Question

我不明白为什么这不起作用以及如何解决它，我尝试了各种各样的事情，比如写

select COUNT(p.OwnerUserId)

但这不起作用，我不明白错误消息。我不使用MS SQL（我使用SQLite和MySQL）。

如何编写此查询以便我可以将QC过滤10或50？（其中QC> 50 AND ...）

基本上将下面的SQL插入到此URL中，运行它，您将在结果中看到1。 http://data.stackexchange.com/stackoverflow/query/new

SELECT
    TOP 100
    p.OwnerUserId  AS [User Link],
    sum(ViewCount) as VC,
    avg(ViewCount) as AVC,
    COUNT(p.OwnerUserId ) as QC

FROM Posts p
join Users on p.OwnerUserId = Users.Id
where PostTypeId = 1 and ViewCount<10000 and CommunityOwnedDate is null
group by p.OwnerUserId
order by AVC desc

Answer 1

您需要使用Having子句来过滤聚合字段

试试这个：

SELECT
    TOP 100
    p.OwnerUserId  AS [User Link],
    sum(ViewCount) as VC,
    avg(ViewCount) as AVC,
    COUNT(p.OwnerUserId ) as QC

FROM Posts p
join Users on p.OwnerUserId = Users.Id
where PostTypeId = 1 and ViewCount<10000 and CommunityOwnedDate is null
group by p.OwnerUserId  
HAVING COUNT(p.OwnerUserId ) > 50
order by AVC desc

Answer 2

使用聚合时，应使用having代替where。

Answer 3

SELECT
    TOP 100
    p.OwnerUserId  AS [User Link],
    sum(ViewCount) as VC,
    avg(ViewCount) as AVC,
    COUNT(p.OwnerUserId ) as QC
FROM Posts p
join Users on p.OwnerUserId = Users.Id
where PostTypeId = 1 and ViewCount<10000 and CommunityOwnedDate is null
group by p.OwnerUserId
HAVING COUNT(p.OwnerUserId ) between 10 and 50   -- <<<<<
order by AVC desc

另一种选择是使其成为子查询

SELECT
    TOP 100
FROM (
SELECT
    p.OwnerUserId  AS [User Link],
    sum(ViewCount) as VC,
    avg(ViewCount) as AVC,
    COUNT(p.OwnerUserId ) as QC
FROM Posts p
join Users on p.OwnerUserId = Users.Id
where PostTypeId = 1 and ViewCount<10000 and CommunityOwnedDate is null
group by p.OwnerUserId
) SQ
WHERE QC >= 50
order by AVC desc

我如何在where语句中使用Count（*）？

3 个答案: