luceneindexnumber.java

《lucene+nutch搜索引擎开发》源代码

package chapter5;

import java.util.Date;
import java.io.*;

import org.apache.lucene.search.Hits;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.RangeQuery;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.store.Directory; 
import org.apache.lucene.store.FSDirectory;
import org.apache.lucene.document.Field;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.Term;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.standard.StandardAnalyzer;
import org.apache.lucene.analysis.SimpleAnalyzer;
import java.text.DecimalFormat; 

public class LuceneIndexNumber {
	
	private static String Dest_Index_Path = "D:\\workshop\\TextIndex";
	
	static protected Integer[] keywords = {120,2,30};
	static protected String[] textdetail = {"数字一","数字二", "数字三"} ;
	
	/*================================================================
	 * 名 称：QueryNumberIndex
	 * 功 能：构造检索查询器，对指定的索引进行查询，找到指定区间值，并输出相应结果。
	 ===============================================================*/
	public static void QueryNumberIndex(){
		
		try {
			IndexSearcher searcher = new IndexSearcher(Dest_Index_Path);  // 生成检索器
			Term termstart = new Term("id","0006");
			Term termend   = new Term("id","0105");
			RangeQuery query = new RangeQuery(termstart,termend,false);   // 生成区间检索对象

			System.out.println(query.toString());

			Hits hits = searcher.search(query);                           // 提交检索
			
			System.out.println("Search result:");
			
			for(int i=0; i < hits.length(); i++)                          // 输出结果
			{
				System.out.println(hits.doc(i).getField("id").stringValue());			
			}
			
		}catch (IOException e) {
				e.printStackTrace();
		}

		System.out.println("Search finaish");		
		
	}
	/*================================================================
	 * 名 称：IndexNumberBuilder
	 * 功 能：构造数字类型信息的磁盘索引，添加内容到指定目录，为后续检索查询做好准备。
	 ===============================================================*/
	public static void IndexNumberBuilder(){
		
		try {
			
			Analyzer TextAnalyzer = new SimpleAnalyzer();                 // 生成分析器
			IndexWriter TextIndex = new IndexWriter(Dest_Index_Path,TextAnalyzer,true); // 索引器对象
	        TextIndex.setUseCompoundFile(true);                           // 使用符合索引
			for(int i = 0; i < 3 ; i++){
				Document document = new Document();                       // 空文档
			    DecimalFormat df = new DecimalFormat("0000");             // 数字化格式
			    String idkey = df.format(keywords[i]);                    // 数字关键字转换成定长字符串
							
				Field field_id = new Field("id", idkey,                   // 添加关键字域
						Field.Store.YES,Field.Index.UN_TOKENIZED);
				System.out.println( idkey );
			
				document.add(field_id);

				Field field_content = new Field("content", textdetail[i], // 添加内容域
						Field.Store.YES,Field.Index.TOKENIZED);
				
				document.add(field_content);
				
				TextIndex.addDocument(document);

			}
			TextIndex.optimize();
			TextIndex.close();
		
		}catch (IOException e) {
				e.printStackTrace();
		}

		System.out.println("Index success");		
	}
	
	/*================================================================
	 * 名 称：main
	 * 功 能：测试Lucene对数字类型内容的索引建立和检索查询功能。
	 ===============================================================*/
	public static void main(String[] args) {
		
		IndexNumberBuilder();         // 创建索引
		QueryNumberIndex();           // 检索内容

	}

}