linq group by having 实现

TO XML ，优化的重点还是在细节，细节决定成败一点也没错

    System.Diagnostics.Stopwatch st1 = new System.Diagnostics.Stopwatch();
            st1.Start();
            XDocument xdom = XDocument.Load(XmlPath);
            var restElements = xdom.Descendants("rest").ToList();
            st1.Stop();
            HttpContext.Current.Response.Write("1.载入XML时间" + st1.ElapsedMilliseconds + "<br>");

载入一个大一点的文件，我这里显示花时间 1700毫秒 左右

           System.Diagnostics.Stopwatch st = new System.Diagnostics.Stopwatch();
            st.Start();
            var Hotels = (from hotels in restElements
                          where  
                          hotels.Element("lat") != null &&   
                          hotels.Element("lat").Value != string.Empty 
                          select new
                          {
                              State = hotels.Element("state").Value,
                              StateCityKey = (hotels.Element("state").Value + "-" + hotels.Element("city").Value).ToUpper()
                          }).ToList();

过滤，，生成匿名对象，然后Tolist();这一步的没什么好说的，关键是不要让他监视XML文件，不要让他延时加载，生成一个KEY

            var hotelsGroup =
                (from hotel in Hotels
                 group hotel by hotel.StateCityKey
                     into h
                     select new
                     {
                         StateCityKey = h.Key,
                         count = h.Count(),
                         hgrou = h
                     } into c
                     where c.count > 2
                     select c).ToList();
            st.Stop();
            HttpContext.Current.Response.Write("2.分组耗费" + st.ElapsedMilliseconds);
            long a = st1.ElapsedMilliseconds + st.ElapsedMilliseconds;
            HttpContext.Current.Response.Write("<br>1+2总耗费 ：" + a + "毫秒<br>");

 分组耗时：40毫秒

通过KEY（StateCityKey）进行分组，注意COUNT，where c.count > 2 对分组以后的数据进行判断，相当于HAVING

用的不是COUNT（） 是count  ，

下面的循环统计输出就不放出来了。。。

 转载请放链接 http://www.cnblogs.com/jacd/archive/2012/04/25/2469389.html