Examples of org.apache.lucene.store.Directory

org.apache.lucene.store.Directory
A Directory is a flat list of files. Files may be written once, when they are created. Once a file is created it may only be opened for read, or deleted. Random access is permitted both when reading and writing.
Java's i/o APIs not used directly, but rather all i/o is through this API. This permits things such as:
- implementation of RAM-based indices;
- implementation indices stored in a database, via JDBC;
- implementation of an index as a single file;
Directory locking is implemented by an instance of {@link LockFactory}, and can be changed for each Directory instance using {@link #setLockFactory}.

    
    //读取本类目录下的text.txt文件
    String content = ContentReader.readText(Chinese.class);


    //接下来是标准的Lucene建立索引和检索的代码
    Directory ramDir = new RAMDirectory();
    IndexWriter writer = new IndexWriter(ramDir, analyzer, MaxFieldLength.UNLIMITED);
    Document doc = new Document();
    Field fd = new Field(FIELD_NAME, content, Field.Store.YES,
        Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS);
    doc.add(fd);

View Full Code Here

    
    //读取本类目录下的text.txt文件
    String content = ContentReader.readText(Chinese.class);


    //接下来是标准的Lucene建立索引和检索的代码
    Directory ramDir = new RAMDirectory();
    IndexWriter writer = new IndexWriter(ramDir, analyzer, MaxFieldLength.UNLIMITED);
    Document doc = new Document();
    Field fd = new Field(FIELD_NAME, content, Field.Store.YES,
        Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS);
    doc.add(fd);

View Full Code Here

import org.apache.lucene.util._TestUtil;


public class TestIndexReader extends LuceneTestCase {
    
    public void testCommitUserData() throws Exception {
      Directory d = newDirectory();


      Map<String,String> commitUserData = new HashMap<String,String>();
      commitUserData.put("foo", "fighters");
      
      // set up writer
      IndexWriter writer = new IndexWriter(d, newIndexWriterConfig(
          TEST_VERSION_CURRENT, new MockAnalyzer(random))
      .setMaxBufferedDocs(2));
      for(int i=0;i<27;i++)
        addDocumentWithFields(writer);
      writer.close();
      
      IndexReader r = IndexReader.open(d, false);
      r.deleteDocument(5);
      r.flush(commitUserData);
      r.close();
      
      SegmentInfos sis = new SegmentInfos();
      sis.read(d);
      IndexReader r2 = IndexReader.open(d, false);
      IndexCommit c = r.getIndexCommit();
      assertEquals(c.getUserData(), commitUserData);


      assertEquals(sis.getCurrentSegmentFileName(), c.getSegmentsFileName());


      assertTrue(c.equals(r.getIndexCommit()));


      // Change the index
      writer = new IndexWriter(d, newIndexWriterConfig(TEST_VERSION_CURRENT,
          new MockAnalyzer(random)).setOpenMode(
              OpenMode.APPEND).setMaxBufferedDocs(2));
      for(int i=0;i<7;i++)
        addDocumentWithFields(writer);
      writer.close();


      IndexReader r3 = r2.reopen();
      assertFalse(c.equals(r3.getIndexCommit()));
      assertFalse(r2.getIndexCommit().isOptimized());
      r3.close();


      writer = new IndexWriter(d, newIndexWriterConfig(TEST_VERSION_CURRENT,
        new MockAnalyzer(random))
        .setOpenMode(OpenMode.APPEND));
      writer.optimize();
      writer.close();


      r3 = r2.reopen();
      assertTrue(r3.getIndexCommit().isOptimized());
      r2.close();
      r3.close();
      d.close();
    }

