如何编写此查询以仅过滤大于 50 的总和？我无法将其与 have 或 sum 结合使用，但我确定有某种方法。
select name,sum(score)
from submissions inner join hacker on submissions.hacker_id = hacker.hacker_id
group by submissions.hacker_id
order by sum(score) desc
having sum(score) > 50
摆弄表格在这里（它们没有什么异常，这个查询没有最后一行运行，但返回每个人的分数和姓名）：http://sqlfiddle.com/#!9/7a660d/16
您的订单应该在您拥有之后。即
select
name,sum(score)
from
submissions
inner join
hacker on submissions.hacker_id = hacker.hacker_id
group by
submissions.hacker_id
having
sum(score) > 50
order by
sum(score) desc
order by 应该在最后
select name,sum(score)
from submissions
inner join hacker on submissions.hacker_id = hacker.hacker_id
group by submissions.hacker_id
having sum(score) > 50
order by sum(score) desc
您也可以在 sum 语句中使用 CASE。这种方法适用于 windows 函数，尽管我还没有针对正常的 sum 聚合函数对其进行测试 选项 1（总和的情况）：
select name,sum(case when score>50 else 0 end)
from submissions inner join hacker on submissions.hacker_id = hacker.hacker_id
group by submissions.hacker_id
order by sum(score) desc
选项 2 Windows 功能：
select name,sum(case when score>50 else 0 end) over (partition by submissions.hacker_id)
from submissions inner join hacker on submissions.hacker_id = hacker.hacker_id
group by submissions.hacker_id
order by 子句略有变化，所以我省略了它，但您可以随时再次添加