Lucene 多种搜索

lucene的搜索相当强大,它提供了很多辅助查询类,各自完成一种特殊的查询,也可以相互组合使用,来完成一些复杂的操作.





Java代码


public class Test{



Analyzer analyzer = new StandardAnalyzer();



RAMDirectory directory = new RAMDirectory();



/**

* 创建索引

*

* @throws IOException

*/

public void index() throws IOException{



IndexWriter indexWriter = new IndexWriter(directory,analyzer,true);



Document doc1 = new Document();



doc1.add(new Field("title","aaabbb",Store.YES,Index.TOKENIZED));



doc1.add(new Field("content","If you would like to help promote OpenOffice",Store.YES,Index.TOKENIZED));



doc1.add(new Field("time","2005",Store.YES,Index.TOKENIZED));



indexWriter.addDocument(doc1);



Document doc2 = new Document();



doc2.add(new Field("title","bbcc",Store.YES,Index.TOKENIZED));



doc2.add(new Field("content","sdfsdfsdfasdds",Store.YES,Index.TOKENIZED));



doc2.add(new Field("time","2007",Store.YES,Index.TOKENIZED));



indexWriter.addDocument(doc2);



indexWriter.optimize();



indexWriter.close();

}



// 按词条搜索

public void termSearcher() throws IOException{



IndexSearcher searcher = new IndexSearcher(directory);



// 查询title中包含aaa

Term term = new Term("title","aaa");



Query query = new TermQuery(term);



searcher.search(query);



searcher.close();

}



// 短语搜索

public void phraseSearcher() throws IOException{



IndexSearcher searcher = new IndexSearcher(directory);



PhraseQuery phraseQuery = new PhraseQuery();



// slop 两个项的位置之间允许的最大间隔,这里would和help中间只隔了like(to会被去掉),所以最大间隔设为1就能找到值

// 对于两个相连的关键词来说,无论坡度设为多少,都可以找到

// 对于不相连的词来说,当两个关键词相隔数小于坡度都可以找到,否则找不到

phraseQuery.setSlop(1);



phraseQuery.add(new Term("content","would"));



phraseQuery.add(new Term("content","help"));



// 如果将help放前面,would放后面,需要将would向后移动3个位置才能到help后面(to不算),所以要设为slop最少要设为3

// phraseQuery.add(new Term("content","help"));



// phraseQuery.add(new Term("content","would"));



// phraseQuery.setSlop(3);



// 短语的评分是根据项之间距离越小,评分越高,否则越小

Hits hits = searcher.search(phraseQuery);



printResult(hits);



searcher.close();

}



// 通配符搜索 WildcardQuery

// 通配符包括’?’匹配一个任意字符和’*’匹配零个或多个任意字符，例如你搜索’use*’，你可能找到’user’或者’uses’：

public void wildcardSearcher() throws IOException{



IndexSearcher searcher = new IndexSearcher(directory);



// 与正则一样,*代表0个或多个字母,?代表0个或一个字母

// WildcardQuery与QueryParser不同的是:WildcardQuery的前缀可以为*,而QueryParser不行

WildcardQuery query = new WildcardQuery(new Term("content","a?bbb*"));



Hits hits = searcher.search(query);



printResult(hits);



searcher.close();

}



// 模糊搜索 FuzzyQuery

public void fuzzySearcher() throws IOException{



IndexSearcher search = new IndexSearcher(directory);



// OpenOffica虽然没被索引,但能找到相近的OpenOffice

FuzzyQuery query = new FuzzyQuery(new Term("content","OpenOffica"));



Hits hits = search.search(query);



printResult(hits);



search.close();

}



// 使用前缀PrefixQuery

public void prefixSearcher() throws IOException{



IndexSearcher search = new IndexSearcher(directory);



// 全部title前缀为a

PrefixQuery query = new PrefixQuery(new Term("title","b"));



Hits hits = search.search(query);



printResult(hits);



search.close();



}



// 范围搜索 RangeQuery

public void rangeSearcher() throws IOException{



IndexSearcher search = new IndexSearcher(directory);



