information_schema库中元数据的用法-CFANZ编程社区

1information_schema就是一个全局虚拟库，库内有很多视图，在mysql启动时自动生成：

mysql> show databases;
+------------------------+
| Database               |
+------------------------+
| information_schema     |
| #mysql50#20210113-1817 |
| mysql                  |
| performance_schema     |
| sys                    |
| tangTest               |
| testku                 |
+------------------------+
7 rows in set (0.02 sec)

2视图可以创建，我们可以将某个复杂的语句定义为一个视图，

1定义视图别名
create view test<100;

2执行查询语句：
select * from test;

3这个information_schema库中有很多的表，其中有一个TABLES表，可以desc tables看一下这个表结构

mysql> use information_schema;
Reading table information for completion of table and column names
You can turn off this feature to get a quicker startup with -A

Database changed
mysql> show tables;
+---------------------------------------+
| Tables_in_information_schema          |
+---------------------------------------+
| CHARACTER_SETS                        |
| TABLES                            |
| COLLATION_CHARACTER_SET_APPLICABILITY |

tables表结构里有很多方法，以下为常用方法：

mysql> desc tables;
+-----------------+---------------------+------+-----+---------+-------+
| Field           | Type                | Null | Key | Default | Extra |
+-----------------+---------------------+------+-----+---------+-------+
| TABLE_SCHEMA    | 表所在的库           | NO   |     |         |       |
| TABLE_NAME      | 表名                | NO   |     |         |       |
| ENGINE          | 表的存储引擎         | YES  |     | NULL    |       |
| TABLE_ROWS      | 表的行数             | YES  |     | NULL    |       |
| AVG_ROW_LENGTH  | 平均行长度           | YES  |     | NULL    |       |
| INDEX_LENGTH    | 索引长度      | YES  |     | NULL    |       |

实例1：

查询所有库对应的所有的表名：

查询testku和tangTest这两个库的所有表名：

mysql> select table_schema,table_name
    -> from information_schema.tables
    -> where table_schema='testku'
    -> union all
    -> select table_schema,table_name
    -> from information_schema.tables
    -> where table_schema='tangTest';
+--------------+---------------+
| table_schema | table_name    |
+--------------+---------------+
| testku       | course        |
| testku       | tanglinux     |
| testku       | tanglinuxcopy |
| tangTest     | AAA           |
+--------------+---------------+
4 rows in set (0.09 sec)

mysql>

以库名进行分组，让表在一行中显示：

mysql> select table_schema,group_concat(table_name)
    -> from information_schema.tables
    -> where table_schema='testku'
    -> group by table_schema;
+--------------+--------------------------------+
| table_schema | group_concat(table_name)       |
+--------------+--------------------------------+
| testku       | course,tanglinux,tanglinuxcopy |
+--------------+--------------------------------+
1 row in set (0.00 sec)

mysql>

2统计一下每个库的表的个数：

mysql> select table_schema,count(table_name)
    -> from information_schema.tables
    -> group by table_schema;
+--------------------+-------------------+
| table_schema       | count(table_name) |
+--------------------+-------------------+
| information_schema |                61 |
| mysql              |                32 |
| performance_schema |                87 |
| sys                |               101 |
| tangTest           |                 1 |
| testku             |                 3 |
+--------------------+-------------------+
6 rows in set (0.13 sec)

例3：

统计一下每个库的真实数据量（单位转换为M）：

每张表的数据量=平均行长度 * 表的行数 + 索引的长度

每张表的数据量=avg_row_length * table_rows + 索引长度

mysql> select table_schema,sum(avg_row_length*table_rows+index_length)/1024/1024
    -> from information_schema.tables
    -> group by table_schema;
+--------------------+-------------------------------------------------------+
| table_schema       | sum(avg_row_length*table_rows+index_length)/1024/1024 |
+--------------------+-------------------------------------------------------+
| information_schema |                                                  NULL |
| mysql              |                                            2.25408268 |
| performance_schema |                                            0.00000000 |
| sys                |                                            0.01562119 |
| tangTest           |                                            0.00000000 |
| testku             |                                            0.09374905 |
+--------------------+-------------------------------------------------------+
6 rows in set (0.02 sec)

生产实例演示：

对所有数据库里的每一张表进行单独备份

如果我们备份某个库里的一张表，我们可以这样写：mysqldump -uroot -p123456 world city >/tmp/world_city.sql

如果我们要利用元数据information_schema实现上面的语句就可以这样写：

mysqld -uroot -p123456 table_schema table_name >/tmp/table_schema_table_name.sql

from information_schema

上面的语句实现了查询所有的库对应所有的表，但我们需要把每张表的备份语句拼接出来，所以就用到了拼接语句concat():

concat("mysqldump -uroot -p123456 ",table_schema," ",table_name," >/tmp/",table_schema,"_",table_name,".sql")

查询出所有库对应表的拼接语句

select concat("mysqldump -uroot -p123456 ",table_schema," ",table_name," >/tmp/",table_schema,"_",table_name,".sql")
from information_schema.tables
where table_schema not in('sys','performance','information_schema')
into outfile '/tmp/bak.sh';

查看所有使用innodb引擎的表：

mysql> select table_schema,table_name,engine from information_schema.tables where engine='innodb';