Lucene 自定义分析器

最新推荐文章于 2021-11-03 14:58:55 发布

weixin_30793643

最新推荐文章于 2021-11-03 14:58:55 发布

阅读量122

点赞数

文章标签： java

原文链接：http://www.cnblogs.com/song-yajun/p/3878781.html

版权

自定义一个NGramAnalyzer:

  1 package org.apache.lucene.analysis.ngram;
  2 
  3 import java.io.BufferedReader;
  4 import java.io.File;
  5 import java.io.FileReader;
  6 import java.io.IOException;
  7 import java.io.Reader;
  8 import java.util.ArrayList;
  9 import java.util.List;
 10 
 11 import org.apache.lucene.analysis.Analyzer;
 12 import org.apache.lucene.analysis.TokenStream;
 13 import org.apache.lucene.analysis.Tokenizer;
 14 import org.apache.lucene.analysis.Analyzer.TokenStreamComponents;
 15 import org.apache.lucene.analysis.core.StopFilter;
 16 import org.apache.lucene.analysis.core.WhitespaceTokenizer;
 17 import org.apache.lucene.analysis.standard.StandardFilter;
 18 import org.apache.lucene.analysis.util.CharArraySet;
 19 import org.apache.lucene.util.Version;
 20 
 21 public class NGramAnalyzer extends Analyzer{
 22     
 23     private Version matchVersion ;
 24     
 25     private int gramSize ;
 26     
 27     private CharArraySet stopwordSet ;
 28     
 29     public NGramAnalyzer(){
 30         
 31     }
 32     
 33     public NGramAnalyzer(Version matchVersion ,int gramSize) {
 34         
 35         this.matchVersion = matchVersion ;
 36         
 37         this.gramSize = gramSize ;
 38 
 39         try {
 40             
 41             List<String> stopWords;
 42             
 43             String stopWordPath = "E://Java program//PatentRetrival//resouce//stopwords.txt";
 44             
 45             stopWords = readStopwords(stopWordPath);
 46             
 47             this.stopwordSet = new CharArraySet(Version.LUCENE_4_9 ,stopWords ,true);
 48             
 49         } catch (IOException e) {
 50             // TODO Auto-generated catch block
 51             e.printStackTrace();
 52         }
 53         
 54         
 55         
 56         
 57     }
 58     
 59     
 60 
 61     @Override
 62     protected TokenStreamComponents createComponents(String fieldName,Reader reader) {
 63         
 64         // TODO Auto-generated method stub
 65         
 66          Tokenizer tokenizer = new NGramTokenizer(Version.LUCENE_4_9, reader,gramSize ,gramSize); 
 67          
 68          TokenStream tokenStream = new NGramTokenFilter(matchVersion, tokenizer);
 69          
 70             //tok = new LowerCaseFilter(matchVersion, tok);
 71          tokenStream = new StopFilter(matchVersion, tokenStream, stopwordSet);
 72          return new TokenStreamComponents(tokenizer ,tokenStream);
 73         
 74     }
 75     
 76     public  List<String> readStopwords(String filepath) throws IOException{
 77         
 78         List<String> stopwords = new ArrayList<String>() ;
 79         
 80         File file = new File(filepath);
 81         
 82         BufferedReader reader = new BufferedReader(new FileReader(file));
 83         
 84         String word;
 85         
 86         while((word = reader.readLine() )!= null){
 87             
 88             
 89              
 90              word = new String(word.getBytes("UTF-8"), "gb2312");
 91              
 92              stopwords.add(word);
 93              
 94              System.out.print(" stopword = " + word);
 95         }
 96         
 97         return stopwords  ;
 98         
 99     }
100 
101 }