lucene.net和（pangu）盘古分词 搜索引擎的简单实现

这篇文章是介绍lucene.net和盘古分词的简单的实现调用  。建立索引，搜索，盘古分词的基类如下：

 

 

using System; using System.Data; using System.Configuration; using System.Linq; using System.Web; using System.Web.Security; using System.Web.UI; using System.Web.UI.HtmlControls; using System.Web.UI.WebControls; using System.Web.UI.WebControls.WebParts; using System.Xml.Linq; using Lucene.Net.Index; using Lucene.Net.Documents; using Lucene.Net.QueryParsers; using Lucene.Net.Analysis; using Lucene.Net.Search; using Lucene.Net.Analysis.Standard; using PanGu; using Lucene.Net.Analysis.PanGu; using StringBulider; using System.Text; using System.Collections; using System.Collections.Generic; /// <summary> ///辅导资料的搜索逻辑 的摘要说明 /// </summary> public class DataSearch { public DataSearch() { // //TODO: 在此处添加构造函数逻辑 // } A_DB_Conn db = null; string index = System.Web.HttpContext.Current.Server.MapPath("~/Gongxin/DataIndex");//索引路径 string wordPath = System.Web.HttpContext.Current.Server.MapPath("~/Dictionaries");//字典路径 //用于存储检索出数据总数 private int _count; public int Count { get { return _count; } set { _count = value; } } //搜索关键字 private string _keyword; public string KeyWord { get { return _keyword;} set{_keyword = value;} } //搜索时间 private double _time; public double Time { get { return _time; } set { _time = value; } } /// <summary> /// 得到数据结果集 /// </summary> /// <returns></returns> public DataTable GetDS() { db = new A_DB_Conn(); string sql = "select N_id,N_title,N_datetime,N_Content from N_Newinfo where N_ifkping=1 and N_id=12 order by N_datetime desc"; DataTable dataTable = null; try { dataTable = db.GetDataSet(sql).Tables[0]; } catch { dataTable = null; } finally { db.close(); db.Dispose(); } return dataTable; } /// <summary> /// 添加索引域 /// </summary> /// <param name="writer"></param> /// <param name="dataTable"></param> public void AddDocument(IndexWriter writer, DataTable dataTable) { for (int i = 0; i < dataTable.Rows.Count; i++) { Document doc = new Document(); doc.Add(new Field("ID", dataTable.Rows[i]["N_id"].ToString(),Field.Store.YES,Field.Index.TOKENIZED)); doc.Add(new Field("N_title",StringCheck.SqlQueryDecode(dataTable.Rows[i]["N_title"].ToString()),Field.Store.YES,Field.Index.TOKENIZED)); doc.Add(new Field("N_datetime", dataTable.Rows[i]["N_datetime"].ToString(),Field.Store.YES,Field.Index.TOKENIZED)); if (StringCheck.SqlQueryDecode(dataTable.Rows[i]["N_Content"].ToString()).Length > 100) { doc.Add(new Field("N_Content", StringCheck.SqlQueryDecode(dataTable.Rows[i]["N_Content"].ToString()).Replace("<p>","").Replace("</p>","").Substring(0, 200), Field.Store.YES, Field.Index.TOKENIZED)); } else { doc.Add(new Field("N_Content", StringCheck.SqlQueryDecode(dataTable.Rows[i]["N_Content"].ToString()).Replace("<p>", "").Replace("</p>", ""), Field.Store.YES, Field.Index.TOKENIZED)); } writer.AddDocument(doc); } } /// <summary> /// 建立索引 /// </summary> public void Index() { PanGu.Segment.Init(); PanGuAnalyzer analyzer = new PanGuAnalyzer(); IndexWriter writer = new IndexWriter(index, analyzer, true); DataTable dataTable = GetDS(); AddDocument(writer, dataTable); writer.Optimize(); writer.Close(); } /// <summary> /// 对药搜索的词进行分词 /// </summary> /// <param name="keyWords">要收索的词</param> /// <param name="ktTokenizer">分词对象</param> /// <returns>分词后的结果</returns> public string GetKeyWordSplitBySpace(string keyWords, PanGuTokenizer ktTokenizer) { StringBuilder builder = new StringBuilder(); ICollection<WordInfo> words = ktTokenizer.SegmentToWordInfos(keyWords); foreach(WordInfo word in words) { if (word == null) { continue; } KeyWord = KeyWord + word+","; builder.AppendFormat("{0}^{1}",word.Word,(int)Math.Pow(3,word.Rank)); } KeyWord = KeyWord.Substring(0, KeyWord.Length - 1); return builder.ToString().Trim(); } /// <summary> /// 检索信息 /// </summary> /// <param name="keyWord">检索关键字</param> /// <param name="pageNumber">当前第几条</param> /// <param name="pageSize">每页显示的条数</param> /// <returns></returns> public string Search(string keyWord, int pageNumber, int pageSize) { string word = GetKeyWordSplitBySpace(keyWord, new PanGuTokenizer()); IndexSearcher search = new IndexSearcher(index); StringBuilder builder = new StringBuilder(); PanGuAnalyzer analyzer = new PanGuAnalyzer(true); //多字段搜索字段 MultiFieldQueryParser parser = new MultiFieldQueryParser(new string[] { "N_title" },analyzer); //分词 Query query = parser.Parse(word); Hits hits = search.Search(query); Count = hits.Length(); int num = 0;//记录每页最后一条 if (Count < pageNumber + pageSize) { num = Count; } else { num = pageSize + pageNumber; } DateTime begin = DateTime.Now; for (int i = pageNumber; i < num; i++) { Document doc = hits.Doc(i); //创建HTMLFormatter,参数为高亮搜索词的HTML代码 PanGu.HighLight.SimpleHTMLFormatter simpleHTMLFormatter = new PanGu.HighLight.SimpleHTMLFormatter("<font color='red'>", "</font>"); //创建高亮，输入HTML代码和 盘古对象Semgent PanGu.HighLight.Highlighter highter = new PanGu.HighLight.Highlighter(simpleHTMLFormatter, new Segment()); //设置每个摘要字段的字符数 highter.FragmentSize = 200; string title = highter.GetBestFragment(keyWord, doc.Get("N_title")); builder.Append("<div class='sea_h_list'>"); if (!string.IsNullOrEmpty(title)) { builder.Append("<h2><a href="D_NewBody.aspx?gxjy=" + UrlEncrpt.Encrypt(doc.Get("ID")) + "" mce_href="D_NewBody.aspx?gxjy=" + UrlEncrpt.Encrypt(doc.Get("ID")) + "" target='_blank'>" + highter.GetBestFragment(keyWord, doc.Get("N_title")) + "</a></h2>"); } else { builder.Append("<h2>" + doc.Get("N_title") + "</h2>"); } builder.Append(doc.Get("N_Content")); DateTime time = Convert.ToDateTime(doc.Get("N_datetime")); builder.Append("<div>"+time.ToString("yyyy-MM-dd")+"</div>"); builder.Append("</div>"); } DateTime end = DateTime.Now; double ts = (end - begin).TotalMilliseconds; Time =ts / 1000.000; search.Close();//关闭检索器 return builder.ToString(); } }

 

前台可以用一个button调用一个button事件来生成索引

 

前台生成索引事件如下。

protected void Button1_Click(object sender, EventArgs e) { search = new NewsSearch(); search.Index(); Response.Write("<Script>alert('索引生成成功！')</Script>"); }

 

 

索引生成成功后，就可以进行搜索了，本人实现搜索的功能时调用了AspNetPager分页控件，具体的搜索的方法如下：

 

 

public void GetInfoList(string keyWord) { try { search = new NewsSearch(); Stopwatch sw = new Stopwatch(); sw.Start(); InfoList = search.Search(keyWord, (this.AspNetPager1.CurrentPageIndex - 1) * this.AspNetPager1.PageSize, this.AspNetPager1.PageSize); sw.Stop(); time = (sw.ElapsedMilliseconds / 1000.000).ToString();//显示所用搜索时间 if (InfoList.Trim().Length == 0) { InfoList = "<div class='rig01_center red'>暂无搜索信息。</div>"; } this.AspNetPager1.RecordCount = search.Count; count = search.Count.ToString();//搜索的条数 strkeyWord = search.KeyWord;//显示搜索关键字 } catch { Response.Redirect("404.html"); } }

 

 

其中 Stopwatch使用了计算搜索所用的时间，将它复制到time 全局变量中,count 全局变量时用来存储搜索的条数。

 

 

搜索出来数据后，前台页面用<%=count%><%=time%> <%=InfoList%>

调用即可。

 

 

 

前台显示效果：

 

 

 

 

本文只是简单的实现了lucene.net 和盘古分词,仅供大家学习，有不足的地方还请见谅，有疑问随时联系，谢谢！