Lucene是apache下的一个开放源代码的全文检索引擎工具包。提供了完整的查询引擎和索引引擎,部分文本分析引擎。Lucene的目的是为软件开发人员提供一个简单易用的工具包,以方便的在目标系统中实现全文检索的功能。

<The rest of contents | 余下全文>

全文检索

将非结构化数据中的一部分信息提取出来,重新组织,使其变得有一定结构,然后对此有一定结构的数据进行搜索,从而达到搜索相对较快的目的。这部分从非结构化数据中提取出的然后重新组织的信息,我们称之索引

例如:字典。字典的拼音表和部首检字表就相当于字典的索引,对每一个字的解释是非结构化的,如果字典没有音节表和部首检字表,在茫茫辞海中找一个字只能顺序扫描。然而字的某些信息可以提取出来进行结构化处理,比如读音,就比较结构化,分声母和韵母,分别只有几种可以一一列举,于是将读音拿出来按一定的顺序排列,每一项读音都指向此字的详细解释的页数。我们搜索时按结构化的拼音搜到读音,然后按其指向的页数,便可找到我们的非结构化数据——也即对字的解释。

这种先建立索引,再对索引进行搜索的过程就叫全文检索**(Full-text Search)**

虽然创建索引的过程也是非常耗时的,但是索引一旦创建就可以多次使用,全文检索主要处理的是查询,所以耗时间创建索引是值得的。对于数据量大、数据结构不固定的数据可采用全文检索方式搜索,比如百度、Google等搜索引擎、论坛站内搜索、电商网站站内搜索等。

Tip:lucene对java环境的最低要求为JAVA1.8

创建索引

第一步:创建一个java工程,并导入jar包。

工程所需jar包:

第二步:创建一个indexwriter对象。

1)指定索引库的存放位置Directory对象

2)指定一个IndexWriterConfig对象。

第二步:创建document对象。

第三步:创建field对象,将field添加到document对象中。

第四步:使用indexwriter对象将document对象写入索引库,此过程进行索引创建。并将索引和document对象写入索引库。

第五步:关闭IndexWriter对象。

代码如下:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
@Test
public void createIndex() throws Exception {
//创建一个Director对象,指定索引库保存的位置
Directory directory = FSDirectory.open(new File("c:\\temp\\index").toPath());
//基于directory对象创建一个IndexWriter对象
IndexWriter indexWriter = new IndexWriter(directory, new IndexWriterConfig());
//读取磁盘上的文件
File dir = new File("C:\\Users\\liu\\Desktop\\searchsource");
//原始文档的路径
File[] files = dir.listFiles();
for (File f : files) {
String name = f.getName();
String filePath = f.getPath();
String fileContent = FileUtils.readFileToString(f, "utf-8");
long fileSize = FileUtils.sizeOf(f);
//创建Field
//第一个参数:域的名称
//第二个参数:域的内容
//第三个参数:是否存储
Field fieldName = new TextField("name", name, Field.Store.YES);
//文件路径域(不分析、不索引、只存储)
Field fieldPath = new TextField("path", filePath, Field.Store.YES);
//文件内容域
Field fieldContent = new TextField("content", fileContent, Field.Store.YES);
//文件大小域
Field fieldSize = new TextField("size", fileSize + "", Field.Store.YES);
//创建文档对象
Document document = new Document();
document.add(fieldName);
document.add(fieldPath);
document.add(fieldContent);
document.add(fieldSize);
//把文档对象写入索引库
indexWriter.addDocument(document);

}
//关闭indexwriter
indexWriter.close();
}

createIndex方法运行后会在c:\\temp\\index文件夹下生成索引文件,如下图:

生成的索引文件为二进制文件,可以使用luke打开索引

查看索引

luck开始界面如下,在indexPath中选择索引存储位置文件夹即可

![](D:\OneDrive - business\myblog\source_posts\lucene操作实例\2019-06-03_232113.jpg)

查询索引

第一步:创建一个Directory对象,也就是索引库存放的位置。

第二步:创建一个indexReader对象,需要指定Directory对象。

第三步:创建一个indexsearcher对象,需要指定IndexReader对象

第四步:创建一个TermQuery对象,指定查询的域和查询的关键词。

第五步:执行查询。

第六步:返回查询结果。遍历查询结果并输出。

第七步:关闭IndexReader对象

代码如下:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
@Test
public void searchIndex() throws Exception {
//指定索引库存放的路径
Directory directory = FSDirectory.open(new File("c:\\temp\\index").toPath());
//创建indexReader对象
IndexReader indexReader = DirectoryReader.open(directory);
//创建indexsearcher对象
IndexSearcher indexSearcher = new IndexSearcher(indexReader);
//创建查询
Query query = new TermQuery(new Term("content", "spring"));

//执行查询
//第一个参数是查询对象,第二个参数是查询结果返回的最大值
TopDocs topDocs = indexSearcher.search(query, 10);
//查询结果的总条数
System.out.println("查询总记录数:" + topDocs.totalHits);
//遍历查询结果
//topDocs.scoreDocs存储了document对象的id

ScoreDoc[] scoreDocs = topDocs.scoreDocs;

for (ScoreDoc doc : scoreDocs) {
//scoreDoc.doc属性就是document对象的id
//根据document的id找到document对象

int docId = doc.doc;
Document document = indexSearcher.doc(docId);
System.out.println(document.get("name"));
System.out.println(document.get("path"));
System.out.println(document.get("size"));
System.out.println(document.get("content"));
System.out.println("--------------------------");
}
//关闭indexreader对象
indexReader.close();

}

部分查询结果如下图: