zoukankan      html  css  js  c++  java
  • Mysql事件调度器学习

    在cassandra数据库中,有一个叫做TTL的功能,即插入一条记录时,可以指定某一字段对应的TTL值,比如30s,那么当TTL到达30s后该条记录就会被自动删除。目前MySQL并未直接提供TTL的功能,但是我们可以通过event schedule功能实现。

    创建一张测试表

    CREATE TABLE `access_token` (
      `token_id` varchar(100) COLLATE utf8_bin NOT NULL DEFAULT '',
      `created_timestamp` bigint(20) DEFAULT NULL,
      `expired_timestamp` bigint(20) DEFAULT NULL,
      `grant_type` varchar(100) COLLATE utf8_bin DEFAULT NULL,
      `refresh_token` varchar(100) COLLATE utf8_bin DEFAULT NULL,
      `scope` varchar(200) COLLATE utf8_bin DEFAULT NULL,
      `user_info` varchar(200) COLLATE utf8_bin DEFAULT NULL,
      PRIMARY KEY (`token_id`)
    ) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_bin;

    首先假设一些记录的expired_timestamp过期时间是“2020-12-31 23:59:59”,将该时间转换成unix时间戳,并乘以1000转换成毫秒存储,其他部分记录的expired_timestamp过期时间设置为当前时间

    select unix_timestamp('2020-12-31 23:59:59')*1000;
    +--------------------------------------------+
    | unix_timestamp('2020-12-31 23:59:59')*1000 |
    +--------------------------------------------+
    |                              1609430399000 |
    +--------------------------------------------+
    1 row in set (0.00 sec)
    select unix_timestamp(current_timestamp)*1000;
    +----------------------------------------+
    | unix_timestamp(current_timestamp)*1000 |
    +----------------------------------------+
    |                          1576319276000 |
    +----------------------------------------+
    1 row in set (0.00 sec)

     插入测试数据

    insert into access_token value('9T8F9EX42G9OXHPIDM98YBK74','1576319276000','1609430399000','mannual','6JL351QA38Z9ONOL9PWJ64A4U','test','david');    #该条测试数据的expired_timestamp是上述的 2020-12-31 23:59:59
    insert into access_token value('AG8FLP3JQKGNCRZY0OPVTYV2W','1576319276000','1576319276000','auto','D7QOBDZPBKEP7TVWUG10XJA0S','ceshi','brain');      #该条测试数据的expired_timestamp的当前时间current_timestamp

     查看测试数据

    select * from access_token;
    +---------------------------+-------------------+-------------------+------------+---------------------------+-------+-----------+
    | token_id                  | created_timestamp | expired_timestamp | grant_type | refresh_token             | scope | user_info |
    +---------------------------+-------------------+-------------------+------------+---------------------------+-------+-----------+
    | 9T8F9EX42G9OXHPIDM98YBK74 |     1576319276000 |     1609430399000 | mannual    | 6JL351QA38Z9ONOL9PWJ64A4U | test  | david     |
    | AG8FLP3JQKGNCRZY0OPVTYV2W |     1576319276000 |     1576319276000 | auto       | D7QOBDZPBKEP7TVWUG10XJA0S | ceshi | brain     |
    +---------------------------+-------------------+-------------------+------------+---------------------------+-------+-----------+
    2 rows in set (0.00 sec)

    将数据转换成timestamp显示

    select token_id, from_unixtime(created_timestamp/1000) as created_timestamp, from_unixtime(expired_timestamp/1000) as expired_timestamp, grant_type, refresh_token , scope, user_info from access_token;
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | token_id                  | created_timestamp        | expired_timestamp        | grant_type | refresh_token             | scope | user_info |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | 9T8F9EX42G9OXHPIDM98YBK74 | 2019-12-14 18:27:56.0000 | 2020-12-31 23:59:59.0000 | mannual    | 6JL351QA38Z9ONOL9PWJ64A4U | test  | david     |
    | AG8FLP3JQKGNCRZY0OPVTYV2W | 2019-12-14 18:27:56.0000 | 2019-12-14 18:27:56.0000 | auto       | D7QOBDZPBKEP7TVWUG10XJA0S | ceshi | brain     |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    2 rows in set (0.01 sec)

     创建event schedule,具体含义可参考https://dev.mysql.com/doc/refman/5.7/en/create-event.html

    create event testdb.delete_access_token_expired_record
        on schedule every 2 minute
        do delete from testdb.access_token where expired_timestamp < unix_timestamp(current_timestamp)*1000 ;

