C#在执行LINQ查询表达式的时候,并不会指定其执行语义,而是将查询表达式转换为遵循查询表达式模式的方法的调用。具体而言,查询表达式将转换为以下名称的调用:Where、Select、SelectMany、Join、GroupJoin、OrderBy、OrderByDescending、ThenBy、ThenByDescending、GroupBy、Cast等等。
如同在前文中提到的用扩展方法和Lambda表达式简化LINQ查询表达式一般,这也是对查询表达式的一种转换。简化后的方法其实就是LINQ查询的实际执行。
本文中用来示例的代码,参数设定都沿用上一篇文章的设定。
从查询表达式到方法调用的转换是一种句法映射,在执行任何类型绑定或重载决策之前发生。该转换可以保证在句法上正确,但不能保证生成语法正确的 C# 代码。转换查询表达式后,以常规方法调用的方式处理生成的方法调用,而这进而可能暴露错误,例如在方法不存在、参数类型错误或方法为泛型方法且类型推断失败这些情况下。
不允许对查询表达式中的范围变量进行赋值。但允许 C# 实现在某些时候以不实施此限制,因为对于此处介绍的句法转换方案,有些时候可能根本无法实施此限制。
某些转换使用由 * 指示的透明标识符注入范围变量。
继续符是指 into 操作符,带有继续符的查询表达式类似如下:
from ···into x ···
转换为
from x in (from ···) ···
示例:
from c in customers
group c by c.Country into g
select new { Country=c.Country , Customer = g.Count()}
转换为
from g in
from c in customers
group c by c.Country
select new { Country = g.Key, CustCount = g.Count() }
最终转换为
customers.
GroupBy(c => c.Country).
Select(g => new { Country = g.Key, CustCount = g.Count() })
显式指定范围变量类型的 from 子句
from T x in e
转换为
from x in (e).Cast<
T>
()
显式指定范围变量类型的 join 子句
join T x in e on k1 equals k2
转换为
join x in (e).Cast<
T>
()
示例:
from Customer in customers
where c.City == "London"
select c
转换为
from c in customers.Cast<Customer>()
where c.City == "London"
select c
最终转换为
customers.Cast<Customer>().Where(c=>c.City=="London")
显式范围变量类型对于查询实现非泛型 IEnumerable 接口的集合很有用,但对于实现泛型IEnumerable 接口的集合没什么用处。如果 customers 属于 ArrayList 类型,则在面的示例中即会如此。
退化查询表达式,是指选择源元素本身的查询,如:
from c in customers
select c
确保查询表达式的结果永不为源对象本身非常重要,因为这样会向查询的客户端暴露源的类型和标识符。因此,在查询表达式为退化查询的时候,可通过在源上显式调用 Select 来保护直接以源代码形式写入的简并查询。然后,由 Select 实施者及其他查询操作员确保这些方法永远不会返回源对象本身。
退化查询表达式如下:
from x in e select x
转换为
(e).Select(x=>x)
示例:
from c in customers
select c
转换为
customers.Select(c => c)
带有另一个 from 子句且后接一个 select 子句的查询表达式
from x1 in e1
from x2 in e2
select v
转换为
(e1).SelectMany( x1 => e2 , ( x1 , x2 ) => v )
带有另一个 from 子句且后接一个 select 子句的查询表达式
from x1 in e1
from x2 in e2
···
转换为
from *
in (e1).SelectMany( x1 => e2 , ( x1 , x2 ) => new { x1 , x2 })
带有 let 子句的查询表达式
from x in e
let y=f
转换为
from *
in (e).Select( x => new { x , y = f })
带有 where 子句的查询表达式
from x in e
where f
···
转换为
from x in (e).Where( x => f )
带有 join 子句(不含 into)且后接 select 子句的查询表达式
from x1 in e1
join x2 in e2 on k1 equals k2
select v
转换为
( e1 ) . Join( e2 , x1 => k1 , x2 => k2 , ( x1 , x2 ) => v )
带有 join 子句(不含 into)且后接除 select 子句之外的其他内容的查询表达式
from x1 in e1
join x2 in e2 on k1 equals k2
…
转换为
from *
in ( e1 ) . Join(
e2 , x1 => k1 , x2 => k2 , ( x1 , x2 ) => new { x1 , x2 })
…
带有 join 子句(含 into)且后接 select 子句的查询表达式
from x1 in e1
join x2 in e2 on k1 equals k2 into g
select v
转换为
( e1 ) . GroupJoin( e2 , x1 => k1 , x2 => k2 , ( x1 , g ) => v )
带有 join 子句(含 into)且后接除 select 子句之外的其他内容的查询表达式
from x1 in e1
join x2 in e2 on k1 equals k2 into g
…
转换为
from *
in ( e1 ) . GroupJoin(
e2 , x1 => k1 , x2 => k2 , ( x1 , g ) => new { x1 , g })
…
带有 orderby 子句的查询表达式
from x in e
orderby k1 , k2 , … , kn
…
转换为
from x in ( e ) .
