站内搜索 lucenenet 和 pangu4lucene分词

注意lucene的版本不要超过3.0

下载地址  lucenenet:   http://lucenenet.apache.org/     pangu4lucene:http://pangusegment.codeplex.com/

分词代码:

private IEnumerable<string> SpliteWord(string word)
    {
        List<string> listResu = new List<string>();
        Analyzer analyzer = new PanGuAnalyzer();
        TokenStream tokenStream = analyzer.TokenStream("", new StringReader(word));
        Lucene.Net.Analysis.Token token = null;
        while ((token = tokenStream.Next()) != null)//reader.Read()//只要还有词，就不返回null
        {
            listResu.Add(token.TermText());//token.TermText()为当前分的词
        }
        return listResu;
    }

建立索引代码:

private void BuildIndex(Model model)
    {
        string indexPath = @"C:\Users\Administrator\Desktop\asp.net\articleManager\articleManagerWeb\article";//注意和磁盘上文件夹的大小写一致，否则会报错。
        FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NativeFSLockFactory());
        bool isUpdate = IndexReader.IndexExists(directory);//判断索引库是否存在
        if (isUpdate)
        {
            //如果索引目录被锁定（比如索引过程中程序异常退出），则首先解锁
            //Lucene.Net在写索引库之前会自动加锁，在close的时候会自动解锁
            //不能多线程执行，只能处理意外被永远锁定的情况
            if (IndexWriter.IsLocked(directory))
            {
                IndexWriter.Unlock(directory);//un-否定。强制解锁
            }
        }
        IndexWriter writer = new IndexWriter(directory, new PanGuAnalyzer(), !isUpdate, Lucene.Net.Index.IndexWriter.MaxFieldLength.UNLIMITED);
        Document document = new Document();//一条Document相当于一条记录
        document.Add(new Field("id", (model.Id).ToString(), Field.Store.YES, Field.Index.NOT_ANALYZED));//每个Document可以有自己的属性（字段），所有字段名都是自定义的，值都是string类型 
        document.Add(new Field("title", model.Title, Field.Store.YES, Field.Index.ANALYZED, Lucene.Net.Documents.Field.TermVector.WITH_POSITIONS_OFFSETS));//与上面不同 这里分词
        document.Add(new Field("msg", model.Msg, Field.Store.YES, Field.Index.ANALYZED, Lucene.Net.Documents.Field.TermVector.WITH_POSITIONS_OFFSETS));
        writer.DeleteDocuments(new Term("id", model.Id.ToString()));//防止存在的数据//delete from t where id=i
        //如果不存在则删除0条
        writer.AddDocument(document);//把文档写入索引库


        writer.Close();
        directory.Close();//不要忘了Close，否则索引结果搜不到
    }

查找代码:

 protected void btnSearch_Click(object sender, EventArgs e)
    {
        string indexPath = @"C:\Users\Administrator\Desktop\asp.net\articleManager\articleManagerWeb\article";

        string kw = txtSearch.Text;
        FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());//打开目录中的文件
        IndexReader reader = IndexReader.Open(directory, true);
        IndexSearcher searcher = new IndexSearcher(reader);
        PhraseQuery query = new PhraseQuery();//查询条件
        foreach (var ky in SpliteWord(kw))
        {
            query.Add(new Term("msg", ky));//where contains("msg",kw)
        }       
        query.SetSlop(100);//两个词的距离大于100（经验值）就不放入搜索结果，因为距离太远相关度就不高了
        TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);//盛放查询结果的容器
        searcher.Search(query, null, collector);//使用query这个查询条件进行搜索，搜索结果放入collector
        //collector.GetTotalHits()总的结果条数
        ScoreDoc[] docs = collector.TopDocs(0, collector.GetTotalHits()).scoreDocs;//从查询结果中取出第m条到第n条的数据

        List<Model> list = new List<Model>();
        for (int i = 0; i < docs.Length; i++)//遍历查询结果
        {
            int docId = docs[i].doc;//拿到文档的id。因为Document可能非常占内存（DataSet和DataReader的区别）
            //所以查询结果中只有id，具体内容需要二次查询
            Document doc = searcher.Doc(docId);//根据id查询内容。放进去的是Document，查出来的还是Document
            //Console.WriteLine(doc.Get("id"));
            //Console.WriteLine(doc.Get("msg"));
            Model result = new Model();
            result.Id = Convert.ToInt64(doc.Get("id"));
            result.Title = doc.Get("title");//只有 Field.Store.YES的字段才能用Get查出来
            result.Msg = doc.Get("msg");//只有 Field.Store.YES的字段才能用Get查出来
            list.Add(result);
        }

        Repeater1.DataSource = list;
        Repeater1.DataBind();
    }