Java Highlighter.getBestTextFragments方法代码示例

import org.apache.lucene.search.highlight.Highlighter; //导入方法依赖的package包/类
/**
 * Performs highlighting for a given query and a given document.
 *
 * @param indexSearcher the IndexSearcher performing the query
 * @param query the Tripod LuceneQuery
 * @param scoreDoc the Lucene ScoreDoc
 * @param doc the Lucene Document
 * @param highlighter the Highlighter to use
 * @param result the QueryResult to add the highlights to
 * @throws IOException if an error occurs performing the highlighting
 * @throws InvalidTokenOffsetsException if an error occurs performing the highlighting
 */
protected void performHighlighting(final IndexSearcher indexSearcher, final Query query, final ScoreDoc scoreDoc,
                                   final Document doc, final Highlighter highlighter, final QR result)
        throws IOException, InvalidTokenOffsetsException {

    if (query.getHighlightFields() == null || query.getHighlightFields().isEmpty()) {
        return;
    }

    final List<Highlight> highlights = new ArrayList<>();
    final List<String> hlFieldNames = getHighlightFieldNames(query, doc);

    // process each field to highlight on
    for (String hlField : hlFieldNames) {
        final String text = doc.get(hlField);
        if (StringUtils.isEmpty(text)) {
            continue;
        }

        final List<String> snippets = new ArrayList<>();
        final Fields tvFields = indexSearcher.getIndexReader().getTermVectors(scoreDoc.doc);
        final int maxStartOffset = highlighter.getMaxDocCharsToAnalyze() -1;

        // get the snippets for the given field
        final TokenStream tokenStream = TokenSources.getTokenStream(hlField, tvFields, text, analyzer, maxStartOffset);
        final TextFragment[] textFragments = highlighter.getBestTextFragments(tokenStream, text, false, 10);
        for (TextFragment textFragment : textFragments) {
            if (textFragment != null && textFragment.getScore() > 0) {
                snippets.add(textFragment.toString());
            }
        }

        // if we have snippets then add a highlight result to the QueryResult
        if (snippets.size() > 0) {
            highlights.add(new Highlight(hlField, snippets));
        }
    }

    result.setHighlights(highlights);
}
 

import org.apache.lucene.search.highlight.Highlighter; //导入方法依赖的package包/类
@Override
protected BenchmarkHighlighter getBenchmarkHighlighter(Query q){
  highlighter = new Highlighter(new SimpleHTMLFormatter(), new QueryScorer(q));
  highlighter.setMaxDocCharsToAnalyze(maxDocCharsToAnalyze);
  return new BenchmarkHighlighter(){
    @Override
    public int doHighlight(IndexReader reader, int doc, String field,
        Document document, Analyzer analyzer, String text) throws Exception {
      TokenStream ts = TokenSources.getAnyTokenStream(reader, doc, field, document, analyzer);
      TextFragment[] frag = highlighter.getBestTextFragments(ts, text, mergeContiguous, maxFrags);
      return frag != null ? frag.length : 0;
    }
  };
}
 

import org.apache.lucene.search.highlight.Highlighter; //导入方法依赖的package包/类
@Override
public BenchmarkHighlighter getBenchmarkHighlighter(Query q) {
  highlighter = new Highlighter(new SimpleHTMLFormatter(), new QueryScorer(q));
  return new BenchmarkHighlighter() {
    @Override
    public int doHighlight(IndexReader reader, int doc, String field, Document document, Analyzer analyzer, String text) throws Exception {
      TokenStream ts = TokenSources.getAnyTokenStream(reader, doc, field, document, analyzer);
      TextFragment[] frag = highlighter.getBestTextFragments(ts, text, mergeContiguous, maxFrags);
      numHighlightedResults += frag != null ? frag.length : 0;
      return frag != null ? frag.length : 0;
    }
  };
}
 

import org.apache.lucene.search.highlight.Highlighter; //导入方法依赖的package包/类
/**
 * NOTE: This method will not preserve the correct field types.
 * 
 * @param preTag
 * @param postTag
 */
public static Document highlight(int docId, Document document, Query query, FieldManager fieldManager,
    IndexReader reader, String preTag, String postTag) throws IOException, InvalidTokenOffsetsException {

  String fieldLessFieldName = fieldManager.getFieldLessFieldName();

  Query fixedQuery = fixSuperQuery(query, null, fieldLessFieldName);

  Analyzer analyzer = fieldManager.getAnalyzerForQuery();

  SimpleHTMLFormatter htmlFormatter = new SimpleHTMLFormatter(preTag, postTag);
  Document result = new Document();
  for (IndexableField f : document) {
    String name = f.name();
    if (fieldLessFieldName.equals(name) || FIELDS_NOT_TO_HIGHLIGHT.contains(name)) {
      result.add(f);
      continue;
    }
    String text = f.stringValue();
    Number numericValue = f.numericValue();

    Query fieldFixedQuery;
    if (fieldManager.isFieldLessIndexed(name)) {
      fieldFixedQuery = fixSuperQuery(query, name, fieldLessFieldName);
    } else {
      fieldFixedQuery = fixedQuery;
    }

    if (numericValue != null) {
      if (shouldNumberBeHighlighted(name, numericValue, fieldFixedQuery)) {
        String numberHighlight = preTag + text + postTag;
        result.add(new StringField(name, numberHighlight, Store.YES));
      }
    } else {
      Highlighter highlighter = new Highlighter(htmlFormatter, new QueryScorer(fieldFixedQuery, name));
      TokenStream tokenStream = TokenSources.getAnyTokenStream(reader, docId, name, analyzer);
      TextFragment[] frag = highlighter.getBestTextFragments(tokenStream, text, false, 10);
      for (int j = 0; j < frag.length; j++) {
        if ((frag[j] != null) && (frag[j].getScore() > 0)) {
          result.add(new StringField(name, frag[j].toString(), Store.YES));
        }
      }
    }
  }
  return result;
}