真爱无限的知识驿站

学习积累技术经验,提升自身能力

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。

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 


//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>();
   &nb

发表评论:

◎欢迎参与讨论,请在这里发表您的看法、交流您的观点。

Powered By Z-BlogPHP 1.7.3

Copyright 2024-2027 pukuimin Rights Reserved.
粤ICP备17100155号