Lucene简单总结

Posted 2020-11-17 lifuwei

Lucene

API

Document

Document:文档对象，是一条原始数据

文档编号	文档内容
1	谷歌地图之父跳槽FaceBook
2	谷歌地图之父加盟FaceBook
3	谷歌地图创始人拉斯离开谷歌加盟Facebook
4	谷歌地图之父跳槽Facebook与Wave项目取消有关
5	谷歌地图之父拉斯加盟社交网站Facebook

==一条记录就是一个document，document的每一个字段就是一个Field==

Field

创建索引

private final static File INDEX_FILE = new File("E:\\DevelopTools\\indexDir");

public static void indexCreate(Document doc) throws Exception {
    // 创建目录对象，指定索引库的存放位置；FSDirectory文件系统；RAMDirectory内存
    Directory dir = FSDirectory.open(INDEX_FILE);
    // 创建分词器对象
    Analyzer analyzer = new IKAnalyzer();
    // 创建索引写入器配置对象，第一个参数版本VerSion.LATEST,第二个参数分词器
    IndexWriterConfig conf = new IndexWriterConfig(Version.LATEST, analyzer);
    // 创建索引写入器
    IndexWriter indexWriter = new IndexWriter(dir, conf);
    // 向索引库写入文档对象
    indexWriter.addDocument(doc);
    // 提交
    indexWriter.commit();
    // 关闭
    indexWriter.close();
}

@Test
public void createTest() throws Exception {
    //
    Document doc = new Document();
    doc.add(new LongField("id", 1, Store.YES));
    doc.add(new TextField("title", "谷歌地图之父跳槽FaceBook", Store.YES));
    doc.add(new TextField("context", "据国外媒体报道，曾先后负责谷歌地图和Wave开发工作的拉斯·拉斯姆森(Lars Rasmussen)已经离开谷歌，并将加盟Facebook。", Store.YES));
    indexCreate(doc);
}

查询索引

public static void indexSearcher(Query query, Integer n) throws IOException {
    // 初始化索引库对象
    Directory dir = FSDirectory.open(INDEX_FILE);
    // 索引读取工具
    IndexReader indexReader = DirectoryReader.open(dir);
    // 索引搜索对象       
    IndexSearcher indexSeracher = new IndexSearcher(indexReader);   
    // 执行搜索操作，返回值topDocs    
    TopDocs topDocs = indexSeracher.search(query, n);
    // 匹配搜索条件的总记录数
    System.out.println("一共命中：" + topDocs.totalHits + "条数据");
    // 获得得分文档数组对象，得分文档对象包含得分和文档编号
    ScoreDoc[] scoreDocs = topDocs.scoreDocs;
    for (ScoreDoc scoreDoc : scoreDocs) {
        int docID = scoreDoc.doc;
        float score = scoreDoc.score;
        System.out.println("文档编号：" + docID);
        System.out.println("文档得分：" + score);
        // 获取文档对象，通过索引读取工具
        Document document = indexReader.document(docID);
        System.out.println("id:" + document.get("id"));
        System.out.println("title:" + document.get("title"));
        System.out.println("context:" + document.get("context"));
    }
    indexReader.close();
}

@Test
public void searchTest() throws Exception {
    //单一字段的查询解析器
    // 创建查询解析器对象
    QueryParser parser = new QueryParser("title", new IKAnalyzer());
    // 创建查询对象
    Query query = parser.parse("谷歌");
    //根据Query搜索，返回评分最高的n条记录
    indexSearcher(query, 10);
    
    /*多字段的查询解析器
    MultiFieldQueryParser parser = new MultiFieldQueryParser(new String[]{"id","title"}, new IKAnalyzer());
    Query query = parser.parse("1");*/
}

各种其他查询方式

//词条查询，查询条件必须是最小粒度不可再分割的内容
Query query = new TermQuery(new Term("title", "谷歌"));

//通配符查询, ?：匹配一个字符, *：匹配多个字符
Query query = new WildcardQuery(new Term("title", "*歌*"));

//模糊查询, 参数：1-词条，查询字段及关键词，关键词允许写错；2-允许写错的最大编辑距离，并且不能大于2（0~2）
Query query = new FuzzyQuery(new Term("title", "facebool"), 1);

//数值范围查询,查询非String类型的数据或者说是一些继承Numeric类的对象的查询,参数1-字段；2-最小值；3-最大值；4-是否包含最小值；5-是否包含最大值
Query query = NumericRangeQuery.newLongRange("id", 2l, 4l, true, true);

//组合查询， 交集： Occur.MUST + Occur.MUST, 并集：Occur.SHOULD + Occur.SHOULD, 非：Occur.MUST_NOT
BooleanQuery query = new BooleanQuery();
Query query1 = NumericRangeQuery.newLongRange("id", 2l, 4l, true, true);
Query query2 = new WildcardQuery(new Term("title", "*歌*"));
query.add(query1, Occur.SHOULD);
query.add(query2, Occur.SHOULD);

修改索引

//本质先删除再添加，先删除所有满足条件的文档，再创建文档， 因此，修改索引通常要根据唯一字段
public static void indexUpdate(Term term, Document doc) throws IOException {
    Directory dir = FSDirectory.open(INDEX_FILE);
    Analyzer analyzer = new IKAnalyzer();
    IndexWriterConfig conf = new IndexWriterConfig(Version.LATEST, analyzer);
    IndexWriter indexWriter = new IndexWriter(dir, conf);
    indexWriter.updateDocument(term, doc);
    indexWriter.commit();
    indexWriter.close();
    }

@Test
public void updateTest() throws Exception {
    Term term = new Term("title", "facebook");
    Document doc = new Document();
    doc.add(new LongField("id", 1L, Store.YES));
    doc.add(new TextField("title", "谷歌地图之父跳槽FaceBook", Store.YES));
    doc.add(new TextField("context", "河马程序员加盟FaceBook", Store.YES));
    indexUpdate(term, doc);
}

删除索引

// 执行删除操作(根据词条)，要求id字段必须是字符串类型
public static void indexDelete(Term term) throws IOException {
    Directory dir = FSDirectory.open(INDEX_FILE);
    Analyzer analyzer = new IKAnalyzer();
    IndexWriterConfig conf = new IndexWriterConfig(Version.LATEST, analyzer);
    IndexWriter indexWriter = new IndexWriter(dir, conf);
    indexWriter.deleteDocuments(term);
    indexWriter.commit();
    indexWriter.close();
    }

public static void indexDeleteAll() throws IOException {
    Directory dir = FSDirectory.open(INDEX_FILE);
    Analyzer analyzer = new IKAnalyzer();
    IndexWriterConfig conf = new IndexWriterConfig(Version.LATEST, analyzer);
    IndexWriter indexWriter = new IndexWriter(dir, conf);
    indexWriter.deleteAll();
    indexWriter.commit();
    indexWriter.close();
    }

@Test
public void deleteTest() throws Exception {
    /*
     * Term term = new Term("context", "facebook"); indexDelete(term);
     */
    indexDeleteAll();
}