DML(Data Manipulation Language)数据操作语言,是指对数据库进行增删改的操作指令,主要有INSERT、UPDATE、DELETE三种,代表插入、更新与删除,这是学习MySQL必要掌握的基本知识。
与之前的章节一致,下方语法中 [] 中内容可以省略。
语法格式如下:
insert into t_name[(column_name1,columnname_2,...)] values (val1,val2);
或者
insert into t_name set column_name1 = val1,column_name2 = val2;
1、字段名称和值需要保证数量一直,类型一直,位置一 一对应,否则可能导致异常。
2、not null的字段需要保证有插入的值,否则会报非空的异常信息。允许null的字段如果不想输入数据,字段和值都不出现,或者value用null代替。
3、数值类型,值不需要用单引号括起来,其他的如字符型或日期类型,值需要用单引号括起来;
4、如果表名后面的column_name 省略不写,则代表覆盖该表的所有字段。值的顺序和表中字段顺序须保持一致。
5、上述第二种语法的写法更繁琐,现在比较少使用。
测试一下:
mysql> desc `user1`;
+---------+--------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+---------+--------------+------+-----+---------+----------------+
| id | bigint(20) | NO | PRI | NULL | auto_increment |
| name | varchar(20) | NO | | NULL | |
| age | int(11) | NO | | 0 | |
| address | varchar(255) | YES | | NULL | |
+---------+--------------+------+-----+---------+----------------+
4 rows in set
mysql> insert into `user1`(name,age,address) values('brand',20,'fuzhou');
Query OK, 1 row affected
mysql> insert into `user1`(age,address) values(20,'fuzhou');
1364 - Field 'name' doesn't have a default value
mysql> insert into `user1` values('sol',21,'xiamen');
1136 - Column count doesn't match value count at row 1
mysql> insert into `user1` values(null,'sol',21,'xiamen');
Query OK, 1 row affected
mysql> select * from `user1`;
+----+-------+-----+---------+
| id | name | age | address |
+----+-------+-----+---------+
| 3 | brand | 20 | fuzhou |
| 4 | sol | 21 | xiamen |
+----+-------+-----+---------+
2 rows in set
语法格式如下:
insert into t_name [(column_name1,column_name2)] values (val1_1,val1_2),(val2_1,val2_2)...);
或者
insert into t_name [(column_name1,column_name2)] select o_name1,o_name2 from o_t_name [where condition];
1、上述第一个语法,values 后面的值个数需要同等配对 column的数量,可以设置多个,逗号隔开,提高数据插入效率。
2、第二个语法,select查询的字段和插入数据的字段数量、顺序、类型需要一致。 insert的字段可以省略,代表插入t_name表所有字段。条件可选。
测试一下:
mysql> insert into `user1`(name,age,address) values('brand',20,'fuzhou'),('sol',21,'xiamen');
Query OK, 2 rows affected
Records: 2 Duplicates: 0 Warnings: 0
mysql> select * from `user1`;
+----+-------+-----+---------+
| id | name | age | address |
+----+-------+-----+---------+
| 5 | brand | 20 | fuzhou |
| 6 | sol | 21 | xiamen |
+----+-------+-----+---------+
2 rows in set
mysql> desc `user2`;
+---------+--------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+---------+--------------+------+-----+---------+----------------+
| id | bigint(20) | NO | PRI | NULL | auto_increment |
| name | varchar(20) | NO | | NULL | |
| age | int(11) | NO | | 0 | |
| address | varchar(255) | YES | | NULL | |
| sex | int(11) | NO | | 1 | |
+---------+--------------+------+-----+---------+----------------+
5 rows in set
mysql> insert into `user2` (name,age,address,sex) select name,age,address,null from `user1`;
Query OK, 2 rows affected
Records: 2 Duplicates: 0 Warnings: 0
mysql> select * from `user2`;
+----+-------+-----+---------+------+
| id | name | age | address | sex |
+----+-------+-----+---------+------+
| 7 | brand | 20 | fuzhou | 1 |
| 8 | sol | 21 | xiamen | 1 |
+----+-------+-----+---------+------+
2 rows in set
语法格式如下:
update t_name [[as] alias] set [ alias.]column_name1 = val1,[alias.]column_name2 = val2 [where condition];
1、alias 是别名的意思,别名越简单识别性越强越好,容易辨认,方便操作,没有别名情况下,表名就是别名
2、as alias 中as也是可选的,where 条件也是可选的,所以用户可以选择需要的,符合特定条件的部分数据进行更新。
测试一下:
mysql> select * from `user2`;
+----+-------+-----+---------+------+
| id | name | age | address | sex |
+----+-------+-----+---------+------+
| 7 | brand | 20 | fuzhou | NULL |
| 8 | sol | 21 | xiamen | NULL |
+----+-------+-----+---------+------+
2 rows in set
mysql> update `user2` as u2 set u2.name = 'hero',u2.age=23,u2.sex=1 where id=7;
Query OK, 1 row affected
Rows matched: 1 Changed: 1 Warnings: 0
mysql> select * from `user2`;
+----+------+-----+---------+------+
| id | name | age | address | sex |
+----+------+-----+---------+------+
| 7 | hero | 23 | fuzhou | 1 |
| 8 | sol | 21 | xiamen | NULL |
+----+------+-----+---------+------+
2 rows in set
还有一种方式是同时更新多个表,使用不同的别名以及一些条件去限制,不过不建议这么做,操作易错,并且不好维护。
语法格式如下:
delete [alias] from t_name [[as] alias] [where condition];
1、跟上面一样,alias代表别名,没有别名情况下,表名就是别名
2、如果表设置了别名,则delete后面必须跟上别名,否则数据库会报异常。
测试一下:
mysql> select * from `user2`;
+----+------+-----+---------+------+
| id | name | age | address | sex |
+----+------+-----+---------+------+
| 7 | hero | 23 | fuzhou | 1 |
| 8 | sol | 21 | xiamen | NULL |
+----+------+-----+---------+------+
2 rows in set
mysql> delete from `user2` as alias where sex=1;
1064 - You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'as alias where sex=1' at line 1
mysql> delete alias from `user2` as alias where sex=1;
Query OK, 1 row affected
mysql> select * from `user2`;
+----+------+-----+---------+------+
| id | name | age | address | sex |
+----+------+-----+---------+------+
| 8 | sol | 21 | xiamen | NULL |
+----+------+-----+---------+------+
1 row in set
3、如果删除表中所有的数据,则后面不带上where条件即可,不过要谨慎使用哟。
mysql> select * from `user2`;
+----+-------+-----+----------+-----+
| id | name | age | address | sex |
+----+-------+-----+----------+-----+
| 8 | sol | 21 | xiamen | 0 |
| 10 | brand | 21 | fuzhou | 1 |
| 11 | helen | 20 | quanzhou | 0 |
+----+-------+-----+----------+-----+
3 rows in set
mysql> delete from `user2`;
Query OK, 3 rows affected
mysql> select * from `user2`;
Empty set
语法格式如下:
truncate t_name;
mysql> select * from `user2`;
+----+-------+-----+----------+-----+
| id | name | age | address | sex |
+----+-------+-----+----------+-----+
| 12 | brand | 21 | fuzhou | 1 |
| 13 | helen | 20 | quanzhou | 0 |
| 14 | sol | 21 | xiamen | 0 |
+----+-------+-----+----------+-----+
3 rows in set
mysql> truncate `user2`;
Query OK, 0 rows affected
mysql> select * from `user2`;
Empty set
看起来跟delete很像,但是重新插入数据会发现,他的自增主键会重新从1开始,但是delete的是直接在原来的所以自增值之后往上加。看下面id字段。
mysql> insert into `user2` (name,age,address,sex) values('brand',21,'fuzhou',1),('helen',20,'quanzhou',0),('sol',21,'xiamen',0);
Query OK, 3 rows affected
Records: 3 Duplicates: 0 Warnings: 0
mysql> select * from `user2`;
+----+-------+-----+----------+-----+
| id | name | age | address | sex |
+----+-------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
+----+-------+-----+----------+-----+
3 rows in set
那 truncate 和 delete有什么区别呢?我们来梳理下。
1、truncate 指的是清空表的数据、释放表的空间,但不删除表的架构定义(表结构)。因为不包含Where条件,所以不是删除具体行,而是将整个表清空了。
2、而delete 语句是删除表中的数据行,可以在后面带上条件控制删除的维度、范围,它每次从表中删除一行,会同时将该行的删除操作作为事务保存在日志中,用于进行可能的回滚操作。
3、truncate 和 delete 一样的地方是:只是删除数据,涉及到的表结构及其列、约束、索引等均不会变。
4、如果被外键 foreign key 约束,不能使用truncate ,只能使用不带where子句的delete语句。
5、truncate 操作会记录在日志中,delete操作会放到 rollback segement 中,执行时要等事务被commit才会生效;所以delete 会触发删除触发器(如果有的话),truncate 不会。
6、如果像上面我们测试的那样,包含自增字段,truncate方式清空之后,自增列的值会被初始化从1开始。
delete方式要分情况判断(如果数据全部delete,数据库未被重启,则按照之前max+1;数据库重启了,则一样会重新开始计算自增列的初始值)。
7、前面章节我们还学过drop,drop语句会删除表包括 结构、数据、依赖该表的约束(constrain),触发器(trigger)索引(index)等。