Linq学习-基本查询
Posted
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了Linq学习-基本查询相关的知识,希望对你有一定的参考价值。
一、本将主要介绍内容
从linq,sql,lambda三个角度比较来学习
select、orderby、分页、group by、distinct、子查询、in的用法
1.select
查询用户和它们的自我介绍
Linq to sql
from a in Blog_UserInfo select new { 真实名字=a.RealName, 自我介绍=a.Introduce }
sql
SELECT [t0].[RealName] AS [真实名字], [t0].[Introduce] AS [自我介绍] FROM [Blog_UserInfo] AS [t0]
Lambda
Blog_UserInfo .Select ( a => new { 真实名字 = a.RealName, 自我介绍 = a.Introduce } )
2.orderby
查询名字里带friend的用户,并排序
Linq to sql
from a in Blog_Users where a.NickName.Contains("Friend") orderby a.UserId ascending, a.CreateTime descending select a --或者 from a in Blog_Users where a.NickName.Contains("Friend") orderby a.UserId,a.CreateTime select a
sql
-- Region Parameters DECLARE @p0 NVarChar(1000) = ‘%Friend%‘ -- EndRegion SELECT [t0].[UserId], [t0].[NickName], [t0].[CreateTime] FROM [Blog_User] AS [t0] WHERE [t0].[NickName] LIKE @p0 ORDER BY [t0].[UserId], [t0].[CreateTime] DESC
Lambda
Blog_Users .Where (a => a.NickName.Contains ("Friend")) .OrderBy (a => a.UserId) .ThenByDescending (a => a.CreateTime)
3.分页
按照每页2条 ,查询第2页的留言表的信息
Linq to sql
(from a in Blog_LeaveMsgs select a).Skip(2).Take(2)
sql
-- Region Parameters DECLARE @p0 Int = 2 DECLARE @p1 Int = 2 -- EndRegion SELECT [t1].[ID], [t1].[ReceiverId], [t1].[LeaverId], [t1].[CreateTime], [t1].[Content] FROM ( SELECT ROW_NUMBER() OVER (ORDER BY [t0].[ID], [t0].[ReceiverId], [t0].[LeaverId], [t0].[CreateTime], [t0].[Content]) AS [ROW_NUMBER], [t0].[ID], [t0].[ReceiverId], [t0].[LeaverId], [t0].[CreateTime], [t0].[Content] FROM [Blog_LeaveMsg] AS [t0] ) AS [t1] WHERE [t1].[ROW_NUMBER] BETWEEN @p0 + 1 AND @p0 + @p1 ORDER BY [t1].[ROW_NUMBER]
Lambda
Blog_LeaveMsgs .Select (a => a) .Skip (2) .Take (2)
4.1分组1(group by字段)
根据用户来分组,查询留言数大于等于3条的用户ID和相应留言数量
Linq to sql
from a in Blog_LeaveMsgs group a by a.LeaverId into b where b.Count() >=3 select new { 朋友ID = b.Key, 留言数 = b.Count() }
sql
-- Region Parameters DECLARE @p0 Int = 3 -- EndRegion SELECT [t1].[LeaverId] AS [朋友ID], [t1].[value2] AS [留言数] FROM ( SELECT COUNT(*) AS [value], COUNT(*) AS [value2], [t0].[LeaverId] FROM [Blog_LeaveMsg] AS [t0] GROUP BY [t0].[LeaverId] ) AS [t1] WHERE [t1].[value] >= @p0
4.2分组2(group by多个字段)
按照接收人和留言人进行分组,查看覆盖的接收人和留言人情况
Linq to sql
from a in Blog_LeaveMsgs group a by new{a.ReceiverId,a.LeaverId} into b select new { 接收人ID=b.Key.ReceiverId, 留言人ID=b.Key.LeaverId }
sql
SELECT [t0].[ReceiverId] AS [接收人ID], [t0].[LeaverId] AS [留言人ID] FROM [Blog_LeaveMsg] AS [t0] GROUP BY [t0].[ReceiverId], [t0].[LeaverId]
Lambda
Blog_LeaveMsgs .GroupBy ( a => new { ReceiverId = a.ReceiverId, LeaverId = a.LeaverId } ) .Select ( b => new { 接收人ID = b.Key.ReceiverId, 留言人ID = b.Key.LeaverId } )
5.distinct
查看留言表中的留言人人数
Linq to sql
(from a in Blog_LeaveMsgs select a.LeaverId) .Distinct()
sql
SELECT DISTINCT [t0].[LeaverId] FROM [Blog_LeaveMsg] AS [t0]
Lambda
Blog_LeaveMsgs .Select (a => a.LeaverId) .Distinct ()
6.子查询
查询留言数量超过4条的用户信息
Linq to sql
from a in Blog_Users where (from b in Blog_LeaveMsgs group b by b.LeaverId into b where b.Count()>=4 select b.Key).Contains(a.UserId) select a
sql
-- Region Parameters DECLARE @p0 Int = 4 -- EndRegion SELECT [t0].[UserId], [t0].[NickName], [t0].[CreateTime] FROM [Blog_User] AS [t0] WHERE EXISTS( SELECT NULL AS [EMPTY] FROM ( SELECT COUNT(*) AS [value], [t1].[LeaverId] FROM [Blog_LeaveMsg] AS [t1] GROUP BY [t1].[LeaverId] ) AS [t2] WHERE ([t2].[LeaverId] = ([t0].[UserId])) AND ([t2].[value] >= @p0) )
Lambda
Blog_Users .Where ( a => Blog_LeaveMsgs .GroupBy (b => b.LeaverId) .Where (b => (b.Count () >= 4)) .Select (b => b.Key) .Contains ((Int32?)(a.UserId)) )
7.in操作
查询制定用户昵称的用户
Linq to sql
from a in Blog_Users where new string[]{"Kimisme","FriendLee"} .Contains(a.NickName) select a
sql
-- Region Parameters DECLARE @p0 NVarChar(1000) = ‘Kimisme‘ DECLARE @p1 NVarChar(1000) = ‘FriendLee‘ -- EndRegion SELECT [t0].[UserId], [t0].[NickName], [t0].[CreateTime] FROM [Blog_User] AS [t0] WHERE [t0].[NickName] IN (@p0, @p1)
Lambda
Blog_Users .Where (a => new String[] { "Kimisme", "FriendLee" } .Contains (a.NickName))
以上是关于Linq学习-基本查询的主要内容,如果未能解决你的问题,请参考以下文章