View Full Code Here

      r3.close();
      d.close();
    }
    
    public void testIsCurrent() throws Exception {
      Directory d = newDirectory();
      IndexWriter writer = new IndexWriter(d, newIndexWriterConfig( 
        TEST_VERSION_CURRENT, new MockAnalyzer(random)));
      addDocumentWithFields(writer);
      writer.close();
      // set up reader:
      IndexReader reader = IndexReader.open(d, false);
      assertTrue(reader.isCurrent());
      // modify index by adding another document:
      writer = new IndexWriter(d, newIndexWriterConfig(TEST_VERSION_CURRENT,
          new MockAnalyzer(random)).setOpenMode(OpenMode.APPEND));
      addDocumentWithFields(writer);
      writer.close();
      assertFalse(reader.isCurrent());
      // re-create index:
      writer = new IndexWriter(d, newIndexWriterConfig(TEST_VERSION_CURRENT,
          new MockAnalyzer(random)).setOpenMode(OpenMode.CREATE));
      addDocumentWithFields(writer);
      writer.close();
      assertFalse(reader.isCurrent());
      reader.close();
      d.close();
    }

View Full Code Here

    /**
     * Tests the IndexReader.getFieldNames implementation
     * @throws Exception on error
     */
    public void testGetFieldNames() throws Exception {
        Directory d = newDirectory();
        // set up writer
        IndexWriter writer = new IndexWriter(
            d,
            newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random))
        );


        Document doc = new Document();
        doc.add(new Field("keyword","test1", Field.Store.YES, Field.Index.NOT_ANALYZED));
        doc.add(new Field("text","test1", Field.Store.YES, Field.Index.ANALYZED));
        doc.add(new Field("unindexed","test1", Field.Store.YES, Field.Index.NO));
        doc.add(new Field("unstored","test1", Field.Store.NO, Field.Index.ANALYZED));
        writer.addDocument(doc);


        writer.close();
        // set up reader
        IndexReader reader = IndexReader.open(d, false);
        Collection<String> fieldNames = reader.getFieldNames(IndexReader.FieldOption.ALL);
        assertTrue(fieldNames.contains("keyword"));
        assertTrue(fieldNames.contains("text"));
        assertTrue(fieldNames.contains("unindexed"));
        assertTrue(fieldNames.contains("unstored"));
        reader.close();
        // add more documents
        writer = new IndexWriter(
            d,
            newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)).
                setOpenMode(OpenMode.APPEND).
                setMergePolicy(newLogMergePolicy())
        );
        // want to get some more segments here
        int mergeFactor = ((LogMergePolicy) writer.getConfig().getMergePolicy()).getMergeFactor();
        for (int i = 0; i < 5*mergeFactor; i++) {
          doc = new Document();
          doc.add(new Field("keyword","test1", Field.Store.YES, Field.Index.NOT_ANALYZED));
          doc.add(new Field("text","test1", Field.Store.YES, Field.Index.ANALYZED));
          doc.add(new Field("unindexed","test1", Field.Store.YES, Field.Index.NO));
          doc.add(new Field("unstored","test1", Field.Store.NO, Field.Index.ANALYZED));
          writer.addDocument(doc);
        }
        // new fields are in some different segments (we hope)
        for (int i = 0; i < 5*mergeFactor; i++) {
          doc = new Document();
          doc.add(new Field("keyword2","test1", Field.Store.YES, Field.Index.NOT_ANALYZED));
          doc.add(new Field("text2","test1", Field.Store.YES, Field.Index.ANALYZED));
          doc.add(new Field("unindexed2","test1", Field.Store.YES, Field.Index.NO));
          doc.add(new Field("unstored2","test1", Field.Store.NO, Field.Index.ANALYZED));
          writer.addDocument(doc);
        }
        // new termvector fields
        for (int i = 0; i < 5*mergeFactor; i++) {
          doc = new Document();
          doc.add(new Field("tvnot","tvnot", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.NO));
          doc.add(new Field("termvector","termvector", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.YES));
          doc.add(new Field("tvoffset","tvoffset", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_OFFSETS));
          doc.add(new Field("tvposition","tvposition", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS));
          doc.add(newField("tvpositionoffset","tvpositionoffset", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS));
          writer.addDocument(doc);
        }
        
