MySQL 常用语句

1 新增

1.1 新增字段

ALTER TABLE shanghai_info ADD created DATETIME;

2 查询

3 修改

3.1 修改表名

alter table ts01 rename to ts01_new;

3.2 修改字段

ALTER TABLE shanghai_info MODIFY created DATETIME DEFAULT NULL COMMENT "新增时间";

3.3 替换字段中的某个字符串

UPDATE tableName SET phone = REPLACE(phone,'.000000','')

4 删除

4.1 清空表数据

• truncate 是整体删除（速度较快）， delete 是逐条删除（速度较慢）。
• truncate 不写服务器 log，delete 写服务器 log，也就是 truncate 效率比 delete 高的原因。
• truncate 不激活 trigger (触发器)，但是会重置 Identity（标识列、自增字段），相当于自增列会被置为初始值，又重新从 1 开始记录，而不是接着原来的 ID 数。而 delete 删除以后，Identity 依旧是接着被删除的最近的那一条记录 ID 加 1 后进行记录。

4.2 删除某字段

ALTER TABLE table_name DROP COLUMN column_name;

5 重复记录操作

5.1 查询重复

  如，查找重复记录是根据单个字段（peopleId）来判断。

SELECT * FROM Tpeople 
  WHERE peopleId IN ( SELECT peopleId FROM Tpeople GROUP BY peopleId HAVING COUNT(peopleId) > 1)

  根据多个字段来判断

SELECT * FROM illegal_info 
  WHERE (illegal_code, city_id, reason) IN ( SELECT illegal_code, city_id, reason FROM illegal_info GROUP BY illegal_code, city_id, reason HAVING COUNT(*) > 1) ORDER BY illegal_code;
 

5.2 删除表中多余的重复记录

  例如，重复记录是根据单个字段（peopleId）来判断，只保留最先增加的记录，下面是保留 ID 最小的记录

DELETE FROM test WHERE id NOT IN (
	SELECT a.id FROM (
		SELECT MIN(id) AS id FROM test GROUP BY username HAVING COUNT(username) > 1 
		)AS a
	) AND username IN (
		SELECT b.username FROM (
			SELECT username FROM test GROUP BY username HAVING COUNT(username) > 1
			) AS b
		); 

  根据多个字段进行删除，只留有 rowid 最小的记录

delete from vitae a
where (a.peopleId,a.seq) in (select peopleId,seq from vitae group by peopleId,seq having count(*) > 1)
and rowid not in (select min(rowid) from vitae group by peopleId,seq having count(*)>1)

6 索引

6.1 添加普通索引

• 存在表时添加普通索引

ALTER TABLE shanghai_info ADD INDEX (car_number);

• 存在表时添加唯一索引

ALTER TABLE llegal_info ADD UNIQUE INDEX [索引名](字段1, 字段2, 字段3);

• 建表时添加索引

CREATE TABLE `hunan_car` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `car_number` varchar(255) DEFAULT NULL COMMENT '车牌号',
  `vin` varchar(255) DEFAULT NULL COMMENT '车架号',
  `eng` varchar(255) DEFAULT NULL COMMENT '发动机',
  `car_type` varchar(255) DEFAULT NULL COMMENT '车辆类型',
  `status` int(11) DEFAULT NULL COMMENT '读取状态，0 未读 / 1 已读',
  `success` int(11) DEFAULT NULL COMMENT '读取成功状态 0 爬取失败 / 1 爬取成功',
  PRIMARY KEY (`id`) USING BTREE,
  INDEX `car_number` (`car_number`) 
) ENGINE=InnoDB AUTO_INCREMENT=110280 DEFAULT CHARSET=utf8;

6.2 添加唯一索引

• 存在表时添加索引

ALTER TABLE shanghai_info ADD UNIQUE (car_number);

• 建表时添加索引

CREATE TABLE `hunan_car` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `car_number` varchar(255) DEFAULT NULL COMMENT '车牌号',
  `vin` varchar(255) DEFAULT NULL COMMENT '车架号',
  `eng` varchar(255) DEFAULT NULL COMMENT '发动机',
  `car_type` varchar(255) DEFAULT NULL COMMENT '车辆类型',
  `status` int(11) DEFAULT NULL COMMENT '读取状态，0 未读 / 1 已读',
  `success` int(11) DEFAULT NULL COMMENT '读取成功状态 0 爬取失败 / 1 爬取成功',
  PRIMARY KEY (`id`) USING BTREE,
  UNIQUE KEY `car_number`` (`car_number``) 
) ENGINE=InnoDB AUTO_INCREMENT=110280 DEFAULT CHARSET=utf8;

6.3 创建组合索引

• 存在表时添加索引

ALTER TABLE illegal_info ADD INDEX idx_reason_money_city_id (reason,money,city_id);

• 建表时添加索引

CREATE TABLE `illegal_info` (
  `id` bigint(11) NOT NULL AUTO_INCREMENT COMMENT '自增id',
  `illegal_code` varchar(6) DEFAULT NULL COMMENT '违章代码，截取第二个字符则为扣分，扣分为7则为12',
  `reason` varchar(300) DEFAULT NULL COMMENT '违章信息',
  `city_id` int(11) DEFAULT '0' COMMENT '城市id',
  `money` varchar(10) DEFAULT NULL COMMENT '罚款',
  PRIMARY KEY (`id`),
  KEY `idx_reason_money_city_id` (`reason`,`money`,`city_id`)
) ENGINE=InnoDB AUTO_INCREMENT=16448 DEFAULT CHARSET=utf8 COMMENT='各地违章信息扣分罚款详情表'

6.4 删除索引

ALTER TABLE shanghai_info DROP INDEX cphm;