Lucene利用Hanlp分词索引和检索

包引入

• pom引入lucene和hanlp
• lucene版本为7.4

• hanlp分词器版本为1.1.6

  <dependency>
      <groupId>org.apache.lucene</groupId>
      <artifactId>lucene-core</artifactId>
      <version>7.4.0</version>
  </dependency>
  
  <dependency>
      <groupId>org.apache.lucene</groupId>
      <artifactId>lucene-queryparser</artifactId>
      <version>7.4.0</version>
  </dependency>
  
  <dependency>
      <groupId>com.hankcs.nlp</groupId>
      <artifactId>hanlp-lucene-plugin</artifactId>
      <version>1.1.6</version>
  </dependency>

  对数据文件进行索引

  public class DocIndex {
  
      // 建立索引文件的目录
      public static String indexPath = "/index";
  
      public static void index(String path, Map<String, SugRecord> map, boolean reIndex) {
  
          indexPath = path + indexPath;
  
          if (!reIndex) {
              return;
          }
  
          IndexWriter writer = null;
          try {
              // 存储索引数据的目录
              Directory dir = FSDirectory.open(Paths.get(indexPath));
              // 创建分析器
              Analyzer analyzer = new HanLPAnalyzer();
              IndexWriterConfig iwc = new IndexWriterConfig(analyzer);
              iwc.setOpenMode(IndexWriterConfig.OpenMode.CREATE);
  
              writer = new IndexWriter(dir, iwc);
              indexDoc(writer, map);
              writer.close();
          } catch (IOException e) {
              e.printStackTrace();
          }
      }
  
      private static void indexDoc(IndexWriter writer, Map<String, SugRecord> map) {
          map.forEach((k, v) -> {
              // 创建一个新的空文档
              Document doc = new Document();
              //创建索引字段
              doc.add(new TextField("id", v.getId(), Field.Store.YES));
              doc.add(new TextField("question", v.getQueryNorm(), Field.Store.NO));
              doc.add(new TextField("query", v.getQuery(), Field.Store.NO));
              // 写文档
              try {
                  writer.addDocument(doc);
              } catch (IOException e) {
                  e.printStackTrace();
              }
          });
          try {
              writer.flush();
              writer.commit();
              writer.close();
          } catch (IOException e) {
              e.printStackTrace();
          }
      }
  }

对索引进行检索

@Slf4j
public class DocSearcher {

    public static List<SugTerm> search(String content) {
        IndexReader reader = null;
        try {
            reader = DirectoryReader.open(FSDirectory.open(Paths.get(DocIndex.indexPath)));
        } catch (IOException e) {
            e.printStackTrace();
        }
        IndexSearcher searcher = new IndexSearcher(reader);

        //或关系，多域检索
        String[] fields = {"question", "query"};
        BooleanClause.Occur[] clauses = {BooleanClause.Occur.SHOULD, BooleanClause.Occur.SHOULD};
        try {
            //指定分词器
            Query query = MultiFieldQueryParser.parse(content, fields, clauses, new HanLPAnalyzer());
            return doPagingSearch(searcher, query);
        } catch (Exception e) {
            e.printStackTrace();
        }
        return null;
    }

    private static List<Term> doPagingSearch(IndexSearcher searcher, Query query) throws IOException {
        // TopDocs保存搜索结果
        TopDocs results = null;
        List<Term> termList = new ArrayList<>();
        //分页数
        results = searcher.search(query, 10);
        ScoreDoc[] hits = results.scoreDocs;
        for (ScoreDoc hit : hits) {
            Document document = searcher.doc(hit.doc);
            String id = Objects.requireNonNull(document).get("id");
            Term term = new Term();
            term.id = id;
            term.score = hit.score;
            term.add(sugTerm);
        }
        log.info(termList.toString());
        return termList;
    }
}