当前位置 : 主页 > 网络推广 > seo >

Lucene.Net 全文索引笔记

来源:互联网 收集:自由互联 发布时间:2021-06-16
Lucene.Net用了又忘...由于现在信息量爆炸,用过的东西用完就忘,只好自己写个笔记来记录一下了... 1: 需要DLL Lucene.Net.dll PanGu.dll PanGu.HighLight.dll PanGu.Lucene.Analyzer.dll 没有的话,可以去我

Lucene.Net用了又忘...由于现在信息量爆炸,用过的东西用完就忘,只好自己写个笔记来记录一下了...

1: 需要DLL

 Lucene.Net.dll

PanGu.dll

PanGu.HighLight.dll

PanGu.Lucene.Analyzer.dll

没有的话,可以去我的资源包里面下,地址如下: http://download.csdn.net/download/kimizhou_blog/10016313


2;生成索引


string indexPath = Context.Server.MapPath("~/App_Data/IndexData");//索引文档保存位置
string commonProductIndexPath = string.Format("{0}/{1}", indexPath, "commonProduct");    //积分商城产品
 //开始处理 积分商城产品索引
CreateCommonProductIndex(commonProductIndexPath);

然后看看CreateCommonProductIndex方法

      /// <summary>
        /// 创建积分商城产品索引
        /// </summary>
        /// <param name="indexPath"></param>
        private void CreateCommonProductIndex(string indexPath)
        {
            FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NativeFSLockFactory());//绑定索引目录
            bool isExist = IndexReader.IndexExists(directory);
            if (isExist)
            {
                if (IndexWriter.IsLocked(directory))
                {
                    IndexWriter.Unlock(directory);
                }
            }
            IndexWriter writer = new IndexWriter(directory, new PanGuAnalyzer(), !isExist, IndexWriter.MaxFieldLength.UNLIMITED);
            writer.DeleteAll();//先删之前的索引
            IList<ProductInfoByIndex> list = Product.GetProductListByIndex();
            foreach (var item in list)
            {
                Document document = new Document(); 
                document.Add(new Field("id", item.ProductID.ToString(), Field.Store.YES, Field.Index.ANALYZED));//--所有字段的值都将以字符串类型保存 因为索引库只存储字符串类型数据
                string Content = string.Format("{0}", item.ProductName);
                document.Add(new Field("Content", Content, Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS));
                writer.AddDocument(document); //文档写入索引库
            }
            writer.Close();//会自动解锁
            directory.Close(); //不要忘了Close,否则索引结果搜不到
        }

 其中
IList<ProductInfoByIndex> list = Product.GetProductListByIndex();方式是去数据中读取这个list对象,这里代码就不贴出来了。到这里你的索引已经创建出来的,那么接下来需要查询和显示

查询是最困难的,各种匹配


3:查询索引并且显示出来


GetProductIndex方法就是获取索引代码如下:

    /// <summary>
        /// 获取积分商品索引
        /// </summary>
        private void GetProductIndex()
        {
            string indexPath = Context.Server.MapPath("~/App_Data/IndexData");//索引文档保存位置
            string commonProductIndexPath = string.Format("{0}/{1}", indexPath, "commonProduct");    //积分商城产品
            FSDirectory directory = FSDirectory.Open(new DirectoryInfo(commonProductIndexPath), new NoLockFactory());
            IndexReader reader = IndexReader.Open(directory, true);
            IndexSearcher searcher = new IndexSearcher(reader);

            BooleanQuery bQuery = new BooleanQuery();
            foreach (string word in SplitContent.SplitWords(Request["SearchKey"]))
            {
                Query queryUseringNatrue = new WildcardQuery(new Term("Content", "*" + word + "*"));
                bQuery.Add(queryUseringNatrue, BooleanClause.Occur.MUST);// MUST 必须
            }

            Sort sort = new Sort(new SortField("id", SortField.FLOAT, true)); //true为降序排序 
            TopDocs docs = searcher.Search(bQuery, (Filter)null, 9999999, sort);
            List<ProductInfoByIndex> proList = new List<ProductInfoByIndex>();
            for (int i = 0; i < docs.totalHits; i++)
            {
                Document doc = searcher.Doc(docs.scoreDocs[i].doc);
                ProductInfoByIndex product = new ProductInfoByIndex();
                product.ProductID = System.Convert.ToInt32(doc.Get("id"));
                product.ProductName = doc.Get("Content");
                //product.ProductName = SplitContent.HightLight(Request["SearchKey"], doc.Get("Content"));
                proList.Add(product);
            }
            productResultList = proList;
            this.Message += string.Format("|{0}条积分商城产品", docs.totalHits);


            //PhraseQuery query = new PhraseQuery();



            //foreach (string word in SplitContent.SplitWords(Request["SearchKey"]))
            //{
            //    query.Add(new Term("Content", word));
            //}
            //query.SetSlop(100);
            //TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);
            //searcher.Search(query, null, collector);
            //ScoreDoc[] docs = collector.TopDocs(0, collector.GetTotalHits()).scoreDocs;

            //List<ProductInfoByIndex> proList = new List<ProductInfoByIndex>();
            //for (int i = 0; i < docs.Length; i++)
            //{
            //    int docId = docs[i].doc;//得到查询结果文档的id(Lucene内部分配的id)
            //    Document doc = searcher.Doc(docId);//根据文档id来获得文档对象Document
            //    ProductInfoByIndex product = new ProductInfoByIndex();
            //    product.ProductID = System.Convert.ToInt32(doc.Get("id"));
            //    //book.ContentDescription = doc.Get("content");//未使用高亮
            //    //搜索关键字高亮显示 使用盘古提供高亮插件
            //    product.ProductName = SplitContent.HightLight(Request["SearchKey"], doc.Get("Content"));
            //    proList.Add(product);
            //}
            //    productResultList = proList;
            //this.Message += string.Format("|{0}条积分商城产品", docs.Length);
        }

