Lucene的CURD使用

说明:由于lucene将我们的文档document分词成为二进制的,无法阅读,所以我们这里用Lucene的官方程序jar包来查看分词后的内容

打开run.bat其实就是一句话,类似于linux系统启动redis类似的命令

java -jar lukeall-4.10.0.jar

公共部分:导入jar包

<dependencies>
    <dependency>
        <groupId>org.apache.lucene</groupId>
        <artifactId>lucene-core</artifactId>
        <version>4.10.2</version>
    </dependency>
    <dependency>
        <groupId>org.apache.lucene</groupId>
        <artifactId>lucene-queries</artifactId>
        <version>4.10.2</version>
    </dependency>
    <dependency>
        <groupId>org.apache.lucene</groupId>
        <artifactId>lucene-test-framework</artifactId>
        <version>4.10.2</version>
    </dependency>
    <dependency>
        <groupId>org.apache.lucene</groupId>
        <artifactId>lucene-analyzers-common</artifactId>
        <version>4.10.2</version>
    </dependency>
    <dependency>
        <groupId>org.apache.lucene</groupId>
        <artifactId>lucene-queryparser</artifactId>
        <version>4.10.2</version>
    </dependency>
    <dependency>
        <groupId>org.apache.lucene</groupId>
        <artifactId>lucene-highlighter</artifactId>
        <version>4.10.2</version>
    </dependency>
    <dependency>
        <groupId>com.janeluo</groupId>
        <artifactId>ikanalyzer</artifactId>
        <version>2012_u6</version>
    </dependency>
    <dependency>
        <groupId>junit</groupId>
        <artifactId>junit</artifactId>
        <version>4.12</version>
    </dependency>

</dependencies>

将IK分词器配置文件拷贝到resource目录下,IK的配置文件提供了Ik强大的拓展功能,能将你需要不分词的当成一个词语

--------------------------------------------增加部分-----------------------------------------------------------------------------------------------

1   添加单个document文档在lucene中

//添加一个documt,相当于添加文章或者文章标题,用ik分析器进行分词存储在一个地方
@Test
public void luceneone() throws Exception {
    //获取检索对象
    Directory directory= FSDirectory.open(new File("f:\\test"));
    IndexWriterConfig indexWriterConfig = new IndexWriterConfig(Version.LATEST,new IKAnalyzer());
    IndexWriter indexWriter= new IndexWriter(directory,indexWriterConfig);
    //获取文档
    Document document = new Document();
    //将文档加入lucene中
    document.add(new TextField("content","黑马程序员蓝瘦香菇隔壁二大爷", Field.Store.YES)); //text类型
    document.add(new IntField("id",1, Field.Store.YES));  //int类型
    document.add(new StringField("title","黑马咋了", Field.Store.YES)); //string类型
    indexWriter.addDocument(document);
    indexWriter.commit();
    indexWriter.close();
}

执行结果:

 

2  添加多个document文档在lucene中

//添加多个document
@Test
public void lucenetwo() throws IOException {
    //创建写入对象
    Directory directory=FSDirectory.open(new File("f:\\test"));
    IndexWriterConfig indexWriterConfig = new IndexWriterConfig(Version.LATEST,new IKAnalyzer());
    IndexWriter indexWriter = new IndexWriter(directory,indexWriterConfig);
    //写入多个文档
    List<Document> list =new ArrayList<Document>();
    //添加五个document
    for(int i=1;i<=5;i++){
    //往一个document中添加faild,内容
    Document document = new Document();
    document.add(new IntField("pid",i, Field.Store.YES));
    document.add(new StringField("pname","格力空调", Field.Store.YES));
    document.add(new TextField("content","我是世界上最好的冰箱,好空调格力造!!!"+i+"号", Field.Store.YES));
    list.add(document);//将一个document放入list中
    }
    indexWriter.addDocuments(list);
    //执行
    indexWriter.commit();
    //关闭资源
    indexWriter.close();

}

 

通过以上两个步骤,添加文档document到lucene已经完成了分词,结果在上图中有展示,,下面我们将模拟用户索引查询分词库中的内容

 

-------------------------------------查询板块-------------------------------------------------------------------------------------------------

