SQL 更新计算列 [英] SQL update a calculated column
问题描述
我需要更新表格中的一列.该列的计算方式如下:
I have a column in a table which I need to update. The column is computed like this:
SELECT CASE WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 27 THEN ((w_staerke/100*70) + (w_technik/100*30))
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 31 THEN ((w_staerke/100*70) + (w_technik/100*30))
END AS marktwert
FROM _spieler;
我想更新该表中所有记录的列.
I want to update column on all the records from that table.
我可以使用类似的东西吗
Can I use something like
UPDATE _spieler SET marktwert = CASE WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 27 THEN ((w_staerke/100*70) + (w_technik/100*30))
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 31 THEN ((w_staerke/100*70) + (w_technik/100*30))
END;
该查询似乎是正确的,但是它将每一行的marktwert"中的值设置为零.
The query seems to be correct, however it sets the value in "marktwert" to zero for every row.
CREATE TABLE `_spieler` (
`id` int(10) NOT NULL,
`vorname` varchar(30) DEFAULT NULL,
`nachname` varchar(30) DEFAULT NULL,
`geburtstag` date NOT NULL,
`w_staerke` tinyint(3) NOT NULL,
`w_technik` tinyint(3) NOT NULL,
`marktwert` int(10) NOT NULL DEFAULT '0',
`age` tinyint(3) DEFAULT NULL,
) ENGINE=InnoDB DEFAULT CHARSET=utf8 AUTO_INCREMENT=535 ;
(1, 'Adam', 'Federici', '1985-01-31', 30, 20, 0, NULL),
(2, 'Ryan', 'Allsop', '1992-06-17', 20, 30, 0, NULL),
(3, 'Tyrone', 'Mings', '1980-03-13', 40, 20, 0, NULL),
(4, 'Joe', 'Bennett', '1990-03-28', 25, 30, 0, NULL),
(5, 'Charlie', 'Daniels', '1986-09-07', 50, 30, 0, NULL);
表定义和一些示例数据
UPDATE _spieler
SET marktwert = CASE WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 27 THEN ((w_staerke/100*70) + (w_technik/100*30)) * 600000
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 31 THEN (((w_staerke/100*70) + (w_technik/100*30)) * 600000) - 5000000
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) > 31 THEN (((w_staerke/100*70) + (w_technik/100*30)) * 600000) - 10000000 END
添加了最终查询
推荐答案
您的问题是您没有为 31 岁以上的玩家 (Spieler) 计算任何 marktwert
值 (geburtstag =birthday)).您的 UPDATE 语句试图将 NULL
写入 marktwert
列,该列定义为 NOT NULL
.这会导致错误.
Your problem is that you do not calculate any marktwert
value for players (Spieler) who is older than 31 years (geburtstag = birthday). Your UPDATE statement is trying to write NULL
into the marktwert
column, which is defined as NOT NULL
. And that results in an error.
解决方案:
1) 在您的 CASE
语句中使用 ELSE
并设置默认值:
1) User ELSE
in your CASE
statement and set a default value:
UPDATE _spieler SET marktwert = CASE
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 27 THEN ((w_staerke/100*70) + (w_technik/100*30))
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 31 THEN ((w_staerke/100*70) + (w_technik/100*30))
ELSE 0
END;
2) 允许列 marktwert
的 NULL
值:
2) Allow NULL
value for column marktwert
:
CREATE TABLE `_spieler` (
...
`marktwert` int(10) NULL DEFAULT '0',
...
)
3) 使用 WHERE
条件:
UPDATE _spieler SET marktwert = CASE
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 27 THEN ((w_staerke/100*70) + (w_technik/100*30))
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 31 THEN ((w_staerke/100*70) + (w_technik/100*30))
END
WHERE TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 31;
更新:您还可以删除 marktwert
列并使用 查看(计算表):
Update: You can also remove the marktwert
column and use a view (calculated table) instead:
CREATE VIEW `_spieler_view` AS SELECT s.*,
CASE
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 27 THEN ((w_staerke/100*70) + (w_technik/100*30))
WHEN TIMESTAMPDIFF(YEAR, geburtstag, NOW()) <= 31 THEN ((w_staerke/100*70) + (w_technik/100*30))
END AS marktwert_calculated
from _spieler s ;
更新 2:
如果您使用 MariaDB,您还可以使用虚拟(计算)列一个>
If you use MariaDB you could also use Virtual (Computed) Columns
这篇关于SQL 更新计算列的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!