给@Mathias Bynens的回答补充一点。当我运行这个时，我得到了一个错误，因为外键检查

mysql -Nse 'SHOW TABLES' <database_name> | while read table; do mysql -e "SET FOREIGN_KEY_CHECKS=0; DROP TABLE $table" <database_name>; done

如果数据库中有视图，则返回一个错误。我必须通过drop view <view_name>手动清除视图;

TB=$( mysql -Bse "show tables from DATABASE" );
for i in ${TB};
    do echo "Truncating table ${i}";
    mysql -e "set foreign_key_checks=0; set unique_checks=0;truncate table DATABASE.${i}; set foreign_key_checks=1; set unique_checks=1";
    sleep 1;
done

--

大卫,

感谢您花时间格式化代码，但这是它应该如何应用。

库尔特

在UNIX或Linux机器上:

确保您处于bash shell中。从命令行运行这些命令，如下所示。

注意:

我将凭据存储在~/.my.cnf文件中，因此不需要在命令行上提供凭据。

注意:

CPM是数据库名称

我只展示了来自每个命令的结果的一小部分示例。

找到你的外键约束:

klarsen@Chaos:~$ mysql -Bse "select concat(table_name, ' depends on ', referenced_table_name)
             from information_schema.referential_constraints
             where constraint_schema = 'cpm'
             order by referenced_table_name"

Approval_external_system依赖于approval_request 地址视客户而定 Customer_identification取决于客户 External_id取决于客户 凭证取决于客户 Email_address取决于客户 Approval_request取决于客户 Customer_status取决于客户 Customer_image取决于客户

列出表和行数:

klarsen@Chaos:~$ mysql -Bse "SELECT table_name, table_rows FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'cpm'" | cat -n

 1  address 297
 2  approval_external_system    0
 3  approval_request    0
 4  country 189
 5  credential  468
 6  customer    6776
 7  customer_identification 5631
 8  customer_image  2
 9  customer_status 13639

截断你的表:

klarsen@Chaos:~$ TB=$( mysql -Bse "show tables from cpm" ); for i in ${TB}; do echo "Truncating table ${i}"; mysql -e "set foreign_key_checks=0; set unique_checks=0;truncate table cpm.${i}; set foreign_key_checks=1; set unique_checks=1"; sleep 1; done

截断表地址 截断表approval_external_system 截断表approval_request 截断表国家 截断表凭据 截断表客户 截断表customer_identification 截断表customer_image 截断表customer_status

验证它是否有效:

klarsen@Chaos:~$ mysql -Bse "SELECT table_name, table_rows FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'cpm'" | cat -n

 1  address 0
 2  approval_external_system    0
 3  approval_request    0
 4  country 0
 5  credential  0
 6  customer    0
 7  customer_identification 0
 8  customer_image  0
 9  customer_status 0
10  email_address   0

在Windows机器上:

注意:

CPM是数据库名称

C:\>for /F "tokens=*" %a IN ('mysql -Bse "show tables" cpm') do mysql -e "set foreign_key_checks=0; set unique_checks=0; truncate table %a; foreign_key_checks=1; set unique_checks=1" cpm

下面是一个截断本地数据库中所有表的过程。

如果它不起作用，请告诉我，我会删除这个答案。

未测试的

CREATE PROCEDURE truncate_all_tables()
BEGIN

   -- Declare local variables
   DECLARE done BOOLEAN DEFAULT 0;
   DECLARE cmd VARCHAR(2000);

   -- Declare the cursor
   DECLARE cmds CURSOR
   FOR
   SELECT CONCAT('TRUNCATE TABLE ', TABLE_NAME) FROM INFORMATION_SCHEMA.TABLES;

   -- Declare continue handler
   DECLARE CONTINUE HANDLER FOR SQLSTATE '02000' SET done=1;

   -- Open the cursor
   OPEN cmds;

   -- Loop through all rows
   REPEAT

      -- Get order number
      FETCH cmds INTO cmd;

      -- Execute the command
      PREPARE stmt FROM cmd;
      EXECUTE stmt;
      DROP PREPARE stmt;

   -- End of loop
   UNTIL done END REPEAT;

   -- Close the cursor
   CLOSE cmds;

END;

如果使用SQL server 2005，有一个隐藏的存储过程，允许您对数据库中的所有表执行一个或一组命令。下面是如何用这个存储过程调用TRUNCATE TABLE:

EXEC [sp_MSforeachtable] @command1="TRUNCATE TABLE ?"

这里有一篇很好的文章进一步阐述。

然而，对于MySql，你可以使用mysqldump并指定——add-drop-tables和——no-data选项来删除和创建所有忽略数据的表。是这样的:

mysqldump -u[USERNAME] -p[PASSWORD] --add-drop-table --no-data [DATABASE]

Mysqldump使用指南

SET FOREIGN_KEY_CHECKS = 0;

SELECT @str := CONCAT('TRUNCATE TABLE ', table_schema, '.', table_name, ';')
FROM   information_schema.tables
WHERE  table_type   = 'BASE TABLE'
  AND  table_schema IN ('db1_name','db2_name');

PREPARE stmt FROM @str;

EXECUTE stmt;

DEALLOCATE PREPARE stmt;

SET FOREIGN_KEY_CHECKS = 1;

溶液(1)

mysql> select group_concat('truncate',' ',table_name,';') from information_schema.tables where table_schema="db_name" into outfile '/tmp/a.txt';
mysql> /tmp/a.txt;

溶液2)

- Export only structure of a db
- drop the database
- import the .sql of structure 

——编辑----

earlier in solution 1, i had mentioned concat() instead of group_concat() which would have not returned the desired result