lucene创建索引 IKAnalyzer中文分词器案列

素颜马尾好姑娘i 2023-10-18 14:42 20阅读 0赞

1.下载需要的jar包和配置，github资源链接：[https://github.com/zhangliqingyun/jarlist/tree/master/lucene][https_github.com_zhangliqingyun_jarlist_tree_master_lucene]

2.创建一个目录索引

//创建一个文件索引

@Test

**public** **void** addFileDirectory() **throws** Exception\{

//创建一个索引目录

Directory directory = FSDirectory.*open*(**new** File("./myindex6"));

//创建一个分词器

IKAnalyzer analyzer = **new** IKAnalyzer();

//创建一个indexWriter

IndexWriter indexWriter = **new** IndexWriter(directory,analyzer,MaxFieldLength.*LIMITED*);

//创建一个文档

Document document = **new** Document();

//创建一个实体类

People people = **new** People();

//向实体类中添加字段数据

people.setId(9);

people.setName("美女真多啊fjhtjhtfgsjtsj方式都会认同一天我要提问");

people.setTitle("这是标题");

//向文档中添加字段数据

Field fieldId = **new** Field("id",people.getId().toString(),Store.*YES*,Index.*ANALYZED*);

Field fieldName = **new** Field("name",people.getName(),Store.*YES*,Index.*ANALYZED*);

Field fieldTitle = **new** Field("title",people.getTitle(),Store.*YES*,Index.*ANALYZED*);

//向文档中添加字段

document.add(fieldId);

document.add(fieldName);

document.add(fieldTitle);

//设置添加文档的相关度更大

document.setBoost(100f);

//向indexWriter中添加文档

indexWriter.addDocument(document);

//关闭indexWriter

indexWriter.close();

3.查询目录索引：

//查询文档索引

@Test

**public** **void** searchIndex() **throws** Exception\{

//定义一个文档索引路径

Directory directory = FSDirectory.*open*(**new** File("./myindex"));

//创建IndexSearcher

IndexSearcher indexSearcher = **new** IndexSearcher(directory);

//定义一个分词器

Analyzer analyzer = **new** StandardAnalyzer(Version.*LUCENE\_30*);

//定义一个查询器

QueryParser queryParser = **new** MultiFieldQueryParser(Version.*LUCENE\_30*, **new** String\[\]\{ "title","name"\}, analyzer);

//给查询条件赋值

Query query = queryParser.parse("lucene");

//向indexSearcher中添加查询条件

TopDocs topDocs = indexSearcher.search(query, 20);

//得到文档集合

ScoreDoc\[\] scoreDocs = topDocs.scoreDocs;

//保存查询到的实体类集合

List<People> list = **new** ArrayList<People>();

//遍历文档集合

**for**(ScoreDoc scoreDoc : scoreDocs)\{

**int** index = scoreDoc.doc;

Document document = indexSearcher.doc(index);

People people = **new** People();

people.setId(Integer.*parseInt*(document.get("id")));

people.setName(document.get("name"));

people.setTitle(document.get("title"));

list.add(people);

//打印输出查询到的结果集

**for**(**int** i = 0;i < list.size();i++)\{

System.*out*.println(list.get(i).getId());

System.*out*.println(list.get(i).getName());

System.*out*.println(list.get(i).getTitle());

4.使用term查询目录索引：

//使用Term查询

@Test

**public** **void** TermQuery() **throws** Exception\{

//查询条件

Term term = **new** Term("name","lucene");

Query query = **new** TermQuery(term);

//定义一个文档索引路径

Directory directory = FSDirectory.*open*(**new** File("./myindex"));

//创建IndexSearcher

IndexSearcher indexSearcher = **new** IndexSearcher(directory);

//向indexSearcher中添加查询条件

TopDocs topDocs = indexSearcher.search(query, 20);

//得到文档集合

ScoreDoc\[\] scoreDocs = topDocs.scoreDocs;