// RangeQuery query = new RangeQuery(beginTime, endTime, false);

// 开始时间,结束时间,最后一个参数表示是否包含边界条件本身,如果为false

RangeQuery query = new RangeQuery(new Term("time","2005"),new Term("time","2007"),true);



Hits hits = search.search(query);



printResult(hits);



search.close();

}





// 与或搜索BooleanQuery

//BooleanClause用于表示布尔查询子句关系的类，包括：BooleanClause.Occur.MUST，BooleanClause.Occur.MUST_NOT，BooleanClause.Occur.SHOULD。有以下6种组合：



//1．MUST和MUST：取得连个查询子句的交集。



//2．MUST和MUST_NOT：表示查询结果中不能包含MUST_NOT所对应得查询子句的检索结果。



//3．MUST_NOT和MUST_NOT：无意义，检索无结果。



//4．SHOULD与MUST、SHOULD与MUST_NOT：SHOULD与MUST连用时，无意义，结果为MUST子句的检索结果。与MUST_NOT连用时，功能同MUST。



//5．SHOULD与SHOULD：表示“或”关系，最终检索结果为所有检索子句的并集。



public void booleanSearcher() throws IOException, ParseException{



IndexSearcher search = new IndexSearcher(directory);



QueryParser qp1 = new QueryParser("title",new StandardAnalyzer());



Query query1 = qp1.parse("aa*");



QueryParser qp2 = new QueryParser("title",new StandardAnalyzer());



Query query2 = qp2.parse("bb*");



BooleanQuery query = new BooleanQuery();



// 搜索结果的title的前双缀可以是aa,或bb

query.add(query1, BooleanClause.Occur.SHOULD);



// BooleanClause.Occur.MUST 必须

// BooleanClause.Occur.MUST_NOT 必须不是

query.add(query2, BooleanClause.Occur.SHOULD);



Hits hits = search.search(query);



printResult(hits);



search.close();



}





// 多关键的搜索 PhrasePrefixQuery

public void phrasePrefixSearcher() throws IOException{



IndexSearcher search = new IndexSearcher(directory);



PhrasePrefixQuery query = new PhrasePrefixQuery();



// 这里两项都有可能首先被匹配

query.add(new Term[]{new Term("content","would"),new Term("content","can")});



// 只有一项必须匹配

query.add(new Term("content","help"));



// If you would like to help promote OpenOffice

// can I help you

// slop因子的作用域为查询中的所有短语

query.setSlop(1);



// 匹配第一项为 would 或 can 第二项为help

// solp设置为1

// If you would like to help promote OpenOffice 除去if to 外,would与help的距离=1

// can I help you 的距离也=1 所以可以搜索出两条数据



Hits hits = search.search(query);



printResult(hits);



search.close();

}



// 在多个域上查询 MultiFieldQueryParser

public void multiFieldSearcher() throws IOException, ParseException{



IndexSearcher search = new IndexSearcher(directory);



// 默认情况下的方式为Occur.SHOULD

// titile可以匹配bb,content可以匹配you

// MultiFieldQueryParser.parse(new String[]{"bb","you"},new String[]{"title","content"}, analyzer);



// titile必须匹配bb,content不能匹配

Query query = MultiFieldQueryParser.parse( new String[]{"bb","you"},new String[]{"title","content"},new BooleanClause.Occur[]{Occur.MUST,Occur.MUST_NOT}, analyzer);



// title中必须包含bb content不能有bb

// Query query = MultiFieldQueryParser.parse( "bb*",new String[]{"title","content"},new BooleanClause.Occur[]{BooleanClause.Occur.MUST,BooleanClause.Occur.MUST_NOT}, analyzer);



Hits hits = search.search(query);



printResult(hits);



search.close();

}



public void printResult(Hits hits) throws IOException{

for(int i = 0; i < hits.length(); i++){

Document d = hits.doc(i);

System.out.println(d.get("title"));

System.out.println(d.get("content"));

System.out.println(d.get("time"));

}

}





}