其中我注释掉的,是另外一种方法,这里我用的效率比较慢的模糊查询 
Query queryUseringNatrue = new WildcardQuery(new Term("Content", "*" + word + "*"));
这个类似 数据库的like '%关键字%'
到这里就已经获取到了所有的索引资料了,是不是很简单,你get到了吗?最后我再给大家介绍索引的几种查询方式:
 第1种:

  //string keyWordUseringNatrue = "营运";
            //if (!string.IsNullOrWhiteSpace(keyWordUseringNatrue))
            //{
            //    QueryParser parseUseringNatrue = new QueryParser("UseringNatrue", new PanGuAnalyzer());
            //    Query query = parseUseringNatrue.Parse(keyWordUseringNatrue);
            //    parseUseringNatrue.SetDefaultOperator(QueryParser.Operator.AND);
            //    bQuery.Add(query, BooleanClause.Occur.MUST);
            //}
            //营运
            //Query queryUseringNatrue = new WildcardQuery(new Term("UseringNatrue", "营运")); 
            //bQuery.Add(queryUseringNatrue, BooleanClause.Occur.MUST);// MUST 必须

这个查询是什么呢?是一般的查询,会查询出运营相关的,但是他和like不一样,他跟分词有关,比如说,“爱” 就查询不出 “可爱” ,pangu有自己的分词,但是这个比较常用,下面汇总一下其它的查询:
其它查询汇总:

//介绍各种Query
            //TermQuery: 首先介绍最基本的查询,如果你想执行一个这样的查询:在content字段中查询包含‘刘备的document”,那么你可以用TermQuery:
            // Term t = new Term("content", "刘备");
            // Query query = new TermQuery(t);

            //BooleanQuery :如果你想这么查询:在content字段中包含”刘备“并且在title字段包含”三国“的document”,那么你可以建立两个TermQuery并把它们用BooleanQuery连接起来:
            //1             TermQuery termQuery1 = new TermQuery(new Term("content", "刘备"));
            //2             TermQuery termQuery2 = new TermQuery(new Term("title", "三国"));
            //3             BooleanQuery booleanQuery = new BooleanQuery();
            //4             booleanQuery.Add(termQuery1, BooleanClause.Occur.SHOULD);
            //5             booleanQuery.Add(termQuery2, BooleanClause.Occur.SHOULD);

            //WildcardQuery :如果你想对某单词进行通配符查询,你可以用WildcardQuery,通配符包括’?’匹配一个任意字符和’*’匹配零个或多个任意字符,例如你搜索’三国*’,你可能找到’三国演义’或者’三国志’:
            //1             Query query = new WildcardQuery(new Term("content", "三国*"));

            //PhraseQuery :你可能对中日关系比较感兴趣,想查找‘中’和‘日’挨得比较近(5个字的距离内)的文章,超过这个距离的不予考虑,你可以
            //1             PhraseQuery query = new PhraseQuery();
            //2             query.SetSlop(5);
            //3             query.Add(new Term("content ", "中"));
            //4             query.Add(new Term("content", "日"));

            //那么它可能搜到“中日合作……”、“中方和日方……”,但是搜不到“中国某高层领导说日本欠扁”
            //PrefixQuery :如果你想搜以‘中’开头的词语,你可以用PrefixQuery:
            //1             PrefixQuery query = new PrefixQuery(new Term("content ", "中"));

            //FuzzyQuery :FuzzyQuery用来搜索相似的term,使用Levenshtein算法。假设你想搜索跟‘wuzza’相似的词语,你可以:
            //1             Query query = new FuzzyQuery(new Term("content", "wuzza"));
            //你可能得到‘fuzzy’和‘wuzzy’。

            //RangeQuery: 另一个常用的Query是RangeQuery,你也许想搜索时间域从20060101到20060130之间的document,你可以用RangeQuery:
            //1             RangeQuery query = new RangeQuery(new Term("time","20060101"), new Term("time","20060130"), true);
            //最后的true表示用闭合区间。


因为各个版本,他们使用的都不太一样,下面介绍一种常用的读取以后显示的方式,其中Sort就是排序

Stopwatch stopwath = new Stopwatch();//秒表 
            Sort sort = new Sort(new SortField("CarPrice", SortField.FLOAT,true)); //true为降序排序 CarPrice为价格 SortField.DOC是?
            TopDocs docs = searcher.Search(bQuery, (Filter)null, 9999999, sort);
            stopwath.Stop();//秒表停止
            long lSearchTime = stopwath.ElapsedMilliseconds;//耗时

            List<CarSourceInfoByIndex> carSourceResult = new List<CarSourceInfoByIndex>();
            for (int i=0;i<docs.totalHits; i++)
            {
                Document doc = searcher.Doc(docs.scoreDocs[i].doc);
                CarSourceInfoByIndex carSource = new CarSourceInfoByIndex()
                {
                    Id = int.Parse(doc.Get("Id")),
                    CarPrice = System.Convert.ToDouble(doc.Get("CarPrice")),
                    Recommended = SplitContent.HightLight(Request["SearchKey"], doc.Get("Content"))
                 };
                carSourceResult.Add(carSource);
            }
            carSourceResultList2 = carSourceResult;
            this.Message += string.Format("{0}条测试", docs.totalHits);
就到这里了,不懂的可以加我QQ 10200454咨询
网友评论