    查看该event信息

    select * from information_schema.eventsG
    *************************** 1. row ***************************
           EVENT_CATALOG: def
            EVENT_SCHEMA: testdb
              EVENT_NAME: delete_access_token_expired_record
                 DEFINER: root@localhost
               TIME_ZONE: SYSTEM
              EVENT_BODY: SQL
        EVENT_DEFINITION: delete from testdb.access_token where expired_timestamp < unix_timestamp(current_timestamp)*1000
              EVENT_TYPE: RECURRING
              EXECUTE_AT: NULL
          INTERVAL_VALUE: 2
          INTERVAL_FIELD: MINUTE
                SQL_MODE: ONLY_FULL_GROUP_BY,STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_ENGINE_SUBSTITUTION
                  STARTS: 2019-12-14 18:38:59
                    ENDS: NULL
                  STATUS: ENABLED
           ON_COMPLETION: NOT PRESERVE
                 CREATED: 2019-12-14 18:38:59
            LAST_ALTERED: 2019-12-14 18:38:59
           LAST_EXECUTED: 2019-12-14 10:38:59
           EVENT_COMMENT: 
              ORIGINATOR: 543306
    CHARACTER_SET_CLIENT: utf8mb4
    COLLATION_CONNECTION: utf8mb4_0900_ai_ci
      DATABASE_COLLATION: utf8mb4_0900_ai_ci
    1 row in set (0.01 sec)

    该event每两分钟运行一次,如果access_token表中字段expired_timestamp的时间戳小于当前时间戳,说明该条记录已经过期,可以删除。

    再次查看access_token表时,发现过期数据已经被删除

    select token_id, from_unixtime(created_timestamp/1000) as created_timestamp, from_unixtime(expired_timestamp/1000) as expired_timestamp, grant_type, refresh_token , scope, user_info from access_token;
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | token_id                  | created_timestamp        | expired_timestamp        | grant_type | refresh_token             | scope | user_info |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | 9T8F9EX42G9OXHPIDM98YBK74 | 2019-12-14 18:27:56.0000 | 2020-12-31 23:59:59.0000 | mannual    | 6JL351QA38Z9ONOL9PWJ64A4U | test  | david     |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    1 row in set (0.00 sec)

    再次插入两条数据,一条数据的expired_timestamp为当前时间戳,一条数据的expired_timestamp为5分钟后

    insert into access_token value('4MXCXCXQRZRY1FVYAO9DG99F9','1576319276000','1576321202000','mannual','EN525M7AW448RBR3386ODM9T3','test','tim');
    
    insert into access_token value('19XKLPXCAYGDJKH9VB7KA9V3Y','1576321336000','1576321504000','auto','3YBVRXTBW7578SVXMEJ9UK67E','ceshi','jhon');

    查看当前数据

    select token_id, from_unixtime(created_timestamp/1000) as created_timestamp, from_unixtime(expired_timestamp/1000) as expired_timestamp, grant_type, refresh_token , scope, user_info from access_token;
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | token_id                  | created_timestamp        | expired_timestamp        | grant_type | refresh_token             | scope | user_info |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | 19XKLPXCAYGDJKH9VB7KA9V3Y | 2019-12-14 19:02:16.0000 | 2019-12-14 19:05:04.0000 | auto       | 3YBVRXTBW7578SVXMEJ9UK67E | ceshi | jhon      |
    | 4MXCXCXQRZRY1FVYAO9DG99F9 | 2019-12-14 18:27:56.0000 | 2019-12-14 19:00:02.0000 | mannual    | EN525M7AW448RBR3386ODM9T3 | test  | tim       |
    | 9T8F9EX42G9OXHPIDM98YBK74 | 2019-12-14 18:27:56.0000 | 2020-12-31 23:59:59.0000 | mannual    | 6JL351QA38Z9ONOL9PWJ64A4U | test  | david     |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    3 rows in set (0.00 sec)

    过两分钟后,再次查看数据时,发现刚才插入access_token为当前时间戳的数据已经被删除

    select token_id, from_unixtime(created_timestamp/1000) as created_timestamp, from_unixtime(expired_timestamp/1000) as expired_timestamp, grant_type, refresh_token , scope, user_info from access_token;
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | token_id                  | created_timestamp        | expired_timestamp        | grant_type | refresh_token             | scope | user_info |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | 19XKLPXCAYGDJKH9VB7KA9V3Y | 2019-12-14 19:02:16.0000 | 2019-12-14 19:05:04.0000 | auto       | 3YBVRXTBW7578SVXMEJ9UK67E | ceshi | jhon      |
    | 9T8F9EX42G9OXHPIDM98YBK74 | 2019-12-14 18:27:56.0000 | 2020-12-31 23:59:59.0000 | mannual    | 6JL351QA38Z9ONOL9PWJ64A4U | test  | david     |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    2 rows in set (0.00 sec)

    五分钟以后再次查看,刚才插入access_token为当前时间五分钟后的时间戳的数据也已经被删除

    mysql> select token_id, from_unixtime(created_timestamp/1000) as created_timestamp, from_unixtime(expired_timestamp/1000) as expired_timestamp, grant_type, refresh_token , scope, user_info from access_token;
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | token_id                  | created_timestamp        | expired_timestamp        | grant_type | refresh_token             | scope | user_info |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    | 9T8F9EX42G9OXHPIDM98YBK74 | 2019-12-14 18:27:56.0000 | 2020-12-31 23:59:59.0000 | mannual    | 6JL351QA38Z9ONOL9PWJ64A4U | test  | david     |
    +---------------------------+--------------------------+--------------------------+------------+---------------------------+-------+-----------+
    1 row in set (0.00 sec)

    禁用该event

    alter event testdb.delete_access_token_expired_record disable;

    再次查看该event

    select * from information_schema.eventsG
    *************************** 1. row ***************************
           EVENT_CATALOG: def
            EVENT_SCHEMA: testdb
              EVENT_NAME: delete_access_token_expired_record
                 DEFINER: root@localhost
               TIME_ZONE: SYSTEM
              EVENT_BODY: SQL
        EVENT_DEFINITION: delete from testdb.access_token where expired_timestamp < unix_timestamp(current_timestamp)*1000
              EVENT_TYPE: RECURRING
              EXECUTE_AT: NULL
          INTERVAL_VALUE: 2
          INTERVAL_FIELD: MINUTE
                SQL_MODE: ONLY_FULL_GROUP_BY,STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_ENGINE_SUBSTITUTION
                  STARTS: 2019-12-14 18:38:59
                    ENDS: NULL
                  STATUS: DISABLED
           ON_COMPLETION: NOT PRESERVE
                 CREATED: 2019-12-14 18:38:59
            LAST_ALTERED: 2019-12-14 18:38:59
           LAST_EXECUTED: 2019-12-14 11:14:59
           EVENT_COMMENT: 
              ORIGINATOR: 543306
    CHARACTER_SET_CLIENT: utf8mb4
    COLLATION_CONNECTION: utf8mb4_0900_ai_ci
      DATABASE_COLLATION: utf8mb4_0900_ai_ci
    1 row in set (0.01 sec)

     创建一个event,每天夜里一点定期你删除过期数据

    CREATE EVENT testdb.delete_access_token_expired_record
        ON SCHEDULE EVERY 24 HOUR STARTS DATE_ADD(DATE_ADD(CURDATE(), INTERVAL 1 DAY), INTERVAL 1 HOUR)
        ON COMPLETION PRESERVE DISABLE ON SLAVE
        DO DELETE FROM testdb.access_token WHERE expired_timestamp < unix_timestamp(CURRENT_TIMESTAMP)*1000 ;
  • 相关阅读:
    NX CAM二次开发-UF_OPER_ask_program_group查询操作所在的程序顺序组tag
    NX CAM二次开发-UF_CAMGEOM_append_items设置切削区域
    NX CAM二次开发-UF_SETUP_create创建加工设置
    NX CAM二次开发-UF_SETUP_ask_setup获取加工设置
    NX CAM二次开发-UF_OPER_ask_oper_type查询指定操作的子类型
    NX CAM二次开发-UF_SETUP_ask_program_root获得程序组视图的根节点
    NX CAM二次开发-UF_OPER_ask_geom_group查询指定操作所在的几何组tag
    NX CAM二次开发-UF_UI_ONT_switch_view切换当前加工导航器指定的视图
    NX CAM二次开发-UF_SETUP_delete_setup删除加工设置
    NX CAM二次开发-UF_UI_PARAM_edit_object编辑CAM加工对象
  • 原文地址:https://www.cnblogs.com/ilifeilong/p/12040579.html
Copyright © 2011-2022 走看看