Question

我可能做错了事，我之前进行了搜索，发现一些变通办法告诉我在MySQL上无法实现，其他人则认为这是由于MySQL优化器造成的，因此您可以将其关闭并继续，但这对我不起作用。

我只想要一个简单的解决方法来解决这个问题。我唯一可以在订单表中找到id_address_delivery和id_address_invoice的表，该数据也位于其他表上，但是可以为null，因此无法选择在其他位置进行搜索。

SET optimizer_switch = 'derived_merge=off';

update orders 
set id_customer = (select id_customer from customer where email like 'foo@foo.com'),
id_address_delivery = (select id_address_delivery from orders where id_customer = (select id_customer from customer where email like 'foo@foo.com') LIMIT 1),
id_address_invoice = (select id_address_invoice from orders where id_customer = (select id_customer from customer where email like 'foo@foo.com') LIMIT 1)
where id_customer = (select id_customer from customer where email like 'foo2@foo2.com');

我得到错误代码：1093。即使我采用

，也无法在mysql工作台上的FROM子句中指定要更新的目标表'orders'

SET optimizer_switch = 'derived_merge=off';

是否有应对这种情况的选择？我感谢设置一些变量，例如：

SET @iad = (select id_address_delivery from orders where id_customer = (select id_customer from customer where email like 'foo@foo.com') LIMIT 1);

然后将该var设置为类似值：

id_address_delivery = @iad;

执行此操作时我没有收到错误响应，但它持续的时间很多（我完全不知道为什么），并且出现tiemout消息（30秒），我尝试将其设置为120秒并获得相同的超时消息。

编辑：

我尝试使用别名，但没有结果。同样的错误：

update orders AS sor
set sor.id_customer = (select id_customer from customer where email like 'foo@foo.com'),
sor.id_address_delivery = (select a.id_address_delivery from orders as a where a.id_customer = (select id_customer from customer where email like 'foo@foo.com') LIMIT 1),
sor.id_address_invoice = (select b.id_address_invoice from orders as b where b.id_customer = (select id_customer from customer where email like 'foo@foo.com') LIMIT 1)
where pso.id_customer = (select id_customer from customer where email like 'foo2@foo2.com');

/ EDIT：

我阅读了一些被标记为同一问题的帖子，但是在其他子句中找到了一些解决方法，并且我不知道如何在我的特殊情况下应用相同的内容。

我该如何进行？谢谢。

Answer 1

作为其他答案，如果您要对表进行UPDATE / INSERT / DELETE，则无法在内部查询中引用该表（但是，您可以引用该外部表中的字段...）

检查以下与您的问题相同的问题，即可解决。

MySQL Error 1093 - Can't specify target table for update in FROM clause

You can't specify target table for update in FROM clause

Answer 2

我终于处理了设置vars的问题，但仍然不知道为什么这会导致超时。我在where子句上添加了空保护，以便在某些var初始化失败时不插入空值。

为您的利益提供了我的工作示例副本：

use dbname;

SET SQL_SAFE_UPDATES = 0;
SET optimizer_switch = 'derived_merge=off';

#-- data asociated with wrong@foomail.com will be associated to correct@foomail.com

SET @delad = (SELECT id_address_delivery FROM orders WHERE customer = (SELECT id_customer FROM customer WHERE email LIKE 'correct@foomail.com') LIMIT 1);
SET @delin = (SELECT id_address_invoice FROM orders WHERE id_customer = (SELECT id_customer FROM customer WHERE email LIKE 'correct@foomail.com') LIMIT 1);
SET @uid = (SELECT id_customer FROM customer WHERE email LIKE 'correct@foomail.com' LIMIT 1);
SET @buid = (SELECT id_customer FROM customer WHERE email LIKE 'wrong@foomail.com' LIMIT 1);

UPDATE orders
SET id_customer = @uid,
id_address_delivery = @delad,
id_address_invoice = @delin
WHERE @uid is not null AND @delad is not null AND @delin is not null AND id_customer = @buid;

UPDATE cart SET
 id_customer = @uid,
 id_address_delivery = @delad,
 id_address_invoice = @delin
 WHERE @buid is not null AND @uid is not null AND @delad is not null AND @delin is not null AND id_customer = @buid;

这只是将子查询值抽象为变量，并将其放在子查询所在的位置。我在不需要此解决方法之前就已经做到了，但是我认为它在sql服务器上（也许）。

希望它可以帮助某人。

Answer 3

create view foo_data as ( select o.id_customer,o.id_address_delivery,o.id_address_invoice from  orders as o inner join customer as c on o.id_customer = c.id_customer where c.email like 'foo@foo.com' limit 1)    


update orders as o inner join foo_data as f on f.id_customer = o.id_customer
set optimizer_switch = 'derived_merge=off',
set o.id_customer = f.id_customer,
set o.id_address_delivery  = f.id_address_delivery ,
set o.id_address_invoice = f.id_address_invoice,
where o.email email like 'foo2@foo2.com';

drop view foo_data;

Please try using creating temp view

mysql错误代码1093：您不能在FROM子句中指定目标表进行更新

3 个答案: