这似乎有效，但我相信还有一种更优雅的解决方案，不需要使用join...

SELECT * FROM cus ; update cus inner join ( select id ,email,row_number() over(order by id asc) as rowid from cus )a on a.id = cus.id set cus.email = concat(a.rowid, a.email) ; SELECT * FROM cus ;

完整测试

-- 创建表 CREATE TABLE cus ( id INTEGER PRIMARY KEY, email TEXT NOT NULL ); -- 插入数据 INSERT INTO cus VALUES (0021, 'Clark'); INSERT INTO cus VALUES (0402, 'Dave'); INSERT INTO cus VALUES (005, 'Ava' ); SELECT * FROM cus ; update cus inner join ( select id ,email,row_number() over(order by id asc) as rowid from cus )a on a.id = cus.id set cus.email = concat(a.rowid, a.email) ; SELECT * FROM cus ;

也许这就是你想要做的：

WITH C AS ( SELECT email,ROW_NUMBER() OVER(ORDER BY id ASC) AS rowid FROM cus ) UPDATE cus JOIN C ON cus.email=C.email SET cus.email=CONCAT(rowid,'email@gmail.com');

将你想要更新的表(cus)与C的cte进行连接，然后进行相应的更新。

这里有一个演示

@QisM对于email不唯一时的语法提出了疑虑，由于OP没有提到，我同意如果email确实不唯一，这不是解决方案。因此，我对语法进行了轻微修改：

WITH C AS ( SELECT id, email, ROW_NUMBER() OVER(ORDER BY id ASC) AS rowid FROM cus ) UPDATE cus JOIN C ON cus.id=C.id AND cus.email=C.email SET cus.email=CONCAT(rowid,'email@gmail.com');

现在cte带有id，并且在JOIN C ON ..条件中，我添加了匹配id的检查。经过测试，如果电子邮件不唯一，这将修复该问题。