3 单个字段查询     QueryParse

//单字段查询
//QueryParse只用用于查询分词后的对象,不能查询不分词的,StringFiled查询不了.
@Test
public void indexSercher() throws Exception {
    //创建查询对象
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(FSDirectory.open(new File("f:\\test"))));
    //执行查询
    //要想得到query必须使用查询
    QueryParser queryParser=new QueryParser("pname",new IKAnalyzer());
    Query query = queryParser.parse("格力空调");
    TopDocs topDocs = indexSearcher.search(query, Integer.MAX_VALUE);
    int totalHits = topDocs.totalHits;//返回得分文档的总条数
    System.out.println(totalHits);
    ScoreDoc[] scoreDocs = topDocs.scoreDocs; //返回得分文档的数组
    for (ScoreDoc scoreDoc : scoreDocs) {
        float score = scoreDoc.score; //根据得分文档查询一条信息的分数
        int id = scoreDoc.doc; //根据得分文档查询自身id(这个id是lucene自己编写的从0开始)
        Document doc = indexSearcher.doc(id);//查询对象根据id查询到自身的文档doc
        //获取这个id对应文档doc的,根据doc获取内容
        //System.out.println(doc);
        String content = doc.get("pname");//获取一个document中的具体信息
        System.out.println("查询的总条数:"+totalHits+"----分数:"+score+"-----文档信息content:"+content);
    }
}

结果

 

4  多个字段查询   MultitFieldQueryParser

//多字段查询,本例中是使用title和pname两个字段查询,只要有一个字段里面有"格力空调",那就返回结果.
//多字段查询解析器 MultitFieldQueryParser
@Test
public void multiFieldQueryParseToLUcene() throws Exception {
    //查询对象
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(FSDirectory.open(new File("F:\\test"))));
    //执行查询
    MultiFieldQueryParser multiFieldQueryParser = new MultiFieldQueryParser(new String[]{"title","pname"},new IKAnalyzer());
    Query query = multiFieldQueryParser.parse("格力空调");
    TopDocs topDocs = indexSearcher.search(query, 20);
    //通过topDocs获取总条数,和总id
    int totalHits = topDocs.totalHits;//获取总条数
    ScoreDoc[] scoreDocs = topDocs.scoreDocs;//总分数的id
    for (ScoreDoc scoreDoc : scoreDocs) {
        float score = scoreDoc.score;//每条的分数
        int id = scoreDoc.doc;//每条的id(lucene自己编的)
        Document doc = indexSearcher.doc(id);//根据id获取具体文档
        System.out.println(doc);
    }
}

结果:

5   词条查询    TermQuery

//词条查询
//词条查询可以查询到分词和不分词的 TermQuery
@Test
public void TermQueryToLucene() throws IOException {
//获取查询对象IndexSERCHER
 IndexSearcher indexSearcher=new IndexSearcher(DirectoryReader.open(FSDirectory.open(new File("f:\\test"))));
//执行查询
TermQuery termQuery=new TermQuery(new Term("title","黑马咋了"));
    TopDocs topDocs = indexSearcher.search(termQuery, Integer.MAX_VALUE);
    int totalHits = topDocs.totalHits;//总条数
    ScoreDoc[] scoreDocs = topDocs.scoreDocs;
    for (ScoreDoc scoreDoc : scoreDocs) {
        float score = scoreDoc.score;//每条查询结果的得分
        int id = scoreDoc.doc;//每条查询结果的id(lucene自己编码)
        Document doc = indexSearcher.doc(id);
        System.out.println(doc);
        String title = doc.get("title");
        System.out.println("共查询到"+totalHits+"条数据.-----"+"title="+title);
    }
}

结果

 

6   通配符查询     WildcardQuery

//通配符查询 WildcardQuery
//*代表0-多个字符  ?代表1个占位符
@Test
public void wilCardQueryToLuene() throws IOException {
    //获取查询对象
    IndexSearcher indexSearcher=new IndexSearcher(DirectoryReader.open(FSDirectory.open(new File("f:\\test"))));
    //执行查询
    WildcardQuery wildcardQuery = new WildcardQuery(new Term("pname", "格力*"));
    TopDocs topDocs = indexSearcher.search(wildcardQuery, Integer.MAX_VALUE);
    int totalHits = topDocs.totalHits;//查询到的总条数
    ScoreDoc[] scoreDocs = topDocs.scoreDocs;//查询到的分数数组
    for (ScoreDoc scoreDoc : scoreDocs) {
        float score = scoreDoc.score;//具体一条查询结果的分数
        int id = scoreDoc.doc;//具体一条查询结果的id
        Document doc = indexSearcher.doc(id);//根据id获取具体的document
        System.out.println(doc);
        String title = doc.get("title");//通过document获取title
        System.out.println("共查询到"+totalHits+"条信息.---结果是:"+title);

    }
}

结果: 因为查询到的document中没有title属性,所以结果是null,如果将java代码中doc.get("title"),换成doc.get("content")就会有内容

 

7  模糊查询  FuzzyQuery

//模糊查询:FuzzyQuery  一个中文占两个字符,默认模糊匹配2个字符
@Test
public void fuzzyQueryToLusene() throws IOException {
    //获取查询对象
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(FSDirectory.open(new File("f:\\test"))));
    //执行查询
    FuzzyQuery fuzzyQuery = new FuzzyQuery(new Term("pname", "格力空"));
    TopDocs topDocs = indexSearcher.search(fuzzyQuery, Integer.MAX_VALUE);
    int totalHits = topDocs.totalHits; //获取总条数
    ScoreDoc[] scoreDocs = topDocs.scoreDocs; //获取查询结果的分数数组
    for (ScoreDoc scoreDoc : scoreDocs) {
        float score = scoreDoc.score;//获取一条查询结果的分数
        int id = scoreDoc.doc;//获取一条查询结果的id(lusene自己排,从0开始)
        //通过id获取doc
        Document doc = indexSearcher.doc(id);
        //System.out.println(doc);
        //获取doc中的内容
        String content = doc.get("content");
        System.out.println("共查询到"+totalHits+"条数据,---结果:"+content);
    }
}

查询结果:

 

8  数字查询   NumericRangeQuery

//数字范围查询 NumericRangeQuery
@Test
public void numQueryToLusene() throws IOException {
    //获取查询对象
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(FSDirectory.open(new File("f:\\test"))));
    //执行查询
    NumericRangeQuery<Integer> numerice = NumericRangeQuery.newIntRange("pid", 0, 1, true, true);//true表示包括,false表示不包括,这里的 意思是包括0,也包括1
    TopDocs topDocs = indexSearcher.search(numerice, 20);
    int totalHits = topDocs.totalHits;//查询到的总数据条数
    ScoreDoc[] scoreDocs = topDocs.scoreDocs;//查询到的分数数组
    for (ScoreDoc scoreDoc : scoreDocs) {
        float score = scoreDoc.score;//查询到的分数
        int id = scoreDoc.doc;//查询到结果的id
        Document doc = indexSearcher.doc(id);//根据id获取文档
        System.out.println(doc);
    }
}

结果:

9  组合查询  BooleanClauses

//组合查询 BooleanClauses  ,组合是各种的查询的交集
@Test
public void booleanQueryToLusene() throws IOException {
    //获取查询对象
    IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(FSDirectory.open(new File("f:\\test"))));
    //执行查询
    BooleanQuery booleanClauses = new BooleanQuery();
    //组合模糊查询,条件为必须
    FuzzyQuery fuzzyQuery = new FuzzyQuery(new Term("pname", "格力空"));
    //再组合一个数字范围的查询结果
    NumericRangeQuery<Integer> intRange = NumericRangeQuery.newIntRange("pid", 0, 2, true, false);
    //在组合查询中加入模糊查询和数字范围查询
    booleanClauses.add(fuzzyQuery, BooleanClause.Occur.MUST);//must表示必须有的
    booleanClauses.add(intRange, BooleanClause.Occur.MUST_NOT);//表示为必须没有
    TopDocs topDocs = indexSearcher.search(booleanClauses, 20);
    int totalHits = topDocs.totalHits;//获取查询总数
    ScoreDoc[] scoreDocs = topDocs.scoreDocs;//获取查询分数的数组
    for (ScoreDoc scoreDoc : scoreDocs) {
        float score = scoreDoc.score; //获取一条查询结果的分数
        int id = scoreDoc.doc;//获取一条查询结果的id;
        //根据id获取具体文档
        Document doc = indexSearcher.doc(id);
        System.out.println(doc);
    }
}

结果