        writer.close();
        // verify fields again
        reader = IndexReader.open(d, false);
        fieldNames = reader.getFieldNames(IndexReader.FieldOption.ALL);
        assertEquals(13, fieldNames.size());    // the following fields
        assertTrue(fieldNames.contains("keyword"));
        assertTrue(fieldNames.contains("text"));
        assertTrue(fieldNames.contains("unindexed"));
        assertTrue(fieldNames.contains("unstored"));
        assertTrue(fieldNames.contains("keyword2"));
        assertTrue(fieldNames.contains("text2"));
        assertTrue(fieldNames.contains("unindexed2"));
        assertTrue(fieldNames.contains("unstored2"));
        assertTrue(fieldNames.contains("tvnot"));
        assertTrue(fieldNames.contains("termvector"));
        assertTrue(fieldNames.contains("tvposition"));
        assertTrue(fieldNames.contains("tvoffset"));
        assertTrue(fieldNames.contains("tvpositionoffset"));
        
        // verify that only indexed fields were returned
        fieldNames = reader.getFieldNames(IndexReader.FieldOption.INDEXED);
        assertEquals(11, fieldNames.size());    // 6 original + the 5 termvector fields 
        assertTrue(fieldNames.contains("keyword"));
        assertTrue(fieldNames.contains("text"));
        assertTrue(fieldNames.contains("unstored"));
        assertTrue(fieldNames.contains("keyword2"));
        assertTrue(fieldNames.contains("text2"));
        assertTrue(fieldNames.contains("unstored2"));
        assertTrue(fieldNames.contains("tvnot"));
        assertTrue(fieldNames.contains("termvector"));
        assertTrue(fieldNames.contains("tvposition"));
        assertTrue(fieldNames.contains("tvoffset"));
        assertTrue(fieldNames.contains("tvpositionoffset"));
        
        // verify that only unindexed fields were returned
        fieldNames = reader.getFieldNames(IndexReader.FieldOption.UNINDEXED);
        assertEquals(2, fieldNames.size());    // the following fields
        assertTrue(fieldNames.contains("unindexed"));
        assertTrue(fieldNames.contains("unindexed2"));
                
        // verify index term vector fields  
        fieldNames = reader.getFieldNames(IndexReader.FieldOption.TERMVECTOR);
        assertEquals(1, fieldNames.size());    // 1 field has term vector only
        assertTrue(fieldNames.contains("termvector"));
        
        fieldNames = reader.getFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION);
        assertEquals(1, fieldNames.size());    // 4 fields are indexed with term vectors
        assertTrue(fieldNames.contains("tvposition"));
        
        fieldNames = reader.getFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET);
        assertEquals(1, fieldNames.size());    // 4 fields are indexed with term vectors
        assertTrue(fieldNames.contains("tvoffset"));
                
        fieldNames = reader.getFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET);
        assertEquals(1, fieldNames.size());    // 4 fields are indexed with term vectors
        assertTrue(fieldNames.contains("tvpositionoffset"));
        reader.close();
        d.close();
    }

View Full Code Here

        reader.close();
        d.close();
    }


  public void testTermVectors() throws Exception {
    Directory d = newDirectory();
    // set up writer
    IndexWriter writer = new IndexWriter(
        d,
        newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)).
            setMergePolicy(newLogMergePolicy())
    );
    // want to get some more segments here
    // new termvector fields
    int mergeFactor = ((LogMergePolicy) writer.getConfig().getMergePolicy()).getMergeFactor();
    for (int i = 0; i < 5 * mergeFactor; i++) {
      Document doc = new Document();
        doc.add(new Field("tvnot","one two two three three three", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.NO));
        doc.add(new Field("termvector","one two two three three three", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.YES));
        doc.add(new Field("tvoffset","one two two three three three", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_OFFSETS));
        doc.add(new Field("tvposition","one two two three three three", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS));
        doc.add(new Field("tvpositionoffset","one two two three three three", Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS));