//保存查询到的实体类集合

List<People> list = **new** ArrayList<People>();

//遍历文档集合

**for**(ScoreDoc scoreDoc : scoreDocs)\{

**int** index = scoreDoc.doc;

Document document = indexSearcher.doc(index);

People people = **new** People();

people.setId(Integer.*parseInt*(document.get("id")));

people.setName(document.get("name"));

people.setTitle(document.get("title"));

list.add(people);

//打印输出查询到的结果集

**for**(**int** i = 0;i < list.size();i++)\{

System.*out*.println(list.get(i).getId());

System.*out*.println(list.get(i).getName());

System.*out*.println(list.get(i).getTitle());

5.查询所有的目录索引

//查询所有的文档

@Test

**public** **void** queryAllDocs() **throws** Exception\{

//查询所有的文档的query

Query query = **new** MatchAllDocsQuery();

//定义一个文档索引路径

Directory directory = FSDirectory.*open*(**new** File("./myindex"));

//创建IndexSearcher

IndexSearcher indexSearcher = **new** IndexSearcher(directory);

//向indexSearcher中添加查询条件

TopDocs topDocs = indexSearcher.search(query, 20);

//得到文档集合

ScoreDoc\[\] scoreDocs = topDocs.scoreDocs;

//保存查询到的实体类集合

List<People> list = **new** ArrayList<People>();

//遍历文档集合

**for**(ScoreDoc scoreDoc : scoreDocs)\{

**int** index = scoreDoc.doc;

Document document = indexSearcher.doc(index);

People people = **new** People();

people.setId(Integer.*parseInt*(document.get("id")));

people.setName(document.get("name"));

people.setTitle(document.get("title"));

list.add(people);

//打印输出查询到的结果集

**for**(**int** i = 0;i < list.size();i++)\{

System.*out*.println(list.get(i).getId());

System.*out*.println(list.get(i).getName());

System.*out*.println(list.get(i).getTitle());

6.按照范围查询目录索引：

//按照范围查询

@Test

**public** **void** queryRange() **throws** Exception\{

//按照范围查询

Query query = NumericRangeQuery.*newLongRange*("id", 1L, 3L, **true**, **true**);

//定义一个文档索引路径

Directory directory = FSDirectory.*open*(**new** File("./myindex2"));

//创建IndexSearcher

IndexSearcher indexSearcher = **new** IndexSearcher(directory);

//向indexSearcher中添加查询条件

TopDocs topDocs = indexSearcher.search(query, 20);

//得到文档集合

ScoreDoc\[\] scoreDocs = topDocs.scoreDocs;

//保存查询到的实体类集合

List<People> list = **new** ArrayList<People>();

//遍历文档集合

**for**(ScoreDoc scoreDoc : scoreDocs)\{

**int** index = scoreDoc.doc;

Document document = indexSearcher.doc(index);

People people = **new** People();

people.setId(NumericUtils.*prefixCodedToInt*(document.get("id")));

people.setName(document.get("name"));

people.setTitle(document.get("title"));

list.add(people);

//打印输出查询到的结果集

**for**(**int** i = 0;i < list.size();i++)\{

System.*out*.println(list.get(i).getId());

System.*out*.println(list.get(i).getName());

System.*out*.println(list.get(i).getTitle());

7.使用boolean查询目录索引：

//通配符查询

@Test

**public** **void** wildcardQuery() **throws** Exception\{

//组织查询语句

Term term = **new** Term("name","l\*");

//按照范围查询

Query query = **new** WildcardQuery(term);

//定义一个文档索引路径

Directory directory = FSDirectory.*open*(**new** File("./myindex3"));

//创建IndexSearcher

IndexSearcher indexSearcher = **new** IndexSearcher(directory);

//向indexSearcher中添加查询条件

TopDocs topDocs = indexSearcher.search(query, 20);

