Lucene是apache軟體基金會4 jakarta專案組的一個子專案，是一個開放原始碼的全文檢索引擎工具包。程式設計師們不僅使用它構建具體的全文檢索應用，而且將之整合到各種系統軟體中去，以及構建Web應用。不同版本的lucene之間還是有一定的差異的。專案裡我使用的是3.0.3版本，以編寫的校園網資料模組為例，對lucene的一些基本原理和用法將直接在下面的原始碼和註釋中介紹。

實現效果如下

1、首先匯入lucene相關的jar包

2、前端搜尋html及返回結果展示

<form id="search-form" action="/schoolnet/other.do?flag=searchdatas&searchpageNow=1"method="post" name="searchdatas" target="blank" onsubmit="return checksearch()">
 

<div class="site-nav">
					<span>當前位置 : </span><a href="/schoolnet/login.do?flag=goIndexUI">首頁</a>
					<a href="/schoolnet/login.do?flag=goHomeUI" title="">個人主頁</a>>>資料搜尋
				</div>
				<c:if test="${action=='searchdatas' }">	<div class="page-news">
					<div class="content">
						<div class="mes_response" id="mes_response">
						<c:set var="heightshow" value="<span style='color:red'>${search}</span>"></c:set>
						<c:set var="allr" value="<span style='color:red'>${allrows}</span>"></c:set>
							<table>
								<tr>
									<th class="news-time">您搜尋的“<span>${heightshow}</span>”， 共有<span>${allr}</span>個結果</div></th>
								</tr>
								<c:forEach var="data" items="${datas }" varStatus="status">
								<tr>
 								   <td>
										<c:if test="${fn:substringAfter(data.docu, '.')=='txt'}"><img src="/schoolnet/images/front/txt.png" width="25px" height="25px"/></c:if>
										<c:if test="${fn:substringAfter(data.docu, '.')=='pdf'}"><img src="/schoolnet/images/front/pdf.gif"  width="25px" height="25px"/></c:if>
										<c:if test="${fn:substringAfter(data.docu, '.')=='doc'}"><img src="/schoolnet/images/front/word.png" width="25px" height="25px"/></c:if>
										<c:if test="${fn:substringAfter(data.docu, '.')=='ppt'}"><img src="/schoolnet/images/front/ppt.png" width="25px" height="25px"/></c:if>
										<c:if test="${fn:substringAfter(data.docu, '.')=='zip'}"><img src="/schoolnet/images/front/zip.ico"  width="25px" height="25px"/></c:if>
										<c:if test="${fn:substringAfter(data.docu, '.')=='docx'}"><img src="/schoolnet/images/front/word.png" width="25px" height="25px"/></c:if>
										<span class="word4"><a
													href="/schoolnet/datas.do?flag=godatasshowUI&datasid=${data.id }&typeid=${data.datasecondtype.id}"
													class="xh" onclick="readgt(this)" id=""
													onmouseover="this.style.color='#333';"
													onmouseout="this.style.color='#666';"
													style="color: #000; font-weight: bold; text-decoration: none">${fn:replace(data.title, search, heightshow)}</a>
										</span><span style="float:right"></span><br/>
										<span style="color: #555">${fn:replace(data.content, search, heightshow)}</span>
									</td>
								</tr>
								</c:forEach>
							</table>
						</div>
 

java程式碼

//path就是當前這個web應用是絕對路徑 D:\apache-tomcat-7.0.27\webapps\schoolnet
		String path=request.getSession().getServletContext().getRealPath("/")+"\\Index\\";
		//lucene中的檔案操作都是通過這Directory來實現的。Directory的實現類可以分為檔案目錄，記憶體目錄和目錄的代理類及工具類。
		//這裡使用最簡單的檔案目錄FSDirectory的子類SimpleFSDirectory
		Directory dir = new SimpleFSDirectory(new File(path));
		//判斷是否已經存在索引目錄，是則在原索引上追加內容，否則覆蓋
		boolean exist = IndexReader.indexExists(dir);
		IndexWriter writer = new IndexWriter(dir, new StandardAnalyzer(
				Version.LUCENE_30), !exist, IndexWriter.MaxFieldLength.LIMITED);
		try {
			//根據專案實際情況，按需建立對應的索引，先查出資料
			List<Datas> dataList=userService.executeQueryByPage("from Datas where dataecondtype.datafirsttype.id=1", null, 0, 1000);
			//將資料寫進索引，一個document實際就是對應一條記錄
			for (int i = 0; i < dataList.size(); i++) {
				//id要儲存,不用索引。因為我們輸入的是想搜尋的內容，id對於查詢暫時用不到，但id佔儲存空間小且後面拿物件很需要，所有要儲存。
				//title(標題)、content(內容)要分詞,索引,但不儲存.由於他們太大了, 
				Document doc = new Document();
				//對id的檢索策略:儲存欄位值，但不索引
				doc.add(new Field("id", dataList.get(i).getId().toString(), Field.Store.YES,
						Field.Index.NOT_ANALYZED));
				//對title的檢索策略：:不儲存欄位值，分詞建索引
				doc.add(new Field("title", dataList.get(i).getTitle(), Field.Store.NO,
						Field.Index.ANALYZED));
				//對content的檢索策略：:不儲存欄位值，分詞建索引
				doc.add(new Field("content", dataList.get(i).getContent(), Field.Store.NO,
						Field.Index.ANALYZED));
				writer.addDocument(doc);
			}
			writer.optimize();
			PrintWriter out = null;
			response.setCharacterEncoding("utf-8");
			response.setContentType("text/html;charset=utf-8");
			try {
				out = response.getWriter();
			} catch (IOException e) {
				// TODO Auto-generated catch block
				e.printStackTrace();
			}
			out.print("<script>alert('建立索引成功');window.history.go(-1);</script>");
		} finally {
			//必須關閉資源
			writer.close();
		}
 

生成對應索引檔案

4、搜尋

	//獲取搜尋關鍵字
		String search=request.getParameter("search");
		int PagesSize=20;
		String pageNow=request.getParameter("searchpageNow");
		int pageNows=Integer.valueOf(pageNow);
		request.setAttribute("searchpageNow", pageNows);
		request.setAttribute("PagesSize", PagesSize);
		request.setAttribute("search", search);
		request.setAttribute("action", "searchdatas");
		//path就是當前這個web應用是絕對路徑 D:\apache-tomcat-7.0.27\webapps\schoolnet
		String path=request.getSession().getServletContext().getRealPath("/")+"\\Index\\";
		File file=new File(path);
		//還沒建立全文索引
		if (!file.exists()) {
			
		}else {
			//lucene中的檔案操作都是通過這Directory來實現的。Directory的實現類可以分為檔案目錄，記憶體目錄和目錄的代理類及工具類。
			//這裡使用最簡單的檔案目錄FSDirectory的子類SimpleFSDirectory
			Directory dir = new SimpleFSDirectory(file);
			//建立dir的檢索器
			Searcher searcher = new IndexSearcher(dir);
			try {
				//建立標準的分析器
				Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_30);
				//構造一個布林查詢
				BooleanQuery bq = new BooleanQuery();
				//使用MultiFieldQueryParser,同時在索引的多個域中搜索同一個關鍵字。搜尋策略：title或content滿足條件即可。
				Query query = MultiFieldQueryParser.parse(Version.LUCENE_30, search,
							new String[]{"title","content"},new BooleanClause.Occur[]{BooleanClause.Occur.SHOULD, BooleanClause.Occur.SHOULD}, analyzer);
				//將query新增到布林查詢中。搜尋策略：title和content必須滿足條件。	
				bq.add(query, BooleanClause.Occur.MUST);
				//獲取相匹配的搜尋條件的前100個搜尋結果
				TopDocs docs = searcher.search(bq, 100);
				//獲取doc裡儲存的id
				List<Integer> list = new ArrayList<Integer>(docs.scoreDocs.length);
				ScoreDoc[] hits = docs.scoreDocs;
				for (int i = 0; i < hits.length; i++) {
					Document d = searcher.doc(hits[i].doc);
					list.add(Integer.valueOf(d.getField("id").stringValue()));
				}
				//根據id獲取對應的實體
				List<Datas> dataList = new ArrayList<Datas>(list.size());
				for (Object id : list) {
					dataList.add((Datas)userService.findById(Datas.class,(Integer) id));
				}
				int allrows=dataList.size();
				int PageCount=(allrows-1)/PagesSize+1;
				if(pageNows<1)
				{
					pageNows=1;
				}
				if(pageNows>PageCount)
				{
					pageNows=PageCount;
				}
				request.setAttribute("datas", dataList);
				request.setAttribute("PageCount", PageCount);
				request.setAttribute("allrows", allrows);
			} finally {
				//必須關閉資源
				searcher.close();
			}
		}