        writer.addDocument(doc);
    }
    writer.close();
    IndexReader reader = IndexReader.open(d, false);
    FieldSortedTermVectorMapper mapper = new FieldSortedTermVectorMapper(new TermVectorEntryFreqSortedComparator());
    reader.getTermFreqVector(0, mapper);
    Map<String,SortedSet<TermVectorEntry>> map = mapper.getFieldToTerms();
    assertTrue("map is null and it shouldn't be", map != null);
    assertTrue("map Size: " + map.size() + " is not: " + 4, map.size() == 4);
    Set<TermVectorEntry> set = map.get("termvector");
    for (Iterator<TermVectorEntry> iterator = set.iterator(); iterator.hasNext();) {
      TermVectorEntry entry =  iterator.next();
      assertTrue("entry is null and it shouldn't be", entry != null);
      if (VERBOSE) System.out.println("Entry: " + entry);
    }
    reader.close();
    d.close();
  }

View Full Code Here


    }


    
    public void testBinaryFields() throws IOException {
        Directory dir = newDirectory();
        byte[] bin = new byte[]{0, 1, 2, 3, 4, 5, 6, 7, 8, 9};
        
        IndexWriter writer = new IndexWriter(dir, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)).setMergePolicy(newLogMergePolicy()));
        
        for (int i = 0; i < 10; i++) {
          addDoc(writer, "document number " + (i + 1));
          addDocumentWithFields(writer);
          addDocumentWithDifferentFields(writer);
          addDocumentWithTermVectorFields(writer);
        }
        writer.close();
        writer = new IndexWriter(dir, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)).setOpenMode(OpenMode.APPEND).setMergePolicy(newLogMergePolicy()));
        Document doc = new Document();
        doc.add(new Field("bin1", bin));
        doc.add(new Field("junk", "junk text", Field.Store.NO, Field.Index.ANALYZED));
        writer.addDocument(doc);
        writer.close();
        IndexReader reader = IndexReader.open(dir, false);
        doc = reader.document(reader.maxDoc() - 1);
        Field[] fields = doc.getFields("bin1");
        assertNotNull(fields);
        assertEquals(1, fields.length);
        Field b1 = fields[0];
        assertTrue(b1.isBinary());
        byte[] data1 = b1.getBinaryValue();
        assertEquals(bin.length, b1.getBinaryLength());
        for (int i = 0; i < bin.length; i++) {
          assertEquals(bin[i], data1[i + b1.getBinaryOffset()]);
        }
        Set<String> lazyFields = new HashSet<String>();
        lazyFields.add("bin1");
        FieldSelector sel = new SetBasedFieldSelector(new HashSet<String>(), lazyFields);
        doc = reader.document(reader.maxDoc() - 1, sel);
        Fieldable[] fieldables = doc.getFieldables("bin1");
        assertNotNull(fieldables);
        assertEquals(1, fieldables.length);
        Fieldable fb1 = fieldables[0];
        assertTrue(fb1.isBinary());
        assertEquals(bin.length, fb1.getBinaryLength());
        data1 = fb1.getBinaryValue();
        assertEquals(bin.length, fb1.getBinaryLength());
        for (int i = 0; i < bin.length; i++) {
          assertEquals(bin[i], data1[i + fb1.getBinaryOffset()]);
        }
        reader.close();
        // force optimize




        writer = new IndexWriter(dir, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)).setOpenMode(OpenMode.APPEND).setMergePolicy(newLogMergePolicy()));
        writer.optimize();
        writer.close();
        reader = IndexReader.open(dir, false);
        doc = reader.document(reader.maxDoc() - 1);
        fields = doc.getFields("bin1");
        assertNotNull(fields);
        assertEquals(1, fields.length);
        b1 = fields[0];
        assertTrue(b1.isBinary());
        data1 = b1.getBinaryValue();
        assertEquals(bin.length, b1.getBinaryLength());
        for (int i = 0; i < bin.length; i++) {
          assertEquals(bin[i], data1[i + b1.getBinaryOffset()]);
        }
        reader.close();
        dir.close();
    }

View Full Code Here

    }


    // Make sure attempts to make changes after reader is
    // closed throws IOException:
    public void testChangesAfterClose() throws IOException {
        Directory dir = newDirectory();


        IndexWriter writer = null;
        IndexReader reader = null;
        Term searchTerm = new Term("content", "aaa");


