lucene in action 第三章（2)

lucene的这种各种各样的查询类型

1、TermQuery 

   最简单的Query类型，某一个field是否含有一个term的value

2、TermRangeQuery 

   由于term在index中是按照字典顺序排列的，可以使用TermRangeQuery查询一个范围内的Term

例如

Queryquery = new TermRangeQuery("city", "aa", "am", true,true);

TopDocs hits = searcher.search(query,20);

可以查血从aa* ab* ..... am*的term。 后面的true和false代表是否包括aa和am

3、NumericRangeQuery 

查询一个数值的范围。  这个必须查血NumericFiled

Query query =NumericRangeQuery.newIntRange("intID", from, to,true,true);

TopDocs hits = searcher.search(query, 20);

4、PrefixQuery 前缀查询

    查询一个term是否满足一个前缀。

    比如 prefix=“bri ”bridge和“bright”都可以满足

Term t = new Term(field, prefix);

Query query = new PrefixQuery(t);

TopDocs hits = searcher.search(query, 20);

5、BooleanQuery联合多个查找

Term t= new Term("contents", "bri");

Queryquery1 = new PrefixQuery(t);

Queryquery2 = NumericRangeQuery.newIntRange("intID", 1, 3, true,true);

//create a boolean query

BooleanQuery query = newBooleanQuery();

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

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

TopDocs hits = searcher.search(query,20);

注意BooleanClause.Occur.MUST是and的意思，BooleanClause.Occur.SHOULD是or的意思，BooleanClause.Occur.MUST_NOT是not的意思

6、PhraseQuery短语查询

     我们想查询一个短语 fox quick 或者 quick fox 或者quick brown fox，或者quickred fox。

     可以使用phraseQuery, PhraseQuery使用Edit distance（ 编辑距离 ） 来量度，编辑距离是一个字符串变化到另一个字符串需要的替换，删除，插入的次数总和。每一次这种操作叫做一次slop。可以使用 setSlop来限制短语slop的最大值。

edit distance如下图

   

比如： quick fox 到quick [xxx] fox 需要 1slop

fox quick 到 quick[xxx] fox 需要 3 slop 先用quick替换 fox，再用fox替换quick，再插入一个xxx总共3次。

    PhraseQuery query = newPhraseQuery();

// set max slop to 10

query.setSlop(10);

query.add(new Term("contents","  quick   "   ));

query.add(new Term("contents","  fox "));

TopDocs hits =searcher.search(query, 20);

7、WildcardQuery通配符查询

    PrefixQuery是WildcardQuery的特殊形式

   *代表一个或者多个，？代表0个或者一个

                                //use wildchard "?ridg*"

WildcardQuery query = newWildcardQuery(new Term("contents", "?ridg*"));

TopDocs hits = searcher.search(query,20);

8、FuzzyQuery  模糊查询

    FuzzyQuery与PhraseQury 一类似都是以Edit distance 来做的，只不过FuzzyQuery是在term内部，而PhraseQuery是在term之间。     

  例如      FuzzyQuery query = newFuzzyQuery(new Term("contents", " Amsteedam"));可以查出  Amsterdam,他们之间的编辑距离是1。

如下

 IndexSearchersearcher = new IndexSearcher(dir);

// "Amsterdam" is similar to"Amsteedam"

FuzzyQuery query = newFuzzyQuery(new Term("contents", "Amsteedam"));

TopDocs hits =searcher.search(query, 20);

showResult(hits,searcher);

packagecharpter3;

importjava.io.File;

importjava.io.IOException;

importorg.apache.lucene.analysis.standard.StandardAnalyzer;

importorg.apache.lucene.document.Document;

importorg.apache.lucene.document.Field;

importorg.apache.lucene.document.Field.TermVector;

importorg.apache.lucene.document.NumericField;

importorg.apache.lucene.index.CorruptIndexException;

importorg.apache.lucene.index.IndexReader;

importorg.apache.lucene.index.IndexWriter;

importorg.apache.lucene.index.Term;

importorg.apache.lucene.queryParser.ParseException;

importorg.apache.lucene.queryParser.QueryParser;

