视图是表的一个映射,是一张虚表,在结构上视图和普通的表没什么区别,一样可以用sql语句来增删改查;视图创建后是一直存在数据库内
创建视图
CREATE [ALGORITHM]={UNDEFINED|MERGE|TEMPTABLE}] VIEW 视图名 [(字段清单)] AS SELECT 语句 [WITH [CASCADED|LOCAL] CHECK OPTION];
-- Merge,意味着视图只是一个规则,语句规则, 当查询视图时,把查询视图的语句(比如where那些)与创建时的语句where子句等合并,分析.再形成一条新的select语句. -- 创建视图的语句: CREATE altorethm=merge VIEW `g2` AS SELECT `goods_id`,`cat_id`,`goods_name`,`shop_price` FROM `goods` ORDER BY `cat_id` ASC,`shop_price` DESC; --查询视图的语句: SELECT * FROM `g2` GROUP BY `cat_id`; -- 最终执行的语句: SELECT `goods_id`,`cat_id`,`goods_name`,`shop_price` FROM `goods` GROUP BY `cat_id` ORDER BY `cat_id` ASC,`shop_price` DESC;
-- temptable是根据创建语句瞬间创建一张临时表,然后查询视图的语句从该临时表查数据. CREATE ALTORETHM=TEMPTABLE VIEW `g2` AS SELECT `goods_id`,`cat_id`,`goods_name`,`shop_price` FROM `goods` ORDER BY `cat_id` ASC,`shop_price` DESC; -- 查询视图的语句: SELECT * FROM `g2` GROUP BY `cat_id`; -- 最终执行的2句话: 取数据并放在临时表,然后去查临时表.
mysql> create view v1 as select * from goods where stock > 100 WITH check OPTION; -- 使用了with check option 这个选项那后面对视图的插入只限于包含stock的表; Query OK, 0 rows affected -- ↓↓ 只允许插入包含stock字段的sql mysql> INSERT into v1(gname) values('zhouzhou'); 1369 - CHECK OPTION failed 'test.v1' mysql> INSERT into v1(gname,stock) values('zhouzhou',8000); Query OK, 1 row affected -- ↓↓ 不影响删除 mysql> delete from v1 where gname = 'ganggang'; Query OK, 2 rows affected -- ↓↓ 貌似也不影响更新 mysql> update v1 set gname = 'test' where gid = 11; Query OK, 1 row affected Rows matched: 1 Changed: 1 Warnings: 0 -- 其实错误,因为更新后查询出来的表字段里面能查询到stock > 100;所以可以更新; -- ↓↓ 如果更新后的结构不能再通过stock>100来查询就不能进行更新; mysql> update v1 set stock = 99; 1369 - CHECK OPTION failed 'test.v1'
以后研究... wait for
CREATE VIEW v1 AS SELECT user_id, nickname, IF (sex = 1, '男', '女') as sex, user_money FROM users ORDER BY user_money DESC LIMIT 10; -- question: 该视图不能被更新? update v1 set user_money=5000 where user_id = 2659; [Err] 1288 - The target table v2 of the UPDATE is not updatable
CREATE VIEW v3 AS SELECT u.user_id, nickname, lhmoney, user_recommend, goods FROM users AS u JOIN COMMENT AS c ON u.user_id = c.user_id ORDER BY user_id ASC WITH CASCADED CHECK OPTION;
创建视图时注意:[Err] 1349 - View's SELECT contains a subquery in the FROM clause -- 不能在SELECT的FROM子句中包含子查询。
删除视图
mysql> drop view v1; -- drop view 视图名; Query OK, 0 rows affected
修改视图
CREATE OR REPLACE语句非常灵活,在视图存在的情况下可对视图进行修改,视图不在的情况下可创建视图,其基本用法和CREATE VIEW 几乎一致
CREATE OR REPLACE ALGORITHM=TEMPTABLE VIEW v2(uid,nickname,money) AS SELECT user_id, nickname, user_money FROM users ORDER BY user_money DESC LIMIT 10;
AlTER VIEW v3(uid,cid,nickname,lhb,content,goods) AS SELECT u.user_id, user_comment_id, nickname, lhmoney, user_recommend, goods FROM users AS u JOIN COMMENT AS c ON u.user_id = c.user_id ORDER BY user_id ASC WITH CASCADED CHECK OPTION;
关于更新视图数据
视图是表的查询结果映射,那原表的数据改变了,视图也当然会跟着做相应的改变。
但,如果视图里面的数据改变了呢?
