Question

我正在尝试根据用户输入频率（每日，每月，每年）返回一些报告数据。我的LINQ如下：

import threading
from time import sleep

a = []

def input_task():
    while True:
        new_val = raw_input("Enter a new value: ")
        a.append(new_val)
        print "Your current list is", a

def clear_task():
    while True:
        del a[:]
        sleep(5)

input_thread = threading.Thread(target=input_task)
clear_thread = threading.Thread(target=clear_task)

input_thread.start()
clear_thread.start()

input_thread.join()
clear_thread.join()

这可以正常工作，但问题是在从SQL检索所有数据后完成分组。未来事件的数量可能增加到数十万，我预计性能会下降。

问题：是否可以编写我的查询，以便我在服务器端级别进行分组？（完全使用LINQ2SQL，而不是降级到LINQ2Object）

Answer 1

我设法找到两种方法，但不是Invoke试验那么小。

0）我用来存储数据的一些POCO

public class AuditReportGroupingDataModelBase
{
    public DateTime GroupTime { get; set; }
    public int CountryId { get; set; }
}

public class AuditReportGroupingDataModel : AuditReportGroupingDataModelBase
{
    public int Count { get; set; }
}

1）丑陋的方式 - 在GroupBy中使用条件运算符

我的少数可能性允许三元运算符使用。但是，对于增加的选项数量，这不能正常工作。

var groupedDataQuery = dataQuery
    .GroupBy(e => new AuditReportGroupingDataModelBase
    {
        GroupTime = (filters.TimeSpanId == (int)TimeSpanEnum.Daily ? e.InsertDay : filters.TimeSpanId == (int)TimeSpanEnum.Monthly ? e.InsertMonth : e.InsertDay).Value,
        CountryId = e.CountryId.Value
    })
    .Select(grp => new AuditReportGroupingDataModel
    {
        GroupTime = grp.Key.GroupTime,
        CountryId = grp.Key.CountryId,
        Count = grp.Count()
    });

这样可行，但会产生一个丑陋且不那么有效的SQL语句：

exec sp_executesql N'SELECT 
    1 AS [C1], 
    [GroupBy1].[K2] AS [C2], 
    [GroupBy1].[K1] AS [CountryId], 
    [GroupBy1].[A1] AS [C3]
    FROM ( SELECT 
        [Filter1].[K1] AS [K1], 
        [Filter1].[K2] AS [K2], 
        COUNT([Filter1].[A1]) AS [A1]
        FROM ( SELECT 
            [Extent1].[CountryId] AS [K1], 
            CASE WHEN (1 = @p__linq__0) THEN [Extent1].[InsertDay] WHEN (2 = @p__linq__1) THEN [Extent1].[InsertMonth] ELSE [Extent1].[InsertDay] END AS [K2], 
            1 AS [A1]
            FROM [dbo].[AppEvent] AS [Extent1]
            WHERE ([Extent1].[EventTypeId] IN (1)) AND ([Extent1].[CountryId] IS NOT NULL)
        )  AS [Filter1]
        GROUP BY [K1], [K2]
    )  AS [GroupBy1]',N'@p__linq__0 int,@p__linq__1 int',@p__linq__0=1,@p__linq__1=1

2）更好的方法 - 基于值的GroupBy表达式

        IQueryable<IGrouping<AuditReportGroupingDataModelBase, AppEvent>> groupedDataQueryInterm = null;
        if (filters.TimeSpanId == (int)TimeSpanEnum.Daily) groupedDataQueryInterm = dataQuery.GroupBy(e => new AuditReportGroupingDataModelBase { GroupTime = e.InsertDay.Value, CountryId = e.CountryId.Value });
        if (filters.TimeSpanId == (int)TimeSpanEnum.Monthly) groupedDataQueryInterm = dataQuery.GroupBy(e => new AuditReportGroupingDataModelBase { GroupTime = e.InsertMonth.Value, CountryId = e.CountryId.Value });
        if (filters.TimeSpanId == (int)TimeSpanEnum.Yearly) groupedDataQueryInterm = dataQuery.GroupBy(e => new AuditReportGroupingDataModelBase { GroupTime = e.InsertYear.Value, CountryId = e.CountryId.Value });
        if (groupedDataQueryInterm == null)
            throw new InvalidEnumArgumentException($@"Invalid value provided to {nameof(filters.TimeSpanId)}");

        var groupedDataQuery = groupedDataQueryInterm
            .Select(grp => new AuditReportGroupingDataModel
            {
                GroupTime = grp.Key.GroupTime,
                CountryId = grp.Key.CountryId,
                Count = grp.Count()
            })

这会产生更好的SQL：

SELECT 
    1 AS [C1], 
    [GroupBy1].[K2] AS [InsertDay], 
    [GroupBy1].[K1] AS [CountryId], 
    [GroupBy1].[A1] AS [C2]
    FROM ( SELECT 
        [Extent1].[CountryId] AS [K1], 
        [Extent1].[InsertDay] AS [K2], 
        COUNT(1) AS [A1]
        FROM [dbo].[AppEvent] AS [Extent1]
        WHERE ([Extent1].[EventTypeId] IN (1)) AND ([Extent1].[CountryId] IS NOT NULL)
        GROUP BY [Extent1].[CountryId], [Extent1].[InsertDay]
    )  AS [GroupBy1]

Linq2SQL组由运行时已知的属性组成

1 个答案: