码迷,mamicode.com
首页 > 数据库 > 详细

CentOS7安装使用MySQL

时间:2017-04-07 01:00:25      阅读:201      评论:0      收藏:0      [点我收藏+]

标签:错误   占位符   指定   发布   完整   sql语句   触发器   set   main   

From: http://www.cnblogs.com/edward2013/p/5313724.html

安装MySQL

添加mysql源

1
# rpm -Uvh http://repo.mysql.com//mysql57-community-release-el7-7.noarch.rpm

安装mysql

1
# yum -y install mysql-community-server

启动mysql并设置为开机自启动服务

1
2
# chkconfig mysqld on
# service mysqld start

检查mysql服务状态

1
# service mysqld status

第一次启动mysql,会在日志文件中生成root用户的一个随机密码,使用下面命令查看该密码

1
# grep ‘temporary password‘ /var/log/mysqld.log

修改root用户密码

1
2
3
4
# mysql -u root -p -h localhost
Enter password:
 
mysql> ALTER USER ‘root‘@‘localhost‘ IDENTIFIED BY ‘Mysql-2016‘;

 创建数据库(数据库名为:crashcourse)

1
mysql> create database crashcourse;

使用crashcourse数据库

1
mysql> use crashcourse;

执行sql脚本(使用source命令)

1
2
mysql> source /root/MySQLCrashCourse/create.sql;
mysql> source /root/MySQLCrashCourse/populate.sql;

查看可用数据库的列表

1
mysql> show databases;

 查看当前数据库内可用表的列表

1
mysql> show tables;

显示表列(表名:customers)

1
mysql> show columns from customers;

 显示服务器错误或警告消息

1
2
mysql> show errors;
mysql> show warnings;

检索数据

检索单个列

1
mysql> select prod_name from products;

检索多个列

1
mysql> select prod_id, prod_name, prod_price from products;

检索所有列(除非你确实需要表中的每个列,否则最好别使用*通配符,因为会降低应用程序的性能)

1
mysql> select * from products;

检索不同的行(使用distinct关键字)

1
mysql> select distinct vend_id from products;

限制结果

1
2
3
4
返回不多于5行
mysql> select prod_name from products limit 5;
检索的开始行和行数
mysql> select prod_name from products limit 5,5;

使用完全限定的表名

1
mysql> select products.prod_name from crashcourse.products;

排序检索数据

排序数据

1
mysql> select prod_name from products order by prod_name;

按多个列排序

1
mysql> select prod_id, prod_price, prod_name from products order by prod_price, prod_name;

指定排序方向(使用desc关键字,desc为倒序,默认是asc,正序)

1
mysql> select prod_id, prod_price, prod_name from products order by prod_price desc;

使用order by和limit的组合

1
mysql> select prod_price from products order by prod_price desc limit 1;

过滤数据

检索单个值

1
mysql> select prod_name, prod_price from products where prod_name = ‘fuses‘;

不匹配检查

1
mysql> select vend_id, prod_name from products where vend_id != 1003;

范围值检查

1
mysql> select prod_name, prod_price from products where prod_price between 5 and 10;

空值检查

1
mysql> select prod_name from products where prod_price is null;

数据过滤

and操作符

1
mysql> select prod_id, prod_price, prod_name from products where vend_id = 1003 and prod_price <= 10;

or操作符

1
mysql> select prod_name, prod_price from products where vend_id = 1003 or prod_id = 1003;

计算次序

1
mysql> select prod_name, prod_price from products where (vend_id = 1002 or vend_id = 1003) and prod_price >= 10;

in操作符

1
mysql> select prod_name, prod_price from products where vend_id in (1002,1003) order by prod_name;

not操作符

1
mysql> select prod_name, prod_price from products where vend_id not in (1002,1003) order by prod_name;

用通配符进行过滤

百分号(%)通配符(匹配0,1或多个字符)

1
mysql> select prod_id, prod_name from products where prod_name like ‘jet%‘;

下划线(_)通配符(匹配1个字符)

1
mysql> select prod_id, prod_name from products where prod_name like ‘_ ton anvil‘;

用正则表达式进行搜索

基本字符匹配

1
mysql> select prod_name from products where prod_name regexp ‘.000‘ order by prod_name;

进行or匹配

1
mysql> select prod_name from products where prod_name regexp ‘1000|2000‘ order by prod_name;

匹配范围

1
mysql> select prod_name from products where prod_name regexp ‘[1-5] Ton‘ order by prod_name;

匹配特殊字符(使用\\进行转义)

1
mysql> select vend_name from vendors where vend_name regexp ‘\\.‘ order by vend_name;

匹配字符类

1
mysql> select prod_name from products where prod_name regexp ‘[[:digit:]]{4}‘ order by prod_name;

匹配多个实例

1
mysql> select prod_name from products where prod_name regexp ‘\\([0-9] sticks?\\)‘ order by prod_name;

定位符(^, $, [[:<:]], [[:>:]])

1
mysql> select prod_name from products where prod_name regexp ‘^[0-9\\.]‘ order by prod_name;

创建计算字段

拼接字段

1
mysql> select concat(vend_name, ‘ (‘, vend_country, ‘)‘) from vendors order by vend_name;

使用别名

1
mysql> select concat(rtrim(vend_name), ‘ (‘, rtrim(vend_country), ‘)‘) as vend_title from vendors order by vend_name;

执行算术计算

1
mysql> select prod_id, quantity, item_price, quantity*item_price as expanded_price from orderitems where order_num = 20005;

使用数据处理函数

文本处理函数

1
2
mysql> select vend_name, upper(vend_name) as vend_name_upcase from vendors order by vend_name;
mysql> select cust_name, cust_contact from customers where soundex(cust_contact) = soundex(‘Y Lie‘);

日期和时间处理函数

1
2
mysql> select cust_id, order_num from orders where date(order_date) between ‘2005-09-01‘ and ‘2005-09-30‘;
mysql> select cust_id, order_num from orders where year(order_date) = 2005 and month(order_date) = 9;

汇总数据

avg()函数

1
mysql> select avg(prod_price) as avg_price from products;

count()函数

1
mysql> select count(*) as num_cust from customers;

max()函数

1
mysql> select max(prod_price) as max_price from products;

min()函数

1
mysql> select min(prod_price) as min_price from products;

sum()函数

1
2
mysql> select sum(quantity) as items_ordered from orderitems where order_num = 20005;
mysql> select sum(item_price*quantity) as total_price from orderitems where order_num = 20005;

聚合不同的值

1
mysql> select avg(distinct prod_price) as avg_price from products where vend_id = 1003;

组合聚集函数

1
mysql> select count(*) as num_items, min(prod_price) as price_min, max(prod_price) as price_max, avg(prod_price) as price_avg from products;

 分组数据

创建分组

1
mysql> select vend_id, count(*) as num_prods from products group by vend_id;

过滤分组(使用having关键字)

1
2
mysql> select cust_id, count(*) as orders from orders group by cust_id having count(*) >= 2;
mysql> select vend_id, count(*) as num_prods from products where prod_price >= 10 group by vend_id having count(*) >= 2;

分组和排序

1
2
3
4
5
mysql> select order_num, sum(quantity*item_price) as ordertotal
    from orderitems
    group by order_num
    having sum(quantity*item_price) >= 50
    order by ordertotal;

使用子查询

利用子查询进行过滤

1
2
3
4
5
6
7
mysql> select cust_name, cust_contact
    from customers
    where cust_id in (select cust_id
     from orders
     where order_num in (select order_num
     from orderitems
     where prod_id = ‘TNT2‘));

作为计算字段使用子查询

1
2
3
4
5
6
mysql> select cust_name, cust_state,
                (select count(*)
                from orders
                where orders.cust_id = customers.cust_id) as orders
            from customers
            order by cust_name;

联结表

创建联结

1
2
3
4
mysql> select vend_name, prod_name, prod_price
            from vendors, products
            where vendors.vend_id = products.vend_id
            order by vend_name, prod_name;

内部联结

1
2
3
mysql> select vend_name, prod_name, prod_price
            from vendors inner join products
            on vendors.vend_id = products.vend_id;

联结多个表

1
2
3
4
5
mysql> select cust_name, cust_contact
            from customers, orders, orderitems
            where customers.cust_id = orders.cust_id
            and orderitems.order_num = orders.order_num
            and prod_id = ‘TNT2‘;

创建高级联结

使用别名

1
2
3
4
5
mysql> select cust_name, cust_contact
            from customers as c, orders as o, orderitems as oi
            where c.cust_id = o.cust_id
            and oi.order_num = o.order_num
            and prod_id = ‘TNT2‘;

自然联结

1
2
3
4
5
mysql> select c.*, o.order_num, o.order_date, oi.prod_id, oi.quantity, oi.item_price
            from customer as c, orders as o, orderitems as oi
            where c.cust_id = o.cust_id
            and oi.order_num = o.order_num
            and prod_id = ‘FB‘;

外部联结

1
2
3
mysql> select customers.cust_id, orders.order_num
            from customers left join orders
            on customers.cust_id = orders.cust_id;

使用带聚集函数的联结

1
2
3
4
5
mysql> select customers.cust_name, customers.cust_id,
                count(orders.order_num) as num_ord
            from customers inner join orders
            on customers.cust_id = orders.cust_id
            group by customers.cust_id;

组合查询

使用union

1
2
3
4
5
6
7
mysql> select vend_id, prod_id, prod_price
            from products
            where prod_price <= 5
            union
            select vend_id, prod_id, prod_price
            from products
            where vend_id in (1001,1002);

包含重复的行(使用union all命令)

1
2
3
4
5
6
7
mysql> select vend_id, prod_id, prod_price
            from products
            where prod_peice <= 5
            union all
            select vend_id, prod_id, prod_price
            from products
            where vend_id in (1001,1002);

对组合查询结果排序

1
2
3
4
5
6
7
8
mysql> select vend_id, prod_id, prod_price
            from products
            where prod_price <=5
            union
            select vend_id, prod_id, prod_price
            from products
            where vend_id in (1001,1002)
            order by vend_id, prod_price;

全文搜索

仅在MyISAM数据库引擎中支持全文搜索,一般在创建表时启用全文搜索

1
2
3
4
5
6
7
8
9
CREATE TABLE productnotes
(
    note_id      int          NOT NULL  AUTO_INCREMENT,
    prod_id      char(10)       NOT NULL,
    note_date  datetime      NOT NULL,
    note_text   text             NULL,
    PRIMARY KEY(note_id),
    FULLTEXT(note_text)
) ENGINE=MyISAM;

进行全文本搜索

1
2
3
mysql> select note_text
            from productnotes
            where match(note_text) against(‘rabbit‘);

使用查询扩展

1
2
3
mysql> select note_text
            from productnotes
            where match(note_text) against(‘anvils‘ with query expansion);

布尔文本搜索

1
2
3
mysql> select note_text
            from productnotes
            where match(note_text) against(‘heavy‘ in boolean mode);

为匹配包含heavy但不包含任意以rope开始的词的行

1
2
3
mysql> select note_text
            from productnotes
            where match(note_text) against(‘heavy -rope*‘ in boolean mode);

插入数据

插入完整的行

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
mysql> insert into customers(cust_name,
    cust_contact,
    cust_email,
    cust_address,
    cust_city,
    cust_state,
    cust_zip,
    cust_country)
values(‘Pep E. LaPew‘,
    Null,
    Null,
    ‘100 Main Street‘,
    ‘Los Angeles‘,
    ‘CA‘,
    ‘90046‘,
    ‘USA‘);

插入多个行

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
mysql> insert into customers(cust_name,
    cust_address,
    cust_city,
    cust_state,
    cust_zip,
    cust_country)
values(
    ‘Pep E. LaPew‘,
    ‘100 Main Street‘,
    ‘Logs Angeles‘,
    ‘CA‘,
    ‘90046‘,
    ‘USA‘
),
(
    ‘M. Martian‘,
    ‘42 Galaxy Way‘,
    ‘New York‘,
    ‘NY‘,
    ‘11213‘,
    ‘USA‘
);

插入检索出的数据

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
mysql> insert into customers(cust_id,
    cust_contact,
    cust_email,
    cust_name,
    cust_address,
    cust_city,
    cust_state,
    cust_zip,
    cust_country)
select cust_id,
    cust_contact,
    cust_email,
    cust_name,
    cust_address,
    cust_city,
    cust_state,
    cust_zip,
    cust_country
from custnew;

更新和删除数据

更新数据

1
2
3
4
mysql> update customers
    set cust_name = ‘The Fudds‘,
        cust_email = ‘elmer@fudd.com‘
where cust_id = 10005;

删除数据

1
2
mysql> delete from customers
    where cust_id = 10006;

创建和操纵表

创建表

1
2
3
4
5
6
7
8
9
10
11
12
mysql> create table customers(
    cust_id            int        NOT NULL AUTO_INCREMENT,
    cust_name     char(50)      NOT NULL,
    cust_address   char(50)     NULL,
    cust_city         char(50)     NULL,
    cust_state       char(5)       NULL,
    char_zip          char(10)     NULL,
    char_country   char(50)    NULL,
    char_contact    char(50)    NULL,
    char_email      char(255)   NULL,
    PRIMARY KEY (cust_id)
) ENGINE=InnoDB;

指定默认值

1
2
3
4
5
6
7
8
mysql> create table orderitems(
    order_num    int                NOT NULL,
    order_item    int                NOT NULL,
    prod_id         char(10)        NOT NULL,
    quantity        int                 NOT NULL   DEFAULT  1,
    item_price    decimal(8,2)   NOT NULL,
    PRIMARY KEY (order_num, order_item)
) ENGINE=InnoDB;

引擎类型

1
2
3
InnoDB是一个可靠的事务处理引擎。
MEMORY在功能上等同于MyISAM,但是数据存储在内存中,速度很快。
MyISAM是性能极高的引擎,支持全文搜索,但不支持事务处理。

更新表

给表添加一列:

1
2
mysql> alter table vendors
    add vend_phone char(20);

删除刚刚添加的列:

1
2
mysql> alter table vendors
    drop column vend_phone;

ALTER TABLE的一种常见用途是定义外键:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
mysql> alter table orderitems
    add constraint fk_orderitems_orders
    foreign key (order_num) references orders (order_num);
 
mysql> alter table orderitems
    add constraint fk_orderitems_products
    foreign key (prod_id) references products (prod_id);
 
mysql> alter table orders
    add constraint fk_orders_customers
    foreign key (cust_id) references customers (cust_id);
 
mysql> alter table products
    add constraint fk_products_vendors
    foreign key (vend_id) references vendors (vend_id);

删除表

1
mysql> drop table customers2;

重命名表

1
mysql> rename table customers2 to customers;

对多个表重命名

1
2
3
mysql> rename table backup_customers to customers,
        backup_vendors to vendors,
        backup_products to products;

使用视图

利用视图简化复杂的联结

1
2
3
4
5
mysql> create view productcustomers as
    select cust_name, cust_contact, prod_id
    from customers, orders, orderitems
    where customers.cust_id = orders.cust_id
        and orderitems.order_num = orders.order_num;

用视图重新格式化检索出的数据

1
2
3
4
5
mysql> create view vendorlocations as
  select concat(rtrim(vend_name), ‘ (‘, rtrim(vend_country), ‘)‘)
        as vend_title
    from vendors
    order by vend_name;

用视图过滤不想要的数据

1
2
3
4
mysql> create view customeremaillist as
    select cust_id, cust_name, cust_email
    from customers
    where cust_email is not null;

使用视图与计算字段

1
2
3
4
5
6
7
mysql> create view orderitemsexpanded as
    select order_num,
            prod_id,
            quantity,
            item_price,
            quantity*item_price as expanded_price
    from orderitems;

使用存储过程

执行存储过程(所有mysql变量都必须以@开始)

1
2
3
mysql> call productpricing(@pricelow,
                           @pricehigh,
                           @priceaverage);

创建存储过程

1
2
3
4
5
mysql> create procedure productpricing()
    begin
        select avg(prod_price) as priceaverage
        from products;
    end;

删除存储过程

1
mysql> drop procedure productpricing;

使用参数

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
mysql> create procedure productpricing(
    out pl decimal(8,2),
    out ph decimal(8,2),
    out pa decimal(8,2)
)
begin
    select min(prod_price)
    into pl
    from products;
    select max(prod_price)
    into ph
    from products;
    select avg(prod_price)
    into pa
    from products;
end;

建立智能存储过程

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
-- Name: ordertotal
-- Parameters: onumber = order number
--                     taxable = 0 if not taxable, 1 if taxable
--                     ototal = order total variable
 
create procedure ordertotal(
    in onumber int,
    in taxable boolean,
    out ototal decimal(8,2)
) comment ‘Obtain order total, optionally adding tax‘
begin
 
    -- Declare variable for total
    declare total decimal(8,2);
    -- Declare tax percentage
    declare taxrate int default 6;
 
    -- Get the order total
    select sum(item_price*quantity)
    from orderitems
    where order_num = onumber
    into total;
 
    -- Is this taxable?
    if taxable then
        -- Yes, so add taxrate to the total
        select total+(total/100*taxrate) into total;
    end if;
 
    -- And finally, save to out variable
    select total into ototal;
 
end;

taxable=0测试结果

1
2
call ordertotal(20005, 0, @total);
select @total;

taxable=1测试结果

1
2
call ordertotal(20005, 1, @total);
select @total;

检查存储过程

1
mysql> show create procedure ordertotal;

限制过程状态结果

1
mysql> show procedure status like ‘ordertotal‘;

使用游标

创建游标

1
2
3
4
5
6
create procedure processorders()
begin
    declare ordernumbers cursor
    for
    select order_num from orders;
end;

打开和关闭游标

1
2
3
4
-- 打开游标
open ordernumbers;
-- 关闭游标
close ordernumbers;

使用游标

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
create procedure processorders()
begin
 
    -- declare local variables
    declare done boolean default 0;
    declare o int;
    declare t decimal(8,2);
 
    -- declare the cursor
    declare ordernumbers cursor
    for
    select order_num from orders;
    -- declare continue handler
    declare continue handler for sqlstate ‘02000‘ set done=1;
 
    -- create a table to store the results
    create table if not exists ordertotals
        (order_num int, total decimal(8,2));
 
    -- open the cursor
    open ordernumbers;
 
    -- loop through all rows
    repeat
 
        -- get order number
        fetch ordernumbers into o;
 
        -- get the total for this order
        call ordertotal(o, 1, t);
 
        -- insert order and total into ordertotals
        insert into ordertotals(order_num, total)
        values(o, t);
 
    -- end of loop
    until done end repeat;
 
    -- close the cursor
    close ordernumbers;
 
end;

使用触发器

创建触发器

1
2
create trigger newproduct after insert on products
for each row select ‘Product added‘;

删除触发器

1
drop trigger newproduct;

使用触发器

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
-- insert触发器
create trigger neworder after insert on orders
for each row select new.order_num;
 
-- delete触发器
create tigger deleteorder before delete on orders
for each row
begin
  insert into archive_orders(order_num, order_date, cust_id)
  value(old.order_num, old.order_date, old.cust_id);
end;
 
-- update触发器
create trigger updatevendor before update on vendors
for each row set new.vend_state = upper(new.vend_state);

管理实务处理

事务(transaction)指一组SQL语句;

回退(rollback)指撤销指定的SQL语句的过程;

提交(commit)指将未存储的SQL语句结果写入数据库表;

保留点(savepoint)指事务处理中设置的临时占位符,你可以对它发布回退。

使用rollback

1
2
3
4
5
6
select * from ordertotals;
start transaction;
delete from ordertotals;
select * from ordertotals;
rollback;
select * from ordertotals;

使用commit

1
2
3
4
start transaction;
delete from orderitems where order_num = 20010;
delete from orders where order_item = 20010;
commit;

使用保留点

1
2
3
savepoint delete1;
 
rollback to delete1;

更改默认的提交行为(设置autocommit为0,mysql不自动提交更改,直到autocommit为设置为1为止)

1
set autocommit=0;

全球化和本地化

字符集为字母和符号的集合;

编码为某个字符集成员的内部表示;

校对为规定字符如何比较的指令。

查看所支持的字符集完整列表

1
show character set;

查看所支持校对的完整列表

1
show collation;

显示数据库所用的字符集和校对

1
2
show variables like ‘character%‘;
show variables like ‘collation%‘;

给表指定字符集和校对

1
2
3
4
5
6
create table mytable
(
  column1  int,
  column2  varchar(10)
) default character set hebrew
  collate hebrew_general_ci;

对每个列设置字符集和校对

1
2
3
4
5
6
7
create table mytable
(
  column1  int,
  column2  varchar(10),
  columm3   varchar(10) character set latin1 collate latin1_general_ci
) default character set hebrew
  collate hebrew_general_ci;

校对顺序排序

1
2
select * from customers
order by lastname, firstname collate latin1_general_cs;

安全管理

不应该在日常的MySQL操作中使用root

获得所有账号列表

1
2
mysql> use mysql;
mysql> select user from user;

创建用户账号

1
mysql> create user ben identified by ‘p@$$w0rd‘;

重命名用户账号

1
mysql> rename user ben to bforta;

删除用户账号

1
mysql> drop user bforta;

查看用户账号权限

1
mysql> show grants for bforta;

给用户账号授予权限

1
mysql> grant select on crashcourse.* to bforta;

撤销用户权限

1
mysql> revoke select on crashcourse.* from bforta;

更改用户口令

1
set password for bforta = password(‘n3w p@$$w0rd‘);

设置自己的口令

1
set password = password(‘n3w p@$$w0rd‘);

数据库维护

备份数据

  • 使用命令行实用程序mysqldump转储所有数据库内容到某个外部文件
  • 使用命令行实用程序mysqlhotcoy从一个数据库复制所有数据
  • 使用backup table或select into outfile转储所有数据到某个外部文件

数据库维护

  • analyze table用来检查表键是否正确
1
mysql> analyze table orders;
  • check table用来针对许多问题对表进行检查
1
mysql> check table orders, orderitems;
  • 如果从一个表中删除大量数据,应该使用optimize table来回收所用的空间,从而优化表的性能

查看日志文件

  • 错误日志。它包含启动和关闭问题以及任意关键错误的细节。通常名为hostname.err
  • 查询日志。它记录所有mysql活动,在诊断问题时非常有用。通常名为hostname.log
  • 二进制日志。它记录更新过数据的所有语句。通常名为hostname.bin
  • 缓慢查询日志。它记录执行缓慢的任何查询。通常名为hostname-slow.log

改善性能

查看当前设置

1
2
mysql> show variables;
mysql> show status;

显示所有活动进程

1
mysql> show processlist;
  • 如果某个任务执行缓慢,则所有的请求都会执行缓慢,可以用kill命令终结该进程。
  • 总是有不止一种方法编写同一条select语句,应该试验联结、并、子查询等,找出最佳的方法。
  • 使用explain语句让mysql解释它如何执行一条select语句。
  • 一般来说,存储过程执行比一条条地执行其中的各条mysql语句快。
  • 应该总是使用正确的数据类型。
  • 决不要检索比需求还要多的数据。换言之,不要用select *(除非你真的需要每个列)。
  • 有的操作(包括insert)支持一个可选的delayed关键字,如果使用它,将把控制立即返回给调用程序,并且一旦有可能就实际执行该操作。
  • 在导入数据时,应该关闭自动提交。你可能还想删除索引(包括fulltext索引),然后在导入完成后再重建它们。
  • 必须索引数据库以改善数据检索的性能。确定索引什么不是一件微不足道的任务,需要分析使用的select语句以找出重复的where和order by子句。如果一个简单的where子句返回结果所花的时间太长,则可以断定其中使用的列(或几个列)就是需要索引的对象。
  • 你的select语句中有一系列复杂的or条件吗?通过使用多条select语句和连接它们的union语句,你能看到极大的性能改进。
  • 索引改善数据检索的性能,但损害数据插入、删除和更新的性能。如果你有一些表,它们收集数据且不经常被搜索,则在有必要之前不要索引它们。(索引可根据需要添加和删除。)
  • like很慢。一般来说,最好是使用fulltext而不是like。
  • 数据库是不断变化的实体。一组优化良好的表一会儿后可能就面目全非了。由于表的使用和内容的更改,理想的优化和配置也会改变。
  • 最重要的规则就是,每条规则在某些条件下都会被打破。

 

CentOS7安装使用MySQL

标签:错误   占位符   指定   发布   完整   sql语句   触发器   set   main   

原文地址:http://www.cnblogs.com/joeblackzqq/p/6676061.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!