearcher searcher = new IndexSearcher(reader);
    TopDocs topDocs = searcher.search(new TermQuery(new Term("_id", "1")), 1);

    assertThat(topDocs.totalHits, equalTo(1));

    FastVectorHighlighter highlighter = new FastVectorHighlighter();
    String fragment = highlighter.getBestFragment(highlighter.getFieldQuery(new TermQuery(new Term("content", "bad"))),
            reader, topDocs.scoreDocs[0].doc, "content", 30);
    assertThat(fragment, notNullValue());
    assertThat(fragment, equalTo("the big <b>bad</b> dog"));
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testVectorHighlighterNoStore() throws Exception {
    Directory dir = new RAMDirectory();
    IndexWriter indexWriter = new IndexWriter(dir, new IndexWriterConfig(Lucene.STANDARD_ANALYZER));

    Document document = new Document();
    document.add(new TextField("_id", "1", Field.Store.YES));
    FieldType vectorsType = new FieldType(TextField.TYPE_NOT_STORED);
    vectorsType.setStoreTermVectors(true);
    vectorsType.setStoreTermVectorPositions(true);
    vectorsType.setStoreTermVectorOffsets(true);
    document.add(new Field("content", "the big bad dog", vectorsType));
    indexWriter.addDocument(document);

    IndexReader reader = DirectoryReader.open(indexWriter);
    IndexSearcher searcher = new IndexSearcher(reader);
    TopDocs topDocs = searcher.search(new TermQuery(new Term("_id", "1")), 1);

    assertThat(topDocs.totalHits, equalTo(1));

    FastVectorHighlighter highlighter = new FastVectorHighlighter();
    String fragment = highlighter.getBestFragment(highlighter.getFieldQuery(new TermQuery(new Term("content", "bad"))),
            reader, topDocs.scoreDocs[0].doc, "content", 30);
    assertThat(fragment, nullValue());
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testVectorHighlighterNoTermVector() throws Exception {
    Directory dir = new RAMDirectory();
    IndexWriter indexWriter = new IndexWriter(dir, new IndexWriterConfig(Lucene.STANDARD_ANALYZER));

    Document document = new Document();
    document.add(new TextField("_id", "1", Field.Store.YES));
    document.add(new TextField("content", "the big bad dog", Field.Store.YES));
    indexWriter.addDocument(document);

    IndexReader reader = DirectoryReader.open(indexWriter);
    IndexSearcher searcher = new IndexSearcher(reader);
    TopDocs topDocs = searcher.search(new TermQuery(new Term("_id", "1")), 1);

    assertThat(topDocs.totalHits, equalTo(1));

    FastVectorHighlighter highlighter = new FastVectorHighlighter();
    String fragment = highlighter.getBestFragment(highlighter.getFieldQuery(new TermQuery(new Term("content", "bad"))),
            reader, topDocs.scoreDocs[0].doc, "content", 30);
    assertThat(fragment, nullValue());
}
 

import org.apache.lucene.index.IndexWriterConfig; //导入依赖的package包/类
public void testSortValues() throws Exception {
    Directory dir = new RAMDirectory();
    IndexWriter indexWriter = new IndexWriter(dir, new IndexWriterConfig(Lucene.STANDARD_ANALYZER));
    for (int i = 0; i < 10; i++) {
        Document document = new Document();
        String text = new String(new char[]{(char) (97 + i), (char) (97 + i)});
        document.add(new TextField("str", text, Field.Store.YES));
        document.add(new SortedDocValuesField("str", new BytesRef(text)));
        indexWriter.addDocument(document);
    }
    IndexReader reader = SlowCompositeReaderWrapper.wrap(DirectoryReader.open(indexWriter));
    IndexSearcher searcher = new IndexSearcher(reader);
    TopFieldDocs docs = searcher.search(new MatchAllDocsQuery(), 10, new Sort(new SortField("str", SortField.Type.STRING)));
    for (int i = 0; i < 10; i++) {
        FieldDoc fieldDoc = (FieldDoc) docs.scoreDocs[i];
        assertThat((BytesRef) fieldDoc.fields[0], equalTo(new BytesRef(new String(new char[]{(char) (97 + i), (char) (97 + i)}))));
    }
}