OrderBy ( x => k1 ) .
ThenBy ( x => k2 ).
… .
ThenBy ( x => kn )
…
如果排序子句指定 descending 方向指示器,则将改为生成对 OrderByDescending 或
ThenByDescending 的调用。
我们假定,在下面的每个查询表达式中没有 let、 where、 join 或 orderby 子句,并且最多只有一个初始 from 子句。
示例1:
from c in customers
from o in c.Orders
select new { c.Name, o.OrderID, o.Total }
转换为
customers.
SelectMany(c => c.Orders,(c,o) => new { c.Name, o.OrderID, o.Total })
示例2:
from c in customers
from o in c.Orders
orderby o.Total descending
select new { c.Name, o.OrderID, o.Total }
转换为
from * in customers.SelectMany(c => c.Orders, (c,o) => new { c, o })
orderby o.Total descending
select new { c.Name, o.OrderID, o.Total }
最终转换为
customers.SelectMany(c => c.Orders, (c,o) => new { c, o }).OrderByDescending(x => x.o.Total).Select(x => new { x.c.Name, x.o.OrderID, x.o.Total })
其中 x 是编译器生成的以其他方式不可见且不可访问的标识符。
示例3:
from o in orders
let t = o.Details.Sum(d => d.UnitPrice * d.Quantity)
where t >= 1000
select new { o.OrderID, Total = t }
转换为
from * in orders.
Select(o => new { o, t = o.Details.Sum(d => d.UnitPrice * d.Quantity) })
where t >= 1000
select new { o.OrderID, Total = t }
最终转换为
orders.Select(o => new { o, t = o.Details.Sum(d => d.UnitPrice * d.Quantity) }).Where(x => x.t >= 1000).Select(x => new { x.o.OrderID, Total = x.t })
其中 x 是编译器生成的以其他方式不可见且不可访问的标识符。
示例4:
from c in customers
join o in orders on c.CustomerID equals o.CustomerID
select new { c.Name, o.OrderDate, o.Total }
转换为
customers.Join(orders, c => c.CustomerID, o => o.CustomerID,(c, o) => new { c.Name, o.OrderDate, o.Total })
示例5:
from c in customers
join o in orders on c.CustomerID equals o.CustomerID into co
let n = co.Count()
where n >= 10
select new { c.Name, OrderCount = n }
转换为
from * in customers.GroupJoin(orders, c => c.CustomerID, o => o.CustomerID,
(c, co) => new { c, co })
let n = co.Count()
where n >= 10
select new { c.Name, OrderCount = n }
最终转换为
customers.GroupJoin(orders, c => c.CustomerID, o => o.CustomerID,(c, co) => new { c, co }).Select(x => new { x, n = x.co.Count() }).Where(y => y.n >= 10).Select(y => new { y.x.c.Name, OrderCount = y.n)
其中 x 和 y 是编译器生成的以其他方式不可见且不可访问的标识符。
示例6:
from o in orders
orderby o.Customer.Name, o.Total descending
select o
转换为
orders.OrderBy(o => o.Customer.Name).ThenByDescending(o => o.Total)
from x in e select v
转换为
( e ) . Select ( x => v )
当 v 为标识符 x 时,转换仅为
( e )
from x in e group v by k
转换为
( e ) . GroupBy ( x => k , x => v )
当 v 为标识符 x 时,转换为
( e ) . GroupBy ( x => k )