一道sql题引发的关于where、having、group by
上原题:
我的语句:
CREATE TRIGGER t on SC FOR INSERT AS DELETE FROM SC WHERE EXISTS(
SELECT * FROM inserted WHERE sno = SC.sno and cno = Sc.cno
AND sno IN (SELECT sno FROM SC WHERE grade <60)
AND sno IN (SELECT sno FROM SC WHERE cno IN (SELECT cno FROM Course WHERE Credit > 4) GROUP BY sno HAVING count(*) > 4)
AND sno not IN (SELECT top(0.1*(SELECT count(*) FROM Stu) sno FROM SC GROUP BY sno ORDER BY AVG(Grade) DESC)
)
总结反思:
- where 可以和 having 并用;where 用来筛选元组,筛选之后的元组才会用于 Group By;Having 作用于分组完成后的结果,其中可以有聚集函数,聚集函数不必出现在Select中
- exists 是看能不能根据元组中的信息从表中匹配出结果,个人认为exists作用于每个元组,把匹配结果不为空的元组选出来
- top 后面可以跟数量,如果想要百分比的数据,可以用百分比去乘以总数来得到想要的数量
- Order By 后面也可跟聚集函数,聚集函数不必出现在Select 中
- Having 用聚集函数进行筛选,Order By根据聚集函数进行排序
本文标题:一道sql题引发的关于where、having、group by
本文链接:https://www.haomeiwen.com/subject/txsugqtx.html
网友评论