        //  add 11 documents with term : aaa
        writer  = new IndexWriter(dir, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)));
        for (int i = 0; i < 11; i++) {
            addDoc(writer, searchTerm.text());
        }
        writer.close();


        reader = IndexReader.open(dir, false);


        // Close reader:
        reader.close();


        // Then, try to make changes:
        try {
          reader.deleteDocument(4);
          fail("deleteDocument after close failed to throw IOException");
        } catch (AlreadyClosedException e) {
          // expected
        }


        try {
          reader.setNorm(5, "aaa", 2.0f);
          fail("setNorm after close failed to throw IOException");
        } catch (AlreadyClosedException e) {
          // expected
        }


        try {
          reader.undeleteAll();
          fail("undeleteAll after close failed to throw IOException");
        } catch (AlreadyClosedException e) {
          // expected
        }
        dir.close();
    }

View Full Code Here

        dir.close();
    }


    // Make sure we get lock obtain failed exception with 2 writers:
    public void testLockObtainFailed() throws IOException {
        Directory dir = newDirectory();


        Term searchTerm = new Term("content", "aaa");


        //  add 11 documents with term : aaa
        IndexWriter writer = new IndexWriter(dir, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)));
        writer.commit();
        for (int i = 0; i < 11; i++) {
            addDoc(writer, searchTerm.text());
        }


        // Create reader:
        IndexReader reader = IndexReader.open(dir, false);


        // Try to make changes
        try {
          reader.deleteDocument(4);
          fail("deleteDocument should have hit LockObtainFailedException");
        } catch (LockObtainFailedException e) {
          // expected
        }


        try {
          reader.setNorm(5, "aaa", 2.0f);
          fail("setNorm should have hit LockObtainFailedException");
        } catch (LockObtainFailedException e) {
          // expected
        }


        try {
          reader.undeleteAll();
          fail("undeleteAll should have hit LockObtainFailedException");
        } catch (LockObtainFailedException e) {
          // expected
        }
        writer.close();
        reader.close();
        dir.close();
    }

View Full Code Here

    }


    // Make sure you can set norms & commit even if a reader
    // is open against the index:
    public void testWritingNorms() throws IOException {
        Directory dir = newDirectory();
        IndexWriter writer;
        IndexReader reader;
        Term searchTerm = new Term("content", "aaa");


        //  add 1 documents with term : aaa
        writer = new IndexWriter(dir, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)));
        addDoc(writer, searchTerm.text());
        writer.close();


        //  now open reader & set norm for doc 0
        reader = IndexReader.open(dir, false);
        reader.setNorm(0, "content", (float) 2.0);


        // we should be holding the write lock now:
        assertTrue("locked", IndexWriter.isLocked(dir));


        reader.commit();


        // we should not be holding the write lock now:
        assertTrue("not locked", !IndexWriter.isLocked(dir));


        // open a 2nd reader:
        IndexReader reader2 = IndexReader.open(dir, false);


        // set norm again for doc 0
        reader.setNorm(0, "content", (float) 3.0);
        assertTrue("locked", IndexWriter.isLocked(dir));


        reader.close();


        // we should not be holding the write lock now:
        assertTrue("not locked", !IndexWriter.isLocked(dir));


        reader2.close();
        dir.close();
    }

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.lucene.store.Directory

com.alimama.mdrill.buffer.CacheKeyBuffer

com.browseengine.bobo.test.FacetNameTest

com.browseengine.bobo.test.FacetNotValuesTest

com.gentics.cr.lucene.indexer.index.LuceneDirectoryFactory

com.gentics.cr.lucene.indexer.index.LuceneDirectoryFactoryTest

com.gentics.cr.lucene.indexer.index.LuceneSingleIndexLocation

com.tamingtext.tagging.LuceneCategoryExtractor

com.tamingtext.tagging.LuceneTagExtractor

org.apache.lucene.benchmark.byTask.tasks.AddIndexesTaskTest

org.apache.lucene.codecs.lucene40.TestBitVector

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.