Question

我是Linq的新手，我确信我已经以复杂的方式解决了这个问题。我试图在Linq中做这样的SQL：

SELECT DISTINCT count(vendor) as vendorCount, reqDate, status, openDate, item, poDate, count(responseDate) as responseCount FROM myTable GROUP BY reqDate, status, openDate, item, poDate HAVING reqDate > openDate

这是我到目前为止所拥有的。



        var groupQuery = (from table in dt.AsEnumerable()
                         group table by new
                         {                                 
                             vendor = table["vendor"], reqdate = table.Field<DateTime>("ReqDate"), status = table["status"],
                             open = table["openDate"],  
                             item = table["item"), 
                             podate = table.Field<DateTime>("PODate"), responsedate =  table.Field<DateTime>("responseDate"), 
                         }
                             into groupedTable
                             where Having(groupedTable.Key.reqdate, groupedTable.Key.openDate) == 1
                             select new
                             {
                                 x = groupedTable.Key,  
                                 y = groupedTable.Count()
                             }).Distinct();



        foreach (var req in groupQuery)
        {
            Console.WriteLine("cols: {0} count: {1} ",
                req.x, req.y);
        }

var groupQuery = (from table in dt.AsEnumerable() group table by new { vendor = table["vendor"], reqdate = table.Field<DateTime>("ReqDate"), status = table["status"], open = table["openDate"], item = table["item"), podate = table.Field<DateTime>("PODate"), responsedate = table.Field<DateTime>("responseDate"), } into groupedTable where Having(groupedTable.Key.reqdate, groupedTable.Key.openDate) == 1 select new { x = groupedTable.Key, y = groupedTable.Count() }).Distinct(); foreach (var req in groupQuery) { Console.WriteLine("cols: {0} count: {1} ", req.x, req.y); }

Having（）是一个函数，它接受两个日期时间参数，如果reqDate大于openDate则返回1。它编译并运行，但它显然不能给我我想要的结果。这可能使用Linq吗？我想将这些数据推送到excel电子表格，所以我希望从这个linq查询创建一个数据表。我最好只从我的数据表创建一个数据视图而不是弄乱Linq？

Answer 1

SQL代码仅按部分字段进行分组，而LINQ语句按所有字段进行分组，因此唯一可以分组的项目是重复项。如果仅按SQL查询分组的字段进行分组，则应获得正确的答案。您的Having()方法字很好，但没有必要且可读性较差。

var groupQuery = (from table in dt.AsEnumerable()
                 group table by new
                 {                                 
                     reqdate = table.Field<DateTime>("ReqDate"),
                     status = table["status"],
                     open = table["openDate"],  
                     item = table["item"), 
                     podate = table.Field<DateTime>("PODate")
                 }
                 into groupedTable
                 where groupedTable.Key.reqdate > groupedTable.Key.openDate
                 select new
                 {
                     x = groupedTable.Key,  
                     VenderCount = groupedTable.Select(t => t["vendor"])
                                               .Distinct()
                                               .Count(),
                     ResponseCount = groupedTable.Select(t => t.Field<DateTime>("responseDate"))
                                                 .Distinct()
                                                 .Count()
                 }).Distinct();

linq group by multiple aggregate

1 个答案: