Lucene.Net+盘古分词->开发自己的搜索引擎

//封装类

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
using Lucene.Net.Analysis;
using Lucene.Net.Index;
using Lucene.Net.Documents;
using System.Reflection;
using Lucene.Net.QueryParsers;
using Lucene.Net.Search;
namespace SearchTest
{
    /// <summary>
    /// 盘古分词在lucene.net中的使用帮助类
    /// 调用PanGuLuceneHelper.instance
    /// </summary>
    public class PanGuLuceneHelper
    {
        private PanGuLuceneHelper() { }

#region 单一实例
        private static PanGuLuceneHelper _instance = null;
        /// <summary>
        /// 单一实例
        /// </summary>
        public static PanGuLuceneHelper instance
        {
            get
            {
                if (_instance == null) _instance = new PanGuLuceneHelper();
                return _instance;
            }
        }
        #endregion

#region 分词测试
        /// <summary>
        /// 分词测试
        /// </summary>
        /// <param name="keyword"></param>
        /// <returns></returns>
        public string Token(string keyword)
        {
            string ret = "";
            System.IO.StringReader reader = new System.IO.StringReader(keyword);
            Lucene.Net.Analysis.TokenStream ts = analyzer.TokenStream(keyword, reader);
            bool hasNext = ts.IncrementToken();
            Lucene.Net.Analysis.Tokenattributes.ITermAttribute ita;
            while (hasNext)
            {
                ita = ts.GetAttribute<Lucene.Net.Analysis.Tokenattributes.ITermAttribute>();
                ret += ita.Term + "|";
                hasNext = ts.IncrementToken();
            }
            ts.CloneAttributes();
            reader.Close();
            analyzer.Close();
            return ret;
        }
        #endregion

#region 创建索引
        /// <summary>
        /// 创建索引
        /// </summary>
        /// <param name="datalist"></param>
        /// <returns></returns>
        public bool CreateIndex(List<MySearchUnit> datalist)
        {
            IndexWriter writer = null;
            try
            {
                writer = new IndexWriter(directory_luce, analyzer, false, IndexWriter.MaxFieldLength.LIMITED);//false表示追加（true表示删除之前的重新写入）
            }
            catch
            {
                writer = new IndexWriter(directory_luce, analyzer, true, IndexWriter.MaxFieldLength.LIMITED);//false表示追加（true表示删除之前的重新写入）
            }
            foreach (MySearchUnit data in datalist)
            {
                CreateIndex(writer, data);
            }
            writer.Optimize();
            writer.Dispose();
            return true;
        }

public bool CreateIndex(IndexWriter writer, MySearchUnit data)
        {
            try
            {

if (data == null) return false;
                Document doc = new Document();
                Type type = data.GetType();//assembly.GetType("Reflect_test.PurchaseOrderHeadManageModel", true, true); //命名空间名称 + 类名

//创建类的实例    
                //object obj = Activator.CreateInstance(type, true);  
                //获取公共属性    
                PropertyInfo[] Propertys = type.GetProperties();
                for (int i = 0; i < Propertys.Length; i++)
                {
                    //Propertys[i].SetValue(Propertys[i], i, null); //设置值
                    PropertyInfo pi = Propertys[i];
                    string name=pi.Name;
                    object objval = pi.GetValue(data, null);
                    string value = objval == null ? "" : objval.ToString(); //值
                    if (name == "id" || name=="flag" )//id在写入索引时必是不分词，否则是模糊搜索和删除，会出现混乱
                    {
                        doc.Add(new Field(name, value, Field.Store.YES, Field.Index.NOT_ANALYZED));//id不分词
                    }
                    else
                    {
                        doc.Add(new Field(name, value, Field.Store.YES, Field.Index.ANALYZED));
                    }
                }
                writer.AddDocument(doc);
            }
            catch (System.IO.FileNotFoundException fnfe)
            {
                throw fnfe;
            }
            return true;
        }
        #endregion

#region 在title和content字段中查询数据
        /// <summary>
        /// 在title和content字段中查询数据
        /// </summary>
        /// <param name="keyword"></param>
        /// <returns></returns>
        public List<MySearchUnit> Search(string keyword)
        {

string[] fileds = { "title", "content" };//查询字段
            //Stopwatch st = new Stopwatch();
            //st.Start();
            QueryParser parser = null;// new QueryParser(Lucene.Net.Util.Version.LUCENE_30, field, analyzer);//一个字段查询
            parser = new MultiFieldQueryParser(version, fileds, analyzer);//多个字段查询
            Query query = parser.Parse(keyword);
            int n = 1000;
            IndexSearcher searcher = new IndexSearcher(directory_luce, true);//true-表示只读
            TopDocs docs = searcher.Search(query, (Filter)null, n);
            if (docs == null || docs.TotalHits == 0)
            {
                return null;
            }
            else
            {
                List<MySearchUnit> list = new List<MySearchUnit>();
                int counter = 1;
                foreach (ScoreDoc sd in docs.ScoreDocs)//遍历搜索到的结果
                {
                    try
                    {
                        Document doc = searcher.Doc(sd.Doc);
                        string id = doc.Get("id");
                        string title = doc.Get("title");
                        string content = doc.Get("content");
                        string flag = doc.Get("flag");
                        string imageurl = doc.Get("imageurl");
                        string updatetime = doc.Get("updatetime");

string createdate = doc.Get("createdate");
                        PanGu.HighLight.SimpleHTMLFormatter simpleHTMLFormatter = new PanGu.HighLight.SimpleHTMLFormatter("<font color=\"red\">", "</font>");
                        PanGu.HighLight.Highlighter highlighter = new PanGu.HighLight.Highlighter(simpleHTMLFormatter, new PanGu.Segment());
                        highlighter.FragmentSize = 50;
                        content = highlighter.GetBestFragment(keyword, content);
                        string titlehighlight = highlighter.GetBestFragment(keyword, title);
                        if (titlehighlight != "") title = titlehighlight;
                        list.Add(new MySearchUnit(id, title, content, flag,imageurl, updatetime));
                    }
                    catch (Exception ex)
                    {
                        Console.WriteLine(ex.Message);
                    }
                    counter++;
                }
                return list;
            }
            //st.Stop();
            //Response.Write("查询时间：" + st.ElapsedMilliseconds + " 毫秒<br/>");

}
        #endregion

#region 在不同的分类下再根据title和content字段中查询数据(分页)
        /// <summary>
        /// 在不同的类型下再根据title和content字段中查询数据(分页)
        /// </summary>
        /// <param name="_flag">分类,传空值查询全部</param>
        /// <param name="keyword"></param>
        /// <param name="PageIndex"></param>
        /// <param name="PageSize"></param>
        /// <param name="TotalCount"></param>
        /// <returns></returns>
        public List<MySearchUnit> Search(string _flag,string keyword, int PageIndex, int PageSize, out int TotalCount)
        {
            if (PageIndex < 1) PageIndex = 1;
            //Stopwatch st = new Stopwatch();
            //st.Start();
            BooleanQuery bq = new BooleanQuery();
            if (_flag != "")
            {
                QueryParser qpflag = new QueryParser(version, "flag", analyzer);
                Query qflag = qpflag.Parse(_flag);
                bq.Add(qflag, Occur.MUST);//与运算
            }
            if (keyword != "")
            {
                string[] fileds = { "title", "content" };//查询字段
                QueryParser parser = null;// new QueryParser(version, field, analyzer);//一个字段查询
                parser = new MultiFieldQueryParser(version, fileds, analyzer);//多个字段查询
                Query queryKeyword = parser.Parse(keyword);
                bq.Add(queryKeyword, Occur.MUST);//与运算
            }
            
            TopScoreDocCollector collector = TopScoreDocCollector.Create(PageIndex * PageSize, false);
            IndexSearcher searcher = new IndexSearcher(directory_luce, true);//true-表示只读
            searcher.Search(bq, collector);
            if (collector == null || collector.TotalHits == 0)
            {
                TotalCount = 0;
                return null;
            }
            else
            {
                int start = PageSize * (PageIndex - 1);
                //结束数
                int limit = PageSize;
                ScoreDoc[] hits = collector.TopDocs(start, limit).ScoreDocs;
                List<MySearchUnit> list = new List<MySearchUnit>();
                int counter = 1;
                TotalCount = collector.TotalHits;
                foreach (ScoreDoc sd in hits)//遍历搜索到的结果
                {
                    try
                    {
                        Document doc = searcher.Doc(sd.Doc);
                        string id = doc.Get("id");
                        string title = doc.Get("title");
                        string content = doc.Get("content");
                        string flag = doc.Get("flag");
                        string imageurl = doc.Get("imageurl");
                        string updatetime = doc.Get("updatetime");

PanGu.HighLight.SimpleHTMLFormatter simpleHTMLFormatter = new PanGu.HighLight.SimpleHTMLFormatter("<font color=\"red\">", "</font>");
                        PanGu.HighLight.Highlighter highlighter = new PanGu.HighLight.Highlighter(simpleHTMLFormatter, new PanGu.Segment());
                        highlighter.FragmentSize = 50;
                        content = highlighter.GetBestFragment(keyword, content);
                        string titlehighlight = highlighter.GetBestFragment(keyword, title);
                        if (titlehighlight != "") title = titlehighlight;
                        list.Add(new MySearchUnit(id, title, content, flag,imageurl, updatetime));
                    }
                    catch (Exception ex)
                    {
                        Console.WriteLine(ex.Message);
                    }
                    counter++;
                }
                return list;
            }
            //st.Stop();
            //Response.Write("查询时间：" + st.ElapsedMilliseconds + " 毫秒<br/>");

}
        #endregion

#region 删除索引数据（根据id）
        /// <summary>
        /// 删除索引数据（根据id）
        /// </summary>
        /// <param name="id"></param>
        /// <returns></returns>
        public bool Delete(string id)
        {
            bool IsSuccess = false;
            Term term = new Term("id", id);
            //Analyzer analyzer = new StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_30);
            //Version version = new Version();
            //MultiFieldQueryParser parser = new MultiFieldQueryParser(version, new string[] { "name", "job" }, analyzer);//多个字段查询
            //Query query = parser.Parse("小王");

//IndexReader reader = IndexReader.Open(directory_luce, false);
            //reader.DeleteDocuments(term);
            //Response.Write("删除记录结果： " + reader.HasDeletions + "<br/>");
            //reader.Dispose();

IndexWriter writer = new IndexWriter(directory_luce, analyzer, false, IndexWriter.MaxFieldLength.LIMITED);
            writer.DeleteDocuments(term); // writer.DeleteDocuments(term)或者writer.DeleteDocuments(query);
            ////writer.DeleteAll();
            writer.Commit();
            //writer.Optimize();//
            IsSuccess = writer.HasDeletions();
            writer.Dispose();
            return IsSuccess;
        }
        #endregion

#region 删除全部索引数据
        /// <summary>
        /// 删除全部索引数据
        /// </summary>
        /// <returns></returns>
        public bool DeleteAll()
        {
            bool IsSuccess = true;
            try
            {
                IndexWriter writer = new IndexWriter(directory_luce, analyzer, false, IndexWriter.MaxFieldLength.LIMITED);
                writer.DeleteAll();
                writer.Commit();
                //writer.Optimize();//
                IsSuccess = writer.HasDeletions();
                writer.Dispose();
            }
            catch
            {
                IsSuccess = false;
            }
            return IsSuccess;
        }
        #endregion

#region directory_luce
        private Lucene.Net.Store.Directory _directory_luce = null;
        /// <summary>
        /// Lucene.Net的目录-参数
        /// </summary>
        public Lucene.Net.Store.Directory directory_luce
        {
            get
            {
                if (_directory_luce == null) _directory_luce = Lucene.Net.Store.FSDirectory.Open(directory);
                return _directory_luce;
            }
        }
        #endregion

#region directory
        private System.IO.DirectoryInfo _directory = null;
        /// <summary>
        /// 索引在硬盘上的目录
        /// </summary>
        public System.IO.DirectoryInfo directory
        {
            get
            {
                if (_directory == null)
                {
                    string dirPath = AppDomain.CurrentDomain.BaseDirectory + "SearchIndex";
                    if (System.IO.Directory.Exists(dirPath) == false) _directory = System.IO.Directory.CreateDirectory(dirPath);
                    else _directory = new System.IO.DirectoryInfo(dirPath);
                }
                return _directory;
            }
        }
        #endregion

#region analyzer
        private Analyzer _analyzer = null;
        /// <summary>
        /// 分析器
        /// </summary>
        public Analyzer analyzer
        {
            get
            {
                //if (_analyzer == null)
                {
                    _analyzer = new Lucene.Net.Analysis.PanGu.PanGuAnalyzer();//盘古分词分析器
                    //_analyzer = new StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_30);//标准分析器
                }
                return _analyzer;
            }
        }
        #endregion

#region version
        private static Lucene.Net.Util.Version _version = Lucene.Net.Util.Version.LUCENE_30;
        /// <summary>
        /// 版本号枚举类
        /// </summary>
        public Lucene.Net.Util.Version version
        {
            get
            {
                return _version;
            }
        }
        #endregion
    }

#region 索引的一个行单元，相当于数据库中的一行数据
    /// <summary>
    /// 索引的一个行单元，相当于数据库中的一行数据
    /// </summary>
    public class MySearchUnit
    {
        public MySearchUnit(string _id, string _title, string _content, string _flag, string _imageurl, string _updatetime)
        {
            this.id = _id;
            this.title = _title;
            this.content = _content;
            this.flag = _flag;
            this.imageurl = _imageurl;
            this.updatetime = _updatetime;
        }
        /// <summary>
        /// 唯一的id号
        /// </summary>
        public string id { get; set; }
        /// <summary>
        /// 标题
        /// </summary>
        public string title { get; set; }
        /// <summary>
        /// 内容
        /// </summary>
        public string content { get; set; }
        /// <summary>
        /// 其他信息
        /// </summary>
        public string flag { get; set; }
        /// <summary>
        /// 图片路径
        /// </summary>
        public string imageurl { get; set; }
        /// <summary>
        /// 时间
        /// </summary>
        public string updatetime { get; set; }
    }
    #endregion
}

原文地址：http://blog.csdn.net/pukuimin1226/article/details/17558247

时间： 2024-10-09 22:11:20

Lucene.Net+盘古分词->开发自己的搜索引擎的相关文章

Lucene.Net+盘古分词

前言 各位朋友,谢谢大家的支持,由于文件过大,有考虑到版权的问题,故没有提供下载,本人已建立一个搜索技术交流群:77570783,源代码已上传至群共享,需要的朋友,请自行下载! 首先自问自答几个问题,以让各位看官了解写此文的目的 什么是站内搜索?与一般搜索的区别?很多网站都有搜索功能,很多都是用SQL语句的Like实现的,但是Like无法做到模糊匹配(例如我搜索“.net学习”,如果有“.net的学习”,Like就无法搜索到,这明显不符合需求,但是站内搜索就能做到),另外Like会造成全盘扫描,

Lucene.Net3.0.3+盘古分词器学习使用

一.Lucene.Net介绍 Lucene.net是Lucene的.net移植版本,是一个开源的全文检索引擎开发包,即它不是一个完整的全文检索引擎,而是一个全文检索引擎的架构,提供了完整的查询引擎和索引引擎.开发人员可以基于Lucene.net实现全文检索的功能. Lucene.net是Apache软件基金会赞助的开源项目,基于Apache License协议. Lucene.net并不是一个爬行搜索引擎,也不会自动地索引内容.我们得先将要索引的文档中的文本抽取出来,然后再将其加到Lucene.

lucene+盘古分词

一般的网站都会有都会有搜索的功能,一般实现搜索主要有三种方案 第一种是最差的,也是最不推荐的,使用数据库的模糊查询例如select * form table where 字段 like XXX,这种查询的缺点很明显: (1)       无法查找几个关键词不连在一起的情况 (2)       全表扫描 效率低下 第二种:使用SqlServer的全文本检索功能 举例:select * form table where msg = ‘江苏南京’ 这是就可以写成select * form table

Lucene.net入门学习（结合盘古分词）

Lucene简介 Lucene是apache软件基金会4 jakarta项目组的一个子项目,是一个开放源代码的全文检索引擎工具包,即它不是一个完整的全文检索引擎,而是一个全文检索引擎的架构,提供了完整的 查询引擎和索引引擎,部分文本分析引擎(英文与德文两种西方语言).Lucene的目的是为软件开发人员提供一个简单易用的工具包,以方便的在目标系统中 实现全文检索的功能,或者是以此为基础建立起完整的全文检索引擎. Lucene.net简介 Lucene.net是Lucene的.net移植版本,是一个

Net Core使用Lucene.Net和盘古分词器 实现全文检索

Lucene.net Lucene.net是Lucene的.net移植版本,是一个开源的全文检索引擎开发包,即它不是一个完整的全文检索引擎,而是一个全文检索引擎的架构,提供了完整的查询引擎和索引引擎,是一个高性能.可伸缩的文本搜索引擎库.它的功能就是负责将文本数据按照某种分词算法进行切词,分词后的结果存储在索引库中,从索引库检索数据的速度非常快.Lucene.net需要有索引库,并且只能进行站内搜索.(来自百度百科) 效果图 盘古分词 如何使用 将PanGu.dIl与PanGu.Lucenet.

Lucene.Net 与 盘古分词

1.关键的一点,Lucene.Net要使用3.0以下的版本,否则与盘古分词接口不一致. 版权声明:本文为博主原创文章,未经博主允许不得转载.

lucene.net 3.0.3、结合盘古分词进行搜索的小例子(转)

lucene.net 3.0.3.结合盘古分词进行搜索的小例子(分页功能) 添加:2013-12-25 更新:2013-12-26 新增分页功能. 更新:2013-12-27 新增按分类查询功能,调整索引行新增记录的图片字段. //封装类 [csharp] view plaincopyprint? using System; using System.Collections.Generic; using System.Linq; using System.Web; using Lucene.Ne

让盘古分词支持最新的Lucene.Net 3.0.3

好多年没升级过的Lucene.Net最近居然升级了,到了3.0.3后接口发生了很大变化,原来好多分词库都不能用了,所以上次我把MMSeg给修改了一下支持了Lucene.Net 3.0.3(参考<基于MMSeg算法的中文分词类库>). 然后我知道了.Net下还有一个盘古分词(http://pangusegment.codeplex.com/),但也不支持Lucene.Net 3.0.3,网上也不少人在问,于是就下载了它的最新代码,基于Lucene.Net 3.0.3更新了,顺便把它的词库给放到d

lucene.net 3.0.3、结合盘古分词进行搜索的小例子（分页功能）

转自:http://blog.csdn.net/pukuimin1226/article/details/17558247 添加:2013-12-25 更新:2013-12-26 新增分页功能. 更新:2013-12-27 新增按分类查询功能,调整索引行新增记录的图片字段. 最新盘古分词dll和词典管理工具下载:http://pangusegment.codeplex.com/ 词典下载:http://pangusegment.codeplex.com/releases/view/47411 L