index.cs

KTDictSeg 简介: KTDictSeg 是由KaiToo搜索开发的一款基于字典的简单中英文分词算法 * 主要功能: 中英文分词

using System;
using System.Collections.Generic;
using System.Text;
using System.IO;

using Lucene.Net.Index;
using Lucene.Net.Documents;
using Lucene.Net.Analysis.Standard;
using Lucene.Net.Search;
using Lucene.Net.QueryParsers;
using Lucene.Net.Analysis;
using Lucene.Net.Highlight;
using SimpleSearch.News;

namespace SimpleSearch
{
    class Index
    {
        public static String INDEX_DIR = @"News";
        private static IndexWriter writer = null;

        public static int MaxMergeFactor
        {
            get
            {
                if (writer != null)
                {
                    return writer.GetMergeFactor();
                }
                else
                {
                    return 0;
                }
            }

            set
            {
                if (writer != null)
                {
                    writer.SetMergeFactor(value);
                }
            }
        }

        public static int MaxMergeDocs
        {
            get
            {
                if (writer != null)
                {
                    return writer.GetMaxMergeDocs();
                }
                else
                {
                    return 0;
                }
            }

            set
            {
                if (writer != null)
                {
                    writer.SetMaxMergeDocs(value);
                }
            }
        }

        public static int MinMergeDocs
        {
            get
            {
                if (writer != null)
                {
                    return writer.GetMaxBufferedDocs();
                }
                else
                {
                    return 0;
                }
            }

            set
            {
                if (writer != null)
                {
                    writer.SetMaxBufferedDocs(value);
                }
            }
        }

        public static void CreateIndex(String indexDir)
        {

            try
            {
                writer = new IndexWriter(indexDir, new Lucene.Net.Analysis.KTDictSeg.KTDictSegAnalyzer(), false);
            }
            catch
            {
                writer = new IndexWriter(indexDir, new Lucene.Net.Analysis.KTDictSeg.KTDictSegAnalyzer(), true);
            }

            //writer.Optimize();
            //writer.Close();
        }

        public static void Rebuild(String indexDir)
        {
            writer = new IndexWriter(indexDir, new Lucene.Net.Analysis.KTDictSeg.KTDictSegAnalyzer(), true);
            writer.Optimize();
            writer.Close();
        }

        public static int IndexString(String indexDir, string url, string title, DateTime time, string content)
        {
            //IndexWriter writer = new IndexWriter(indexDir, new Lucene.Net.Analysis.KTDictSeg.KTDictSegAnalyzer(), false);

            Document doc = new Document();

            Field field = new Field("url", url, Field.Store.YES, Field.Index.NO);
            doc.Add(field);
            field = new Field("title", title, Field.Store.YES, Field.Index.TOKENIZED);
            doc.Add(field);
            field = new Field("time", time.ToString("yyyyMMdd"), Field.Store.YES, Field.Index.UN_TOKENIZED);
            doc.Add(field);
            field = new Field("contents", content, Field.Store.YES, Field.Index.TOKENIZED);
            doc.Add(field);

            writer.AddDocument(doc);

            int num = writer.DocCount();
            //writer.Optimize();
            //writer.Close();
            return num;
        }

        public static void Close()
        {
            writer.Optimize();
            writer.Close();
        }

        public static List<TNews> Search(String indexDir, String q)
        {
            IndexSearcher search = new IndexSearcher(indexDir);

            QueryParser queryParser = new QueryParser("contents", new Lucene.Net.Analysis.KTDictSeg.KTDictSegAnalyzer());

            Query query = queryParser.Parse(q);

            Hits hits = search.Search(query);

            List<TNews> result = new List<TNews>();

            for (int i = 0; i < hits.Length(); i++)
            {
                try
                {
                    TNews news = new TNews();
                    news.Title = hits.Doc(i).Get("title");
                    news.Content = hits.Doc(i).Get("contents");
                    news.Url = hits.Doc(i).Get("url");
                    String strTime = hits.Doc(i).Get("time");
                    news.Time = DateTime.ParseExact(strTime, "yyyyMMdd", null);

                    // 高亮显示设置
                    //TermQuery tQuery = new TermQuery(new Term("contents", q));

                    SimpleHTMLFormatter simpleHTMLFormatter = new SimpleHTMLFormatter("<font color=\"red\">", "</font>");
                    Highlighter highlighter = new Highlighter(simpleHTMLFormatter, new QueryScorer(query));
                    //关键内容显示大小设置 
                    highlighter.SetTextFragmenter(new SimpleFragmenter(50));
                    //取出高亮显示内容
                    Lucene.Net.Analysis.KTDictSeg.KTDictSegAnalyzer analyzer = new Lucene.Net.Analysis.KTDictSeg.KTDictSegAnalyzer();
                    TokenStream tokenStream = analyzer.TokenStream("contents", new StringReader(news.Content));
                    news.Abstract = highlighter.GetBestFragment(tokenStream, news.Content);
                    result.Add(news);
                }
                catch(Exception e)
                {
                }
            }

            search.Close();
            return result;
        }
    }
}