importorg.apache.lucene.search.BooleanClause;

importorg.apache.lucene.search.BooleanQuery;

importorg.apache.lucene.search.FuzzyQuery;

importorg.apache.lucene.search.IndexSearcher;

importorg.apache.lucene.search.NumericRangeQuery;

importorg.apache.lucene.search.PhraseQuery;

importorg.apache.lucene.search.PrefixQuery;

importorg.apache.lucene.search.Query;

importorg.apache.lucene.search.ScoreDoc;

importorg.apache.lucene.search.TermQuery;

importorg.apache.lucene.search.TermRangeQuery;

importorg.apache.lucene.search.TopDocs;

importorg.apache.lucene.search.WildcardQuery;

importorg.apache.lucene.store.Directory;

importorg.apache.lucene.store.FSDirectory;

importorg.apache.lucene.util.Version;

public classQuerys {

privateIndexWriter writer;

protectedString[] ids = { "1", "2", "3" };

protectedString[] unindexed = { "Netherlands", "Italy", "China"};

protectedString[] unstored = { "Amsterdam has a lot ofbridge",

"Venice haslots of canals", "Amsterddam bridges are a lot"};

protectedString[] text = { "Amsterdam", "Venice", "Aeijing"};

privateDirectory dir = null;

privateIndexReader indexReader = null;

publicQuerys(String indexDir) throws IOException {

dir =FSDirectory.open(new File(indexDir));

this.writer =new IndexWriter(dir, new StandardAnalyzer(

Version.LUCENE_36), true,IndexWriter.MaxFieldLength.UNLIMITED);

this.writer.setInfoStream(System.out);

// create aindex reader instance

indexReader =IndexReader.open(dir);

}

public voidaddDocuments() throws CorruptIndexException, IOException{

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

Document doc =new Document();

NumericFieldnfield = new NumericField("intID", 10);

nfield.setIntValue(i);

doc.add(nfield);

doc.add(newField("id", ids[i], Field.Store.YES,

Field.Index.NOT_ANALYZED));

doc.add(newField("country", unindexed[i], Field.Store.YES,

Field.Index.NO));

doc.add(newField("contents", unstored[i], Field.Store.YES,

Field.Index.ANALYZED));

doc.add(newField("city", text[i], Field.Store.YES,

Field.Index.ANALYZED));

writer.addDocument(doc);

}

System.out.println("docs = " +writer.numDocs());

}

public voidindex() throws CorruptIndexException, IOException{

this.addDocuments();

this.commit();

}

public voidexpressionQuery() throws CorruptIndexException,IOException,

ParseException{

IndexSearchersearcher = new IndexSearcher(this.indexReader);

QueryParserpraser = newQueryParser(Version.LUCENE_CURRENT,

"contents",new StandardAnalyzer(Version.LUCENE_CURRENT));

//note

Query query =praser.parse("+bridge -Amsterdam");

System.out.println("query = " +query.toString());

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidtermQuery(String fieldName, String q)

throwsCorruptIndexException, IOException, ParseException{

//IndexSearcher searcher = newIndexSearcher(dir);

// build aindexSearch on a indexReader

IndexSearchersearcher = new IndexSearcher(this.indexReader);

Term t = newTerm(fieldName, q.toLowerCase());

Query query =new TermQuery(t);

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidtermRangeQuery(String fieldName, String q)

throwsCorruptIndexException, IOException, ParseException{

IndexSearchersearcher = new IndexSearcher(dir);

Query query =new TermRangeQuery("city", "aa", "am", true,true);

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidnumericRangeQuery(int from, int to)

throwsCorruptIndexException, IOException, ParseException{

IndexSearchersearcher = new IndexSearcher(dir);

Query query =NumericRangeQuery.newIntRange("intID", from, to,true,true);

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidprefixQuery(String field, String prefix)

throwsCorruptIndexException, IOException, ParseException{

IndexSearchersearcher = new IndexSearcher(dir);

Term t = newTerm(field, prefix);

Query query =new PrefixQuery(t);

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidbooleanQuery() throws CorruptIndexException,IOException,

