学步园

      lucene3.0已于2009-11-25发布啦，但网上的入门实例都是针对lucene3.0以前的，相对于以前的版本，貌似改动不小。 
      本人从头开始学习lucene，现在用的是《lucene in action中文版》，结合lucene3.0文档写了个入门实例，可供像我一样直接从lucene3.0开始学习的新手参考！ 

入门实例： 
lucene3.0.0的jar包和《三国演义》电子书的下载网址：lucene3.0.0的学习资料
1.预处理：先把网上下载的一个《三国演义》电子书“三国演义.txt”（可用其他代替，呵呵）切割成多个小文件。 

package com.taobao.zhujiadun.basic;<br /> import java.io.BufferedReader;<br /> import java.io.BufferedWriter;<br /> import java.io.File;<br /> import java.io.FileNotFoundException;<br /> import java.io.FileReader;<br /> import java.io.FileWriter;<br /> import java.io.IOException;<br /> /**<br /> * @author ht<br /> * 预处理<br /> *<br /> */<br /> public class FilePreprocess {<br /> public static void main(String[] arg){<br /> String outputpath = "D://test//small//";//小文件存放路径<br /> String filename = "D://test//三国演义.txt";//原文件存放路径<br /> if(!new File(outputpath).exists()){<br /> new File(outputpath).mkdirs();<br /> }<br /> splitToSmallFiles(new File(filename), outputpath);<br /> }</p> <p> /**大文件切割为小的<br /> * @param file<br /> * @param outputpath<br /> */<br /> public static void splitToSmallFiles(File file ,String outputpath){<br /> int filePointer = 0;<br /> int MAX_SIZE = 10240;<br /> String filename = "output";<br /> BufferedWriter writer = null;</p> <p> try {<br /> BufferedReader reader = new BufferedReader(new FileReader(file));<br /> StringBuffer buffer = new StringBuffer();<br /> String line = reader.readLine();<br /> while(line != null){<br /> buffer.append(line).append("/r/n");<br /> if(buffer.toString().getBytes().length>=MAX_SIZE){<br /> writer = new BufferedWriter(new FileWriter(outputpath+filename+filePointer+".txt"));<br /> writer.write(buffer.toString());<br /> writer.close();<br /> filePointer++;<br /> buffer=new StringBuffer();<br /> }<br /> line = reader.readLine();<br /> }<br /> writer = new BufferedWriter(new FileWriter(outputpath+filename+filePointer+".txt"));<br /> writer.write(buffer.toString());<br /> writer.close();<br /> System.out.println("The file hava splited to small files !");<br /> } catch (FileNotFoundException e) {<br /> System.out.println("file not found !");<br /> e.printStackTrace();<br /> } catch (IOException e) {<br /> e.printStackTrace();<br /> }<br /> }<br /> }<br />  

2.用lucene3.0生成索引类:用lencene3.0对生成的多个小文件进行索引，中文分词用的是lucene3.0自带的StandardAnalyzer. 

