Java Weight.extractTerms方法代码示例

import org.apache.lucene.search.Weight; //导入方法依赖的package包/类
@Override
protected SpanQuery convertUnknownQuery(String field, Query query) {
  if (query instanceof CommonTermsQuery) {

    // specialized since rewriting would change the result query
    // this query is TermContext sensitive.
    CommonTermsQuery ctq = (CommonTermsQuery) query;

    Set<Term> terms = new HashSet<>();
    try {
      Weight w = ctq.createWeight(searcher, false, 1.0f);
      w.extractTerms(terms);
    } catch (IOException e) {
      throw new RuntimeException("IOException on searcher!!!", e);
    }
    List<SpanQuery> spanQs = new LinkedList<>();

    for (Term term : terms) {
      if (term.field().equals(field)) {
        spanQs.add(new SpanTermQuery(term));
      }
    }
    if (spanQs.size() == 0) {
      return getEmptySpanQuery();
    } else if (spanQs.size() == 1) {
      return spanQs.get(0);
    } else {
      return new SpanOrQuery(spanQs.toArray(new SpanQuery[spanQs.size()]));
    }
  }
  super.convertUnknownQuery(field, query);
  return null;
}
 

import org.apache.lucene.search.Weight; //导入方法依赖的package包/类
/**
 * Simple utility class to perform basic term frequency/document frequency
 * counts on the individual terms within a query.  This relies on
 * IndexReader and does not perform any concordance search/retrieval;
 * it is, therefore, very fast.
 * <p>
 * If you want to visit more than basic terms (e.g. SpanNear),
 * see {@link org.apache.lucene.search.concordance.windowvisitor.TargetVisitor}
 *
 * @param query query
 * @param searcher searcher
 * @return target term results
 * @throws java.io.IOException if there is an IOException from the searcher
 */
public SimpleTargetTermResults searchSingleTerms(Query query, IndexSearcher searcher)
    throws IOException {
  Query tmpQ = query.rewrite(searcher.getIndexReader());
  Set<Term> terms = new HashSet<>();
  Weight weight = tmpQ.createWeight(searcher, false, 1.0f);//default boost 1.0f.
  weight.extractTerms(terms);

  Map<String, Integer> dfs = new HashMap<>();
  Map<String, Integer> tfs = new HashMap<>();

  for (Term t : terms) {
    String targ = t.text();
    int docFreq = searcher.getIndexReader().docFreq(t);
    if (docFreq == 0) {
      continue;
    }
    Integer i = Integer.valueOf(docFreq);
    dfs.put(targ, i);

    long tf = searcher.getIndexReader().totalTermFreq(t);
    tfs.put(targ, (int) tf);
  }

  return new SimpleTargetTermResults(dfs, tfs);
}
 

import org.apache.lucene.search.Weight; //导入方法依赖的package包/类
@Override
public void extractTerms(Set<Term> terms) {
    for (Weight w : weights) {
        w.extractTerms(terms);
    }
}