//得到文档集合

ScoreDoc\[\] scoreDocs = topDocs.scoreDocs;

//保存查询到的实体类集合

List<People> list = **new** ArrayList<People>();

//遍历文档集合

**for**(ScoreDoc scoreDoc : scoreDocs)\{

**int** index = scoreDoc.doc;

Document document = indexSearcher.doc(index);

People people = **new** People();

people.setId(Integer.*parseInt*(document.get("id")));

people.setName(document.get("name"));

people.setTitle(document.get("title"));

list.add(people);

//打印输出查询到的结果集

**for**(**int** i = 0;i < list.size();i++)\{

System.*out*.println(list.get(i).getId());

System.*out*.println(list.get(i).getName());

System.*out*.println(list.get(i).getTitle());

8.高亮显示查询：

//高亮查询

@Test

**public** **void** highLighterQuery() **throws** Exception\{

//组织查询语句

Term term = **new** Term("name","北京");

TermQuery termQuery = **new** TermQuery(term);

Term term2 = **new** Term("name","美女");

TermQuery termQuery2 = **new** TermQuery(term2);

Term term3 = **new** Term("name","北京美女");

TermQuery termQuery3 = **new** TermQuery(term3);

BooleanQuery booleanQuery = **new** BooleanQuery();

booleanQuery.add(termQuery, Occur.*SHOULD*);

booleanQuery.add(termQuery2,Occur.*SHOULD*);

booleanQuery.add(termQuery3,Occur.*SHOULD*);

//索引查询的路径

Directory directory = FSDirectory.*open*(**new** File("./myindex6"));

//创建一个indexSearcher

IndexSearcher indexSearcher = **new** IndexSearcher(directory);

//设置高亮显示的样式

Formatter formatter = **new** SimpleHTMLFormatter("<font color='red'>","</font>");

//设置一个结果查询

Scorer scorer = **new** QueryScorer(booleanQuery);

//创建一个高亮器

Highlighter highLighter = **new** Highlighter(formatter, scorer);

//搜索的摘要

Fragmenter fragmenter = **new** SimpleFragmenter(10);

//设置高亮摘要

highLighter.setTextFragmenter(fragmenter);

//添加查询条件

TopDocs topDocs = indexSearcher.search(booleanQuery, 20);

System.*out*.println(topDocs.totalHits);

//赋值到结果集中

ScoreDoc\[\] scoreDocs = topDocs.scoreDocs;

//保存people的结果集

List<People> list = **new** ArrayList<People>();

//遍历结果集

**for**(ScoreDoc scoreDoc : scoreDocs)\{

**int** index = scoreDoc.doc;

System.*out*.println("相关度得分:"\+scoreDoc.score);

Document document = indexSearcher.doc(index);

People people = **new** People();

people.setId(Integer.*parseInt*(document.get("id")));

people.setName(document.get("name"));

people.setTitle(document.get("title"));

Analyzer analyzer = **new** IKAnalyzer();

String name = highLighter.getBestFragment(analyzer, "name", people.getName());

people.setName(name);

list.add(people);

//遍历打印输出list中存储的值

**for**(**int** i = 0;i < list.size();i++)\{

System.*out*.println(list.get(i).getId());

System.*out*.println(list.get(i).getName());

System.*out*.println(list.get(i).getTitle());

indexSearcher.close();

9.创建内存索引

//创建一个内存索引库

@Test

**public** **void** addMemoryIndex() **throws** Exception\{

//创建一个内存索引库目录

Directory ramDirectory = **new** RAMDirectory();

//创建一个分词器

Analyzer analyzer = **new** StandardAnalyzer(Version.*LUCENE\_30*);

//创建一个IndexWriter

IndexWriter indexWriter = **new** IndexWriter(ramDirectory, analyzer, MaxFieldLength.*LIMITED*);

//创建一个实体类people

People people = **new** People();

//给实体类people添加字段值

people.setId(1);

people.setAge(24);

people.setName("lucene是一个搜搜服务器");

//创建一个添加文档

Document document = **new** Document();

//创建文档的添加字段

Field fieldId = **new** Field("id",people.getId().toString(),Store.*YES*,Index.*ANALYZED*);

Field fieldAge = **new** Field("age",people.getAge().toString(),Store.*YES*,Index.*ANALYZED*);

Field fieldName = **new** Field("name",people.getName(),Store.*YES*,Index.*ANALYZED*);

document.add(fieldId);

document.add(fieldAge);

document.add(fieldName);

indexWriter.addDocument(document);

indexWriter.close();

//调用查询索引的方法

searchIndex(ramDirectory);

10.删除目录索引：

//删除索引

@Test

**public** **void** deleteIndex() **throws** Exception\{

//创建一个索引目录

Directory directory = FSDirectory.*open*(**new** File("./myindex5"));

//创建一个分词器

Analyzer analyzer = **new** StandardAnalyzer(Version.*LUCENE\_30*);

//创建一个indexwriter

IndexWriter indexWriter = **new** IndexWriter(directory,analyzer,MaxFieldLength.*LIMITED*);

//组织删除的条件语句

Term term = **new** Term("name","hello");

indexWriter.deleteDocuments(term);

indexWriter.close();

11.更新目录索引（先删除，后增加）：

//更新索引

@Test

**public** **void** updateIndex() **throws** Exception\{

//创建一个索引目录

Directory directory = FSDirectory.*open*(**new** File("./myindex5"));

//创建一个分词器

Analyzer analyzer = **new** StandardAnalyzer(Version.*LUCENE\_30*);

//创建一个indexwriter

IndexWriter indexWriter = **new** IndexWriter(directory,analyzer,MaxFieldLength.*LIMITED*);

//要更新的查询条件

Term term = **new** Term("name","hello");

//创建一个实体类对象

People people = **new** People();

//给实体对象赋值

people.setAge(24);

people.setId(2);

people.setName("这是更新后的hello");

//更新为的新文档

Document document = **new** Document();

//给文档赋值

Field fieldId = **new** Field("id",people.getId().toString(),Store.*YES*,Index.*ANALYZED*);

Field fieldAge = **new** Field("age",people.getAge().toString(),Store.*YES*,Index.*ANALYZED*);

Field fieldName = **new** Field("name",people.getName(),Store.*YES*,Index.*ANALYZED*);

//把字段添加到文档中

document.add(fieldId);

document.add(fieldAge);

document.add(fieldName);

//执行更新操作

indexWriter.updateDocument(term, document);

//关闭indexWriter

indexWriter.close();

12.优化目录索引（合并相同项）：

//优化（合并相同的数据）

@Test

**public** **void** optimize() **throws** Exception\{

//操作索引的目录

Directory directory = FSDirectory.*open*(**new** File("./myindex5"));

//创建一个分词器

Analyzer analyzer = **new** StandardAnalyzer(Version.*LUCENE\_30*);

//创建一个indexWriter

IndexWriter indexWriter = **new** IndexWriter(directory,analyzer,MaxFieldLength.*LIMITED*);

//优化indexWriter

indexWriter.optimize();

indexWriter.close();

13.中分分词器：

①导jar包：

![20190221154100825.png][]

②导入配置文件：

![20190221154114978.png][]

③配置IKAnalyzer.cfg.xml内容

![2019022115413022.png][]

[https_github.com_zhangliqingyun_jarlist_tree_master_lucene]: https://github.com/zhangliqingyun/jarlist/tree/master/lucene
[20190221154100825.png]: https://img-blog.csdnimg.cn/20190221154100825.png
[20190221154114978.png]: https://img-blog.csdnimg.cn/20190221154114978.png
[2019022115413022.png]: https://img-blog.csdnimg.cn/2019022115413022.png