Lucene的全文搜索及高亮显示

目录

测试的效果如下

1.引入maven依赖

2.索引目录存放数据

3.controller 

4.处理类

---

测试的效果如下

 

 

1.引入maven依赖

 <!-- lucene核心库 -->
        <dependency>
            <groupId>org.apache.lucene</groupId>
            <artifactId>lucene-core</artifactId>
            <version>8.2.0</version>
        </dependency>
        <!-- lucene的默认分词器库 -->
        <dependency>
            <groupId>org.apache.lucene</groupId>
            <artifactId>lucene-analyzers-common</artifactId>
            <version>8.2.0</version>
        </dependency>
        <!-- Lucene的查询解析器 -->
        <dependency>
        <groupId>org.apache.lucene</groupId>
        <artifactId>lucene-queryparser</artifactId>
        <version>8.2.0</version>
        </dependency>
        <!-- lucene的高亮显示 -->
        <dependency>
            <groupId>org.apache.lucene</groupId>
            <artifactId>lucene-highlighter</artifactId>
            <version>8.2.0</version>
        </dependency>

2.索引目录存放数据

@Component
public class IndexBuilder {


    //建立相应的索引目录
    public void buildIndex(List<User> users) throws IOException {
        Directory directory = FSDirectory.open(Paths.get("./index"));
        Analyzer analyzer = new StandardAnalyzer();
        IndexWriterConfig config = new IndexWriterConfig(analyzer);
        config.setOpenMode(IndexWriterConfig.OpenMode.CREATE);

        IndexWriter indexWriter = new IndexWriter(directory, config);
        
        for (User user:users){
            Document doc = new Document();
            doc.add(new TextField("uid", user.getUid(), Field.Store.YES));
            doc.add(new TextField("username", user.getUsername(), Field.Store.YES));
            doc.add(new TextField("password", user.getPassword(), Field.Store.YES));
            doc.add(new TextField("name", user.getName(), Field.Store.YES));
            doc.add(new TextField("email", user.getEmail(), Field.Store.YES));
            doc.add(new TextField("phone", user.getPhone(), Field.Store.YES));
            doc.add(new TextField("sex", user.getSex(), Field.Store.YES));
            doc.add(new TextField("state", user.getState(), Field.Store.YES));
            doc.add(new TextField("code", user.getCode(), Field.Store.YES));
            doc.add(new TextField("addr", user.getAddr(), Field.Store.YES));
            indexWriter.addDocument(doc);
        }


        indexWriter.close();
    }

}

3.controller 

  @Autowired
    private SearchService searchService;

    @RequestMapping(value = "/search")
    public String search(@RequestParam String keyword, Model model) throws IOException, ParseException, InvalidTokenOffsetsException {
        List<User> search = searchService.search(keyword);
        model.addAttribute("obj",search);
        return "hello/search";
    }

4.处理类

@Component
public class SearchService {
    @Autowired
    private UserMapper mapper;
    @Autowired
    private IndexBuilder indexBuilder;

    public List<User> search(String keyword) throws IOException, ParseException, InvalidTokenOffsetsException {
        //查包含aaa
        //把查到的数据丢到索引目录,用作测试，，这个索引目录应该是提前建好的
        List<User> users = mapper.selectList(null);
        indexBuilder.buildIndex(users);

        //1.打开索引目录得到indexReader用于读取文件
        List<User> results = new ArrayList<>();
        Directory directory = FSDirectory.open(Paths.get("./index"));
        IndexReader indexReader = DirectoryReader.open(directory);


        //2.构建查询器，得到query，用于表示我要哪些字段
        Analyzer analyzer = new StandardAnalyzer();

        //查单个字段，比如只要content字段
        //按传过来的值，在index目录下对content搜索
//        QueryParser parser = new QueryParser("content", analyzer);
//        Query query = parser.parse(keyword);

        //多个字段
        String[] fields = {"uid", "username","password","name","email","phone","sex","state","code","addr"}; // 定义多个字段
        MultiFieldQueryParser parser = new MultiFieldQueryParser(fields, analyzer);
        Query query = parser.parse(keyword);

        //3.得到10条文档数据
        IndexSearcher indexSearcher = new IndexSearcher(indexReader);
        TopDocs topDocs = indexSearcher.search(query, 10);//查十个
        ScoreDoc[] scoreDocs = topDocs.scoreDocs;

        //3.对文档数据进行高亮处理
        results=headle(query,scoreDocs,indexSearcher,analyzer);

        indexReader.close();
        directory.close();

        return results;
    }

    //对查询结果进行处理及高亮显示
    public List<User> headle(Query query,ScoreDoc[] scoreDocs,IndexSearcher searcher,Analyzer analyzer ) throws IOException, InvalidTokenOffsetsException {
        // 查询高亮，对于name字段中符合keyWord的高亮
        QueryScorer score = new QueryScorer(query);
        List<User> results = new ArrayList<>();
        // 定制高亮标签
        SimpleHTMLFormatter simpleHTMLFormatter = new SimpleHTMLFormatter("<span style='color:red'>", "</span>");
        // 高亮分析器
        Highlighter highlighter = new Highlighter(simpleHTMLFormatter, score);

        //对每个文档对象进行遍历
        for (ScoreDoc scoreDoc : scoreDocs) {
            User user=new User();
            // 取出这一条文档编号
            int docId = scoreDoc.doc;
            // 根据编号去找文档
            Document doc = searcher.doc(docId);
            //一个文档对应数据库的一条记录，fields是这条记录的所以字段字段，循环字段
            List<IndexableField> fields = doc.getFields();
            for (IndexableField f : fields) {
                //我这里对他的所有字段做高亮处理，可以通过if语句筛选
                //f.name()//是这个字段的名称，如uid 等等
                //如                                                uid    1
                TokenStream tokenStream = analyzer.tokenStream(f.name(), new StringReader(doc.get(f.name())));
                // 获取高亮的片段
                String fieldContent = highlighter.getBestFragment(tokenStream, doc.get(f.name()));
                //如果fieldContent==null，说明这个字段无需高亮显示
                try {
                    Class<User> aClass = User.class;
                    Field field=aClass.getDeclaredField(f.name());
                    Method setNameMethod = aClass.getMethod("set" + f.name().substring(0, 1).toUpperCase() + f.name().substring(1), field.getType());
                    //私有方法
                    setNameMethod.setAccessible(true);
                    if (fieldContent!=null)
                        //通过反射调用对应的set方法
                        setNameMethod.invoke(user,fieldContent);
                    else
                        setNameMethod.invoke(user,doc.get(f.name()));

                } catch (Exception e) {
                    e.printStackTrace();
                }
            }
            results.add(user);
        }
        return results;
    }

}