ParseException{

IndexSearchersearcher = new IndexSearcher(dir);

Term t = newTerm("contents", "bri");

Query query1 =new PrefixQuery(t);

Query query2 =NumericRangeQuery.newIntRange("intID", 1, 3, true,true);

// create aboolean query

BooleanQueryquery = new BooleanQuery();

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

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

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidphraseQuery() throws CorruptIndexException,IOException,

ParseException{

IndexSearchersearcher = new IndexSearcher(dir);

PhraseQueryquery = new PhraseQuery();

// set maxslop to 10

query.setSlop(10);

query.add(newTerm("contents", "lot"));

query.add(newTerm("contents", "bridges"));

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidwildCardQuery() throws CorruptIndexException,IOException,

ParseException{

IndexSearchersearcher = new IndexSearcher(dir);

// usewildchard "?ridg*"

WildcardQueryquery = new WildcardQuery(new Term("contents","?ridg*"));

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidfuzzyQuery() throws CorruptIndexException,IOException,

ParseException{

IndexSearchersearcher = new IndexSearcher(dir);

// "Amsterdam"is similar to "Amsteedam"

FuzzyQueryquery = new FuzzyQuery(new Term("contents","Amsteedam"));

TopDocs hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidtestReopen() throws ParseException, IOException{

IndexSearchersearcher = new IndexSearcher(this.indexReader);

QueryParserpraser = newQueryParser(Version.LUCENE_CURRENT,

"contents",new StandardAnalyzer(Version.LUCENE_CURRENT));

//note

Query query =praser.parse("+bridge -Amsterdam");

System.out.println("query = " +query.toString());

TopDocs hits =searcher.search(query, 20);

// reopen aindex and will cover current modification ofindex.

IndexReadernewReader = indexReader.reopen();

if(indexReader != newReader) {

indexReader =newReader;

// ifindexReader is changed , searcher must beconstructed.

searcher.close();

searcher =null;

searcher = newIndexSearcher(this.indexReader);

}

hits =searcher.search(query, 20);

showResult(hits, searcher);

}

public voidtestTopDocs() throws CorruptIndexException, IOException{

IndexSearchersearcher = new IndexSearcher(dir);

// "Amsterdam"is similar to "Amsteedam"

FuzzyQueryquery = new FuzzyQuery(new Term("contents","Amsteedam"));

TopDocs hits =searcher.search(query, 20);

System.out.println("search result:");

for (ScoreDocdoc : hits.scoreDocs) {

//閸欐牕绶遍崨鎴掕厬閻ㄥ嫭鏋冨锟�

Document d =searcher.doc(doc.doc);

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

}

}

public voidcommit() throws CorruptIndexException, IOException{

this.writer.commit();

}

public voidshowResult(TopDocs hits, IndexSearcher searcher){

try{

System.out.println("search result:");

for (ScoreDocdoc : hits.scoreDocs) {

//閸欐牕绶遍崨鎴掕厬閻ㄥ嫭鏋冨锟�

Document d =searcher.doc(doc.doc);

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

}

} catch(Exception e) {

e.printStackTrace();

}

}

public staticvoid main(String[] args) throws IOException, ParseException{

// TODOAuto-generated method stub

Querys ci =new Querys("charpter2-1");

ci.index();

System.out.println("----------termQuery--------------");

ci.termQuery("city", "Venice");

System.out.println("----------termRangeQuery--------------");

ci.termRangeQuery(null, null);

System.out.println("----------numericRangeQuery--------------");

ci.numericRangeQuery(1, 5);

System.out.println("----------prefixQuery--------------");

ci.prefixQuery("contents", "bri");

System.out.println("----------booleanQuery--------------");

ci.booleanQuery();

System.out.println("----------phraseQuery--------------");

ci.phraseQuery();

System.out.println("----------wildCardQuery--------------");

ci.wildCardQuery();

System.out.println("----------fuzzyQuery--------------");

ci.fuzzyQuery();

System.out.println("----------expressionQuery--------------");

ci.expressionQuery();

System.out.println("----------testreopen--------------");

ci.testReopen();

}

}