package com.taobao.zhujiadun.basic;<br /> import java.io.File;<br /> import java.io.FileReader;<br /> import java.io.IOException;<br /> import java.util.Date;<br /> import org.apache.lucene.analysis.standard.StandardAnalyzer;<br /> import org.apache.lucene.document.Document;<br /> import org.apache.lucene.document.Field;<br /> import org.apache.lucene.index.IndexWriter;<br /> import org.apache.lucene.store.FSDirectory;<br /> import org.apache.lucene.util.Version;<br /> /**<br /> * @author ht<br /> * 索引生成<br /> *<br /> */<br /> public class Indexer {<br /> private static String INDEX_DIR = "D://test//index";//索引存放目录<br /> private static String DATA_DIR = "D://test//small//";//小文件存放的目录</p> <p> public static void main(String[] args) throws Exception {<br /> long start = new Date().getTime();<br /> int numIndexed = index(new File(INDEX_DIR), new File(DATA_DIR));//调用index方法<br /> long end = new Date().getTime();<br /> System.out.println("Indexing " + numIndexed + " files took " + (end - start) + " milliseconds");<br /> }<br /> /**索引dataDir下的.txt文件，并储存在indexDir下，返回索引的文件数量<br /> * @param indexDir<br /> * @param dataDir<br /> * @return int<br /> * @throws IOException<br /> */<br /> public static int index(File indexDir, File dataDir) throws IOException {<br /> if (!dataDir.exists() || !dataDir.isDirectory()) {<br /> throw new IOException(dataDir + " does not exist or is not a directory");<br /> }<br /> IndexWriter writer = new IndexWriter(FSDirectory.open(indexDir), new StandardAnalyzer(Version.LUCENE_CURRENT), true,<br /> IndexWriter.MaxFieldLength.LIMITED);//有变化的地方</p> <p> indexDirectory(writer, dataDir);<br /> int numIndexed = writer.numDocs();<br /> writer.optimize();<br /> writer.close();<br /> return numIndexed;<br /> }<br /> /**循环遍历目录下的所有.txt文件并进行索引<br /> * @param writer<br /> * @param dir<br /> * @throws IOException<br /> */<br /> private static void indexDirectory(IndexWriter writer, File dir) throws IOException {<br /> File[] files = dir.listFiles();<br /> for (int i = 0; i < files.length; i++) {<br /> File f = files[i];<br /> if (f.isDirectory()) {<br /> indexDirectory(writer, f); // recurse<br /> }<br /> else if (f.getName().endsWith(".txt")) {<br /> indexFile(writer, f);<br /> }<br /> }<br /> }<br /> /**对单个txt文件进行索引<br /> * @param writer<br /> * @param f<br /> * @throws IOException<br /> */<br /> private static void indexFile(IndexWriter writer, File f)throws IOException {<br /> if (f.isHidden() || !f.exists() || !f.canRead()) {<br /> return;<br /> }<br /> System.out.println("Indexing " + f.getCanonicalPath());<br /> Document doc = new Document();<br /> doc.add(new Field("contents",new FileReader(f)));//有变化的地方<br /> doc.add(new Field("filename",f.getCanonicalPath(),Field.Store.YES, Field.Index.ANALYZED));//有变化的地方<br /> writer.addDocument(doc);<br /> }<br /> } 

3.查询类：查询“玄德”！ 

package com.taobao.zhujiadun.basic;<br /> import java.io.File;<br /> import java.util.Date;<br /> import org.apache.lucene.analysis.standard.StandardAnalyzer;<br /> import org.apache.lucene.document.Document;<br /> import org.apache.lucene.queryParser.QueryParser;<br /> import org.apache.lucene.search.IndexSearcher;<br /> import org.apache.lucene.search.Query;<br /> import org.apache.lucene.search.ScoreDoc;<br /> import org.apache.lucene.search.TopScoreDocCollector;<br /> import org.apache.lucene.store.FSDirectory;<br /> import org.apache.lucene.util.Version;<br /> /**<br /> * @author ht<br /> * 查询<br /> *<br /> */<br /> public class Searcher {<br /> private static String INDEX_DIR = "D://test//index//";//索引所在的路径<br /> private static String KEYWORD = "玄德";//关键词<br /> private static int TOP_NUM = 100;//显示前100条结果</p> <p> public static void main(String[] args) throws Exception {<br /> File indexDir = new File(INDEX_DIR);<br /> if (!indexDir.exists() || !indexDir.isDirectory()) {<br /> throw new Exception(indexDir +" does not exist or is not a directory.");<br /> }<br /> search(indexDir, KEYWORD);//调用search方法进行查询<br /> }<br /> /**查询<br /> * @param indexDir<br /> * @param q<br /> * @throws Exception<br /> */<br /> public static void search(File indexDir, String q) throws Exception {<br /> IndexSearcher is = new IndexSearcher(FSDirectory.open(indexDir),true);//read-only<br /> String field = "contents";</p> <p> QueryParser parser = new QueryParser(Version.LUCENE_CURRENT, field, new StandardAnalyzer(Version.LUCENE_CURRENT));//有变化的地方<br /> Query query = parser.parse(q);</p> <p> TopScoreDocCollector collector = TopScoreDocCollector.create(TOP_NUM , false);//有变化的地方</p> <p> long start = new Date().getTime();// start time</p> <p> is.search(query, collector);<br /> ScoreDoc[] hits = collector.topDocs().scoreDocs;</p> <p> System.out.println(hits.length);<br /> for (int i = 0; i < hits.length; i++) {<br /> Document doc = is.doc(hits[i].doc);//new method is.doc()<br /> System.out.println(doc.getField("filename")+" "+hits[i].toString()+" ");<br /> }<br /> long end = new Date().getTime();//end time</p> <p> System.out.println("Found " + collector.getTotalHits() +<br /> " document(s) (in " + (end - start) +<br /> " milliseconds) that matched query '" +<br /> q + "':");<br /> }<br /> }