C# LINQtoSQL为类似的GROUPBY表达式生成不同的查询

我注意到，在LINQtoSQL中使用GroupBy时，将引用Id作为键与使用实际导航属性作为键在结果查询中存在差异

例1： Employees.GroupByx=>x.CompanyId.Selectg=>g.Count

结果SQL:

SELECT COUNT(*) AS [value]
FROM [Employees] AS [t0]
GROUP BY [t0].[CompanyId]

SELECT [t1].[value]
FROM (
    SELECT COUNT(*) AS [value], [t0].[DivisionDeductionID]
    FROM [CheckDeductions] AS [t0]
    GROUP BY [t0].[DivisionDeductionID]
) AS [t1]
LEFT OUTER JOIN [DivisionDeductions] AS [t2] ON [t2].[DivisionDeductionID] = [t1].[DivisionDeductionID]

例2： Employees.GroupByx=>x.Company.Selectg=>g.Count

结果SQL:

SELECT COUNT(*) AS [value]
FROM [Employees] AS [t0]
GROUP BY [t0].[CompanyId]

SELECT [t1].[value]
FROM (
    SELECT COUNT(*) AS [value], [t0].[DivisionDeductionID]
    FROM [CheckDeductions] AS [t0]
    GROUP BY [t0].[DivisionDeductionID]
) AS [t1]
LEFT OUTER JOIN [DivisionDeductions] AS [t2] ON [t2].[DivisionDeductionID] = [t1].[DivisionDeductionID]

---

看看示例2，很明显，除了左连接本身之外，从未使用过[t2]。为什么LINQtoSQL没有检测到这一点，而只是使用与示例1相同的查询？无论如何，它是按ID字段分组的。

这看起来像是EF的SQL生成器错过了优化查询的机会：事实上，由于[t2]没有在外部联接之外使用，它可能会与嵌套的select一起被丢弃

EF编写者似乎为[t2]添加了一个联接，因为他们不想区分以下两种情况：1导航属性仅用于其主键，因此可以使用相应的FK替换；2查询实际从中提取其他字段

考虑到RDBMS优化了不必要的连接，这种做法是完全合理的