Java FieldType.setStoreTermVectorPositions方法代码示例

import org.apache.lucene.document.FieldType; //导入方法依赖的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.document.FieldType; //导入方法依赖的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.document.FieldType; //导入方法依赖的package包/类
public void testFieldTypeToTermVectorString() throws Exception {
    FieldType ft = new FieldType();
    ft.setStoreTermVectorOffsets(false);
    ft.setStoreTermVectorPayloads(true);
    ft.setStoreTermVectors(true);
    ft.setStoreTermVectorPositions(true);
    String ftOpts = FieldMapper.termVectorOptionsToString(ft);
    assertThat("with_positions_payloads", equalTo(ftOpts));
    AllFieldMapper.Builder builder = new AllFieldMapper.Builder(null);
    boolean exceptiontrown = false;
    try {
        TypeParsers.parseTermVector("", ftOpts, builder);
    } catch (MapperParsingException e) {
        exceptiontrown = true;
    }
    assertThat("TypeParsers.parseTermVector should accept string with_positions_payloads but does not.", exceptiontrown, equalTo(false));
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
public LuceneUtil(JasperReportsContext jasperReportsContext, boolean isCaseSensitive, boolean isWholeWordsOnly, boolean removeAccents) {
	this.isCaseSensitive = isCaseSensitive;
	this.isWholeWordsOnly = isWholeWordsOnly;
	this.removeAccents = removeAccents;

	this.noneSelector = JRStyledTextAttributeSelector.getNoneSelector(jasperReportsContext);
	this.styledTextUtil = JRStyledTextUtil.getInstance(jasperReportsContext);
	
	fieldType = new FieldType();
	fieldType.setIndexed(true);
	fieldType.setTokenized(true);
	fieldType.setStored(true);
	fieldType.setStoreTermVectors(true);
	fieldType.setStoreTermVectorPositions(true);
	fieldType.setStoreTermVectorOffsets(true);
	fieldType.freeze();
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
private void doTestMixedPostings(Codec codec) throws Exception {
  Directory dir = newDirectory();
  IndexWriterConfig iwc = newIndexWriterConfig(new MockAnalyzer(random()));
  iwc.setCodec(codec);
  RandomIndexWriter iw = new RandomIndexWriter(random(), dir, iwc);
  Document doc = new Document();
  FieldType ft = new FieldType(TextField.TYPE_NOT_STORED);
  // turn on vectors for the checkindex cross-check
  ft.setStoreTermVectors(true);
  ft.setStoreTermVectorOffsets(true);
  ft.setStoreTermVectorPositions(true);
  Field idField = new Field("id", "", ft);
  Field dateField = new Field("date", "", ft);
  doc.add(idField);
  doc.add(dateField);
  for (int i = 0; i < 100; i++) {
    idField.setStringValue(Integer.toString(random().nextInt(50)));
    dateField.setStringValue(Integer.toString(random().nextInt(100)));
    iw.addDocument(doc);
  }
  iw.close();
  dir.close(); // checkindex
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
public void testFlushWithNoMerging() throws IOException {
  Directory dir = newDirectory();
  IndexWriter writer = new IndexWriter(
      dir,
      newIndexWriterConfig(new MockAnalyzer(random()))
          .setMaxBufferedDocs(2)
          .setMergePolicy(newLogMergePolicy(10))
  );
  Document doc = new Document();
  FieldType customType = new FieldType(TextField.TYPE_STORED);
  customType.setStoreTermVectors(true);
  customType.setStoreTermVectorPositions(true);
  customType.setStoreTermVectorOffsets(true);
  doc.add(newField("field", "aaa", customType));
  for(int i=0;i<19;i++)
    writer.addDocument(doc);
  writer.flush(false, true);
  writer.close();
  SegmentInfos sis = new SegmentInfos();
  sis.read(dir);
  // Since we flushed w/o allowing merging we should now
  // have 10 segments
  assertEquals(10, sis.size());
  dir.close();
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
private void writeEmptyTermVector(TermVectorsResponse outResponse) throws IOException {

        Directory dir = newDirectory();
        IndexWriterConfig conf = new IndexWriterConfig(new StandardAnalyzer());
        conf.setOpenMode(OpenMode.CREATE);
        IndexWriter writer = new IndexWriter(dir, conf);
        FieldType type = new FieldType(TextField.TYPE_STORED);
        type.setStoreTermVectorOffsets(true);
        type.setStoreTermVectorPayloads(false);
        type.setStoreTermVectorPositions(true);
        type.setStoreTermVectors(true);
        type.freeze();
        Document d = new Document();
        d.add(new Field("id", "abc", StringField.TYPE_STORED));

        writer.updateDocument(new Term("id", "abc"), d);
        writer.commit();
        writer.close();
        DirectoryReader dr = DirectoryReader.open(dir);
        IndexSearcher s = new IndexSearcher(dr);
        TopDocs search = s.search(new TermQuery(new Term("id", "abc")), 1);
        ScoreDoc[] scoreDocs = search.scoreDocs;
        int doc = scoreDocs[0].doc;
        Fields fields = dr.getTermVectors(doc);
        EnumSet<Flag> flags = EnumSet.of(Flag.Positions, Flag.Offsets);
        outResponse.setFields(fields, null, flags, fields);
        outResponse.setExists(true);
        dr.close();
        dir.close();

    }
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
private void writeStandardTermVector(TermVectorsResponse outResponse) throws IOException {

        Directory dir = newDirectory();
        IndexWriterConfig conf = new IndexWriterConfig(new StandardAnalyzer());

        conf.setOpenMode(OpenMode.CREATE);
        IndexWriter writer = new IndexWriter(dir, conf);
        FieldType type = new FieldType(TextField.TYPE_STORED);
        type.setStoreTermVectorOffsets(true);
        type.setStoreTermVectorPayloads(false);
        type.setStoreTermVectorPositions(true);
        type.setStoreTermVectors(true);
        type.freeze();
        Document d = new Document();
        d.add(new Field("id", "abc", StringField.TYPE_STORED));
        d.add(new Field("title", "the1 quick brown fox jumps over  the1 lazy dog", type));
        d.add(new Field("desc", "the1 quick brown fox jumps over  the1 lazy dog", type));

        writer.updateDocument(new Term("id", "abc"), d);
        writer.commit();
        writer.close();
        DirectoryReader dr = DirectoryReader.open(dir);
        IndexSearcher s = new IndexSearcher(dr);
        TopDocs search = s.search(new TermQuery(new Term("id", "abc")), 1);
        ScoreDoc[] scoreDocs = search.scoreDocs;
        int doc = scoreDocs[0].doc;
        Fields termVectors = dr.getTermVectors(doc);
        EnumSet<Flag> flags = EnumSet.of(Flag.Positions, Flag.Offsets);
        outResponse.setFields(termVectors, null, flags, termVectors);
        dr.close();
        dir.close();

    }
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
public void testTermVectorStringGenerationWithoutPositions() throws Exception {
    FieldType ft = new FieldType();
    ft.setStoreTermVectorOffsets(true);
    ft.setStoreTermVectorPayloads(true);
    ft.setStoreTermVectors(true);
    ft.setStoreTermVectorPositions(false);
    String ftOpts = FieldMapper.termVectorOptionsToString(ft);
    assertThat(ftOpts, equalTo("with_offsets"));
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
private void writeStandardTermVector(TermVectorsResponse outResponse) throws IOException {

        Directory dir = LuceneTestCase.newDirectory();
        IndexWriterConfig conf = new IndexWriterConfig(new StandardAnalyzer());

        conf.setOpenMode(OpenMode.CREATE);
        IndexWriter writer = new IndexWriter(dir, conf);
        FieldType type = new FieldType(TextField.TYPE_STORED);
        type.setStoreTermVectorOffsets(true);
        type.setStoreTermVectorPayloads(false);
        type.setStoreTermVectorPositions(true);
        type.setStoreTermVectors(true);
        type.freeze();
        Document d = new Document();
        d.add(new Field("id", "abc", StringField.TYPE_STORED));
        d.add(new Field("plaintext", "the1 quick brown fox jumps over  the1 lazy dog comment", type));
        d.add(new Field("desc", "the1 quick brown fox jumps over  the1 lazy dog comment", type));

        writer.updateDocument(new Term("id", "abc"), d);
        writer.commit();
        writer.close();
        DirectoryReader dr = DirectoryReader.open(dir);
        IndexSearcher s = new IndexSearcher(dr);
        TopDocs search = s.search(new TermQuery(new Term("id", "abc")), 1);
        ScoreDoc[] scoreDocs = search.scoreDocs;
        int doc = scoreDocs[0].doc;
        Fields termVectors = dr.getTermVectors(doc);
        EnumSet<Flag> flags = EnumSet.of(Flag.Positions, Flag.Offsets);
        outResponse.setFields(termVectors, null, flags, termVectors);
        dr.close();
        dir.close();

    }
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
protected void makeUnstoredIndex() throws Exception {
  IndexWriter writer = new IndexWriter(dir, new IndexWriterConfig(
      TEST_VERSION_CURRENT, analyzerW).setOpenMode(OpenMode.CREATE));
  Document doc = new Document();
  FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
  customType.setStoreTermVectors(true);
  customType.setStoreTermVectorOffsets(true);
  customType.setStoreTermVectorPositions(true);
  doc.add( new Field( F, "aaa", customType) );
  //doc.add( new Field( F, "aaa", Store.NO, Index.ANALYZED, TermVector.WITH_POSITIONS_OFFSETS ) );
  writer.addDocument( doc );
  writer.close();
  if (reader != null) reader.close();
  reader = DirectoryReader.open(dir);
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
private Document newDocument() {
  Document doc = new Document();
  for (IndexOptions option : FieldInfo.IndexOptions.values()) {
    FieldType ft = new FieldType(TextField.TYPE_NOT_STORED);
    // turn on tvs for a cross-check, since we rely upon checkindex in this test (for now)
    ft.setStoreTermVectors(true);
    ft.setStoreTermVectorOffsets(true);
    ft.setStoreTermVectorPositions(true);
    ft.setStoreTermVectorPayloads(true);
    ft.setIndexOptions(option);
    doc.add(new Field(option.toString(), "", ft));
  }
  return doc;
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
/**
 * {@inheritDoc}
 */
@Override
public IndexableField createField(SchemaField field, Object value, float boost) {
  final boolean indexed = field.indexed();
  final boolean stored = field.stored();
  final boolean docValues = field.hasDocValues();

  if (!indexed && !stored && !docValues) {
    if (log.isTraceEnabled())
      log.trace("Ignoring unindexed/unstored field: " + field);
    return null;
  }
  final Integer intValue = stringValueToIntValue(value.toString());
  if (intValue == null || intValue.equals(DEFAULT_VALUE))
    throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "Unknown value for enum field: " + value.toString());

  String intAsString =  intValue.toString();
  final FieldType newType = new FieldType();

  newType.setIndexed(field.indexed());
  newType.setTokenized(field.isTokenized());
  newType.setStored(field.stored());
  newType.setOmitNorms(field.omitNorms());
  newType.setIndexOptions(getIndexOptions(field, intAsString));
  newType.setStoreTermVectors(field.storeTermVector());
  newType.setStoreTermVectorOffsets(field.storeTermOffsets());
  newType.setStoreTermVectorPositions(field.storeTermPositions());
  newType.setNumericType(FieldType.NumericType.INT);
  newType.setNumericPrecisionStep(DEFAULT_PRECISION_STEP);

  final org.apache.lucene.document.Field f;
  f = new org.apache.lucene.document.IntField(field.getName(), intValue.intValue(), newType);

  f.setBoost(boost);
  return f;
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
public void testSparseSpan() throws IOException, InvalidTokenOffsetsException {
  final String TEXT = "the fox did not jump";
  final Directory directory = newDirectory();
  final IndexWriter indexWriter = new IndexWriter(directory,
      newIndexWriterConfig(new MockAnalyzer(random(), MockTokenizer.WHITESPACE, false)));
  try {
    final Document document = new Document();
    FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
    customType.setStoreTermVectorOffsets(true);
    customType.setStoreTermVectorPositions(true);
    customType.setStoreTermVectors(true);
    document.add(new Field(FIELD, new TokenStreamSparse(), customType));
    indexWriter.addDocument(document);
  } finally {
    indexWriter.close();
  }
  final IndexReader indexReader = DirectoryReader.open(directory);
  try {
    assertEquals(1, indexReader.numDocs());
    final IndexSearcher indexSearcher = newSearcher(indexReader);
    final Query phraseQuery = new SpanNearQuery(new SpanQuery[] {
        new SpanTermQuery(new Term(FIELD, "did")),
        new SpanTermQuery(new Term(FIELD, "jump")) }, 0, true);

    TopDocs hits = indexSearcher.search(phraseQuery, 1);
    assertEquals(0, hits.totalHits);
    final Highlighter highlighter = new Highlighter(
        new SimpleHTMLFormatter(), new SimpleHTMLEncoder(),
        new QueryScorer(phraseQuery));
    final TokenStream tokenStream = TokenSources
        .getTokenStream(indexReader.getTermVector(
            0, FIELD), false);
    assertEquals(
        highlighter.getBestFragment(new TokenStreamSparse(), TEXT),
        highlighter.getBestFragment(tokenStream, TEXT));
  } finally {
    indexReader.close();
    directory.close();
  }
}
 

import org.apache.lucene.document.FieldType; //导入方法依赖的package包/类
public void testPositionIncrementGapEmptyField() throws Exception {
  Directory dir = newDirectory();
  MockAnalyzer analyzer = new MockAnalyzer(random());
  analyzer.setPositionIncrementGap( 100 );
  IndexWriter w = new IndexWriter(dir, newIndexWriterConfig(analyzer));
  Document doc = new Document();
  FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
  customType.setStoreTermVectors(true);
  customType.setStoreTermVectorPositions(true);
  Field f = newField("field", "", customType);
  Field f2 = newField("field", "crunch man", customType);
  doc.add(f);
  doc.add(f2);
  w.addDocument(doc);
  w.close();

  IndexReader r = DirectoryReader.open(dir);
  Terms tpv = r.getTermVectors(0).terms("field");
  TermsEnum termsEnum = tpv.iterator(null);
  assertNotNull(termsEnum.next());
  DocsAndPositionsEnum dpEnum = termsEnum.docsAndPositions(null, null);
  assertNotNull(dpEnum);
  assertTrue(dpEnum.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
  assertEquals(1, dpEnum.freq());
  assertEquals(100, dpEnum.nextPosition());

  assertNotNull(termsEnum.next());
  dpEnum = termsEnum.docsAndPositions(null, dpEnum);
  assertNotNull(dpEnum);
  assertTrue(dpEnum.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
  assertEquals(1, dpEnum.freq());
  assertEquals(101, dpEnum.nextPosition());
  assertNull(termsEnum.next());

  r.close();
  dir.close();
}