首先,视图增删改是会影响表的;也就是说,可以在诸如UPDATE、DELETE或INSERT等语句中使用它们,以更新基表的内容。但是,视图并不是总是能增删改的。对于可更新的视图,在视图中的行和基表中的行之间必须具有一对一的关系。还有一些特定的其他结构,这类结构会使得视图 不可更新 。更具体地讲,如果视图包含下述结构中的任何一种,那么它就是不可更新的:
mysql> update v4 set sex=5 where user_id = 32; -- 一定程度上保证了基表数据的安全性 1288 - The target table v4 of the UPDATE is not updatable
关于视图的可插入性:insert
如果视图满足关于视图列的下述额外要求,可更新的视图也是可插入的:
查看视图
查看视图字段结构
mysql> describe v3; +----------------+-------------+------+-----+---------+-------+ | Field | Type | Null | Key | Default | Extra | +----------------+-------------+------+-----+---------+-------+ | user_id | int(11) | NO | | 0 | | | nickname | varchar(25) | NO | | | | | lhmoney | int(11) | NO | | 0 | | | user_recommend | longtext | YES | | NULL | | | goods | int(10) | NO | | 0 | | +----------------+-------------+------+-----+---------+-------+
查看视图基本信息
mysql> show table status like 'v3'/G -- 该命令与查看普通表的命令是一样的,而此处大多数为NULL正是证明视图是一张虚表; *************************** 1. row *************************** Name: v3 Engine: NULL Version: NULL Row_format: NULL Rows: NULL Avg_row_length: NULL Data_length: NULL Max_data_length: NULL Index_length: NULL Data_free: NULL Auto_increment: NULL Create_time: NULL Update_time: NULL Check_time: NULL Collation: NULL Checksum: NULL Create_options: NULL Comment: VIEW 1 row in set (0.01 sec) mysql> show table status like 'users'/G *************************** 1. row *************************** Name: users Engine: InnoDB Version: 10 Row_format: Compact Rows: 171650 Avg_row_length: 82 Data_length: 14172160 Max_data_length: 0 Index_length: 11567104 Data_free: 66060288 Auto_increment: 169781 Create_time: 2015-05-01 23:03:29 Update_time: NULL Check_time: NULL Collation: utf8_general_ci Checksum: NULL Create_options: Comment: 用户信息表 1 row in set (0.01 sec)
mysql> show create view v3/G *************************** 1. row *************************** View: v3 Create View: CREATE ALGORITHM=UNDEFINED DEFINER=`root`@`localhost` SQL SECURITY DEFINER VIE W `v3` AS select `u`.`user_id` AS `user_id`,`u`.`nickname` AS `nickname`,`u`.`lhmoney` AS `lhmoney`, `c`.`user_recommend` AS `user_recommend`,`c`.`goods` AS `goods` from (`users` `u` join `comment` `c` on((`u`.`user_id` = `c`.`user_id`))) order by `u`.`user_id` WITH CASCADED CHECK OPTION character_set_client: utf8 collation_connection: utf8_general_ci 1 row in set (0.01 sec)
mysql> SELECT * FROM information_schema.views/G *************************** 1. row *************************** TABLE_CATALOG: def TABLE_SCHEMA: test TABLE_NAME: v1 VIEW_DEFINITION: select `test`.`users`.`user_id` AS `user_id`,`test`.`users`.`nickname` AS `nic kname`,if((`test`.`users`.`sex` = 1),'男','女') AS `sex`,`test`.`users`.`user_money` AS `user_money` from `test`.`users` order by `test`.`users`.`user_money` desc limit 10 CHECK_OPTION: NONE IS_UPDATABLE: NO DEFINER: root@localhost SECURITY_TYPE: DEFINER CHARACTER_SET_CLIENT: utf8 COLLATION_CONNECTION: utf8_general_ci
information_schema是Mysql自带的一个数据库,详细请参考我的另外一边博文《探寻mysql自带的数据库》
思路1:
News表,分成四表,每表给一个id
Newsid, 1,2,3,4
News1,news2,news3,news4表
把一张表的数据分散到4张表里,分散的方法很多,
最常用可以用id取模来计算.
Id%4+1 = [1,2,3,4]
比如
```sql
$_GET['id'] = 17,
17%4 + 1 = 2,
$tableName = 'news'.'2'
Select * from news2 where id = 17;```
思路2:
还可以用视图, 把4张表形成一张视图
Create view news as select from n1 union select from n2 union.........