首先去官网下载Elasticsearch 2.3.4安装包,解压后，在cmd命令行进入安装目录，再进入 bin目录，运行elasticsearch.bat命令。
elasticsearch插件elasticsearch-head安装：
bin目录下执行命令plugin -install mobz/elasticsearch-head
然后开始.net编程,构建控制台应用程序
Program.cs代码如下:
class Program { static void Main(string[] args) { Console.WriteLine("*Program 开始运行 : " + DateTime.Now); var business = new Business(); var swRead = new Stopwatch(); //swRead.Start(); //business.AddToDb();//sqlserver数据库增加数据 //swRead.Stop(); //Console.WriteLine("DB 写入时间 : " + swRead.ElapsedMilliseconds); //swRead.Reset(); //swRead.Start(); //business.AddToElasticIndex(); //swRead.Stop(); //Console.WriteLine("ES 写入时间 : " + swRead.ElapsedMilliseconds); var sw = new Stopwatch(); sw.Start(); var personsFromDB = business.GetFromDB(); sw.Stop(); Console.WriteLine("DB 读时间 : " + sw.ElapsedMilliseconds); sw.Reset(); sw.Start(); var personsFromEs = business.GetFromES(); sw.Stop(); Console.WriteLine("ES 读时间 : " + sw.ElapsedMilliseconds); Console.ReadLine(); } }
BLL层的Business.cs类:
public class Business { private List<PersonDetail> _personList = new List<PersonDetail>(); //SQLSERVER数据库 PersonDbProvider dbProvider = new PersonDbProvider(); //ElasticSearch ESProvider esProvider = new ESProvider(); public void AddToDb() { _personList = Util.Get10000PersonDetails();//辅助类,生成10000条数据 foreach (var personDetail in _personList) { dbProvider.AddPerson(personDetail); } } public void AddToElasticIndex() { _personList = Util.Get10000PersonDetailsWithID(); foreach (var personDetail in _personList) { esProvider.Index(personDetail); } } public List<PersonDetail> GetFromDB() { return dbProvider.GetAllPersonDetails(); } public List<PersonDetail> GetFromES() { return esProvider.GetAll(); } }
PersonDbProvider.cs和ElasticSearchProvider.cs以及Util.cs,Setting.cs类:
public class PersonDbProvider { public bool AddPerson(PersonDetail personDetail) { try { //数据库上下文 using (var db = new PersonContext()) { db.PersonDetails.Add(personDetail); db.SaveChanges(); return true; } } catch (Exception) { return false; } } public List<PersonDetail> GetAllPersonDetails() { try { using (var db = new PersonContext()) { return db.PersonDetails.ToList(); } } catch (Exception) { return null; } } }
public class ESProvider { public static ElasticClient client = new ElasticClient(Setting.ConnectionSettings); public bool Index(PersonDetail person) { var client = new ElasticClient(Setting.ConnectionSettings); try { //添加数据 //在调用下面的index方法的时候，如果没有指定使用哪个index，ElasticSearch会直接使用我们在setting中的defaultIndex，如果没有，则会自动创建 var index = client.Index(person); return index.Created; } catch (Exception ex) { Console.WriteLine(" Excepton Message : " + ex.Message); } return false; } public List<PersonDetail> GetAll() { var searchResults = client.Search<PersonDetail>(s => s .From(0) .Size(10000) ); return searchResults.Documents.ToList(); } public List<PersonDetail> GetEntities(string keyword) { var client = new ElasticClient(Setting.ConnectionSettings); #region 全文搜索 keyword = String.Format("*{0}*", keyword); //默认的Operator是Or,当keyword是类似于"One Two"之类的中间有空格的时候，会被当成两个关键词搜索，然后搜索结果进行or运算 //所以我们需要根据需求来调整Operator var searchResults = client.Search<PersonDetail>(s => s .Index("elastic-search-app") .Query(q => q.QueryString(qs => qs.Query(keyword).DefaultOperator(Operator.And))) ); //-------------------------------------------------------------------------------------- //另外由于ES是分词搜索，所以当我们要用"One"来搜索完整的单词"JustOne"的时候，就必须在"One"外面添加**，类似于SQL里面的%keyword%，但是这样的做法会导致在用完整的单词来搜索的时候搜索不到结果，所以我们需要使用下面的方式 //wholeKeyword = keyword; //keyword = String.Format("*{0}*", keyword); //QueryContainer query = new QueryStringQuery() { Query = keyword, DefaultOperator = Operator.And }; //if (!String.IsNullOrEmpty(wholeKeyword)) //{ // QueryContainer wholeWordQuery = new QueryStringQuery() { Query = wholeKeyword }; // query = query || wholeWordQuery; //} //var searchResults = client.Search<Person>(s => s // .Index("zhixiao-application") // .Query(query) //); #endregion #region 指定属性搜索 //使用term Query //Term是一个被索引的精确值，也就是说Foo, foo, FOO是不相等的，因此 //在使用term query的时候要注意，term query在搜索的Field已经被索引的时候，是不支持大写的。 // QueryContainer query2 = new TermQuery { Field = item.Key, Value = item.Value.ToLower() }; //-------------------------------------------------------------------------------------- //var searchResults = client.Search<PersonDetail>(s => s // .Index("elastic-search-app") // .Query(q => q.Term(t => t.OnField(f => f.LastName == "keyword"))) //); //效果同上 //QueryContainer termQuery = new TermQuery { Field = "lastname", Value = "keyword" }; //var searchResults = client.Search<PersonDetail>(s => s // .Index("elastic-search-app") // .Query(termQuery) //); //-------------------------------------------------------------------------------------- //使用 Query String query //QueryString query一般用于全文搜索，但是也可以用于单个属性的搜索（设置DefaultField属性），queryString query可以不区分大小写。QueryString还有一个好处就是我们可以搜索一个term中的一部分， //例如lastname为"t Boterhuis 1"，那么我们可以用"terhuis"搜索到这个数据（虽然需要在外面包上**），在term query里面就做不到，因为ES把每一个属性的值都分析成一个个单独的term，提高了搜索的效率。 //keyword = "t Boterhuis 2"; //QueryContainer wholeWordQuery = new QueryStringQuery() { Query = keyword, DefaultOperator = Operator.And }; //var searchResults = client.Search<PersonDetail>(s => s // .Index("elastic-search-app") // .Query(wholeWordQuery) //); #endregion return searchResults.Documents.ToList(); } public List<PersonDetail> Sort(string keyword) { // 首先我们把原先的索引先删除了 var response = client.DeleteIndex( new DeleteIndexRequest(new IndexNameMarker() { Name = "elastic-search-app", Type = typeof(PersonDetail) })); //然后重新创建索引 var indexResult = client.CreateIndex("PD-application"); var response1 = client.Map<PersonDetail>(m => m.MapFromAttributes()); IEnumerable<PersonDetail> persons = new List<PersonDetail> { new PersonDetail() { Id = 4, FirstName = "Boterhuis-040", LastName = "Gusto-040", }, new PersonDetail() { Id = 5, FirstName = "sales@historichousehotels.com", LastName = "t Boterhuis 1", }, new PersonDetail() { Id = 6, FirstName = "Aberdeen #110", LastName = "sales@historichousehotels.com", }, new PersonDetail() { Id = 7, FirstName = "Aberdeen #110", LastName = "t Boterhuis 2", }, }; foreach (var person in persons) { client.Index(person); } var searchResults = client.Search<PersonDetail>(s => s .Index("PD-application") .Sort(sort => sort.OnField(f => f.Id).Order(SortOrder.Ascending)) ); return searchResults.Documents.ToList(); } }