SQL Server中的集合运算包括UNION
(合并),EXCEPT
(差集)和INTERSECT
(相交)三种。
集合运算的基本使用
1.UNION
(合并两个查询结果集,隐式DINSTINCT,删除重复行)
1
2
3
4
|
--合并两个提取表/派生表(derived table), 返回结果为:[a,b,c,d,e] SELECT FC FROM ( VALUES ( 'a' ),( 'b' ),( 'c' ),( 'e' )) Table1 (FC) UNION SELECT FC FROM ( VALUES ( 'a' ),( 'b' ),( 'c' ),( 'd' )) Table2 (FC) |
2.UNION ALL
(简单合并两个查询结果集,不删除重复行)
1
2
3
4
|
--提取表/派生表(derived table)可以是多列,列名、顺序可以不同,但列数必须相同 SELECT * FROM ( VALUES ( 'a' , 'Anna' ),( 'b' , 'Bob' ),( 'c' , 'Cassie' ),( 'e' , 'Elina' )) Table1 (FC, Name ) UNION ALL SELECT * FROM ( VALUES ( 'a' , 'Anna' ),( 'b' , 'Bob' ),( 'c' , 'Cassie' ),( 'd' , 'David' )) Table2 (FC, Name ) |
3.EXCEPT
(返回出现在第一个结果集但不出现在第二个结果集中的所有行)
1
2
3
4
|
--返回结果为:[e] SELECT FC FROM ( VALUES ( 'a' ),( 'b' ),( 'c' ),( 'e' )) Table1 (FC) EXCEPT SELECT FC FROM ( VALUES ( 'a' ),( 'b' ),( 'c' ),( 'd' )) Table2 (FC) |
4.INTERSECT
(返回第一个查询结果集和第二个查询结果集共有的部分)
1
2
3
4
|
--返回结果为:[a,b,c] SELECT FC FROM ( VALUES ( 'a' ),( 'b' ),( 'c' ),( 'e' )) Table1 (FC) INTERSECT SELECT FC FROM ( VALUES ( 'a' ),( 'b' ),( 'c' ),( 'd' )) Table2 (FC) |
集合运算的使用场景
1.使用UNION
代替Where
子句中的OR
,查询速度更快
1
2
3
4
5
6
7
|
--使用Where子句 + OR SELECT name , population, area FROM world WHERE area > 3000000 OR population > 25000000 --使用UNION SELECT name , population, area FROM world WHERE area > 3000000 UNION SELECT name , population, area FROM world WHERE population > 25000000 |
2.使用EXCEPT
和INTERSECT
, 过滤出列表中不存在/存在于数据库中的项
假设存在表Customers
, 数据如下表所示
cust_id | cust_name | cust_address | cust_city | cust_state | cust_country | cust_contact | cust_email |
---|---|---|---|---|---|---|---|
1000000001 | Village Toys | 200 Maple Lane | Detroit | MI | USA | John Smith | sales@villagetoys.com |
1000000002 | Kids Place | 333 South Lake Drive | Columbus | OH | USA | Michelle Green | NULL |
1000000003 | Fun4All | 1 Sunny Place | Muncie | IN | USA | Jim Jones | jjones@fun4all.com |
1000000004 | Fun4All | 829 Riverside Drive | Phoenix | AZ | USA | Denise L. Stephens | dstephens@fun4all.com |
1000000005 | The Toy Store | 4545 53rd Street | Chicago | IL | USA | Kim Howard | NULL |
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
--过滤出列表中不存在于数据库中的项 --返回结果为['1000000006','1000000007'] SELECT [Id] AS [cust_id] FROM ( VALUES ( '1000000004' ),( '1000000005' ),( '1000000006' ),( '1000000007' ) ) dt ([Id]) EXCEPT SELECT [cust_id] FROM [Customers] --过滤出列表中存在于数据库中的项 --返回结果为['1000000004','1000000005'] SELECT [Id] AS [cust_id] FROM ( VALUES ( '1000000004' ),( '1000000005' ),( '1000000006' ),( '1000000007' ) ) dt ([Id]) INTERSECT SELECT [cust_id] FROM [Customers] |
1
2
3
4
5
6
7
8
9
10
11
|
--对于SQLServer 2008以前的版本 SELECT [Id] AS [cust_id] FROM ( SELECT '1000000004' UNION ALL SELECT '1000000005' UNION ALL SELECT '1000000006' UNION ALL SELECT '1000000007' ) dt ([Id]) INTERSECT --EXCEPT SELECT [cust_id] FROM [Customers] |
1
2
3
4
5
6
7
8
9
10
11
|
//使用C#动态生成SQL语句 var list = new List<string>(){ "1000000004" , "1000000005" , "1000000006" , "1000000007" }; string sqlQuery = string.Format($@ " SELECT [Id] AS [cust_id] FROM ( VALUES('{string.Join(" '),( '", list)}' ) ) dt ([Id] INTERSECT --EXCEPT SELECT [cust_id] FROM [Customers]" ); |
更多参考
Set Operators - EXCEPT and INTERSECT
到此这篇关于SQL Server中的集合运算: UNION, EXCEPT和INTERSECT的文章就介绍到这了,更多相关SQL Server中的集合运算内容请搜索服务器之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持服务器之家!
原文链接:https://www.cnblogs.com/makesense/p/13456100.html