Sql 计算在其他对象修改日期之后创建的对象数

我正在尝试统计t-SQL查询中特定日期之后创建的所有帖子：

select a.ParentID as [Post Link]
     , count(DISTINCT a.id) as [Answers after edition]
     , q.LastEditDate as [Last Edition]
     , a.creationdate as [Answered On]
     , DATEDIFF(hour, q.LastEditDate, a.creationdate) as [Hours after edition]
from posts a 
inner join posts q on q.id = a.ParentID
where q.posttypeid = 1 -- Q
and a.posttypeid = 2 -- A
and q.LastEditorUserId = ##userid##
and a.creationdate > q.LastEditDate
and q.AnswerCount > 1
group by a.ParentID

其结果是：

Column 'posts.LastEditDate' is invalid in the select list because it is not contained in either an aggregate function or the GROUP BY clause.

此查询是对其他查询的修改：

select q.id as [Post Link]
     , q.owneruserid as [User Linkl]
     , q.LastEditDate as [Last Edition]
     , a.creationdate as [Answered On]
     , DATEDIFF(hour, q.LastEditDate, a.creationdate) as [Hours after edition]
from posts q 
inner join posts a on a.id = q.AcceptedAnswerId
where q.posttypeid = 1 -- Q
and a.posttypeid = 2 -- A
and q.LastEditorUserId = ##userid##
and a.creationdate > q.LastEditDate
order by [Hours after edition]

变化在于后者仅比较存在AcceptedAnswerId的帖子。我想做同样的事情，但这次将具有相同q.Id的分组

这正是我想要的：

select a.id as [Answer Link]
     , q.owneruserid as [User Linkl]
     , q.LastEditDate as [Last Edition]
     , a.creationdate as [Answered On]
     , DATEDIFF(hour, q.LastEditDate, a.creationdate) as [Hours after edition]
from posts a 
inner join posts q on q.id = a.ParentID
where q.posttypeid = 1 -- Q
and a.posttypeid = 2 -- A
and q.LastEditorUserId = ##userid##
and a.creationdate > q.LastEditDate
and q.AnswerCount > 1

但不按ParentId分组

Rhis正是我所期望的：

Post Link   | Answer Count | Hours after edition
http://link |            3 |                   3

---

我不太想在编辑后保留几个小时，但如果能保留得更好的话。

将评论转化为答案。在您的第一段代码中：

 , q.LastEditDate as [Last Edition]
 , a.creationdate as [Answered On]

您在这方面遇到了一个错误，因为它们不在GROUPBY语句中，并且脚本不知道如何处理多个值。如果您将它们包括在group by中，您可能会为每个父ID获得几行。我认为您希望获取最大日期作为最后一次编辑，最小日期作为第一次创建

---

顺便说一下……我想你一定是在MSSQL上。如果您在MySQL中，它会在没有GROUPBY的情况下执行您的查询，并随机填充这两个字段。哎呀

其中是列上的筛选器，是聚合上的筛选器。因为我们正在将它们移动到maxdate或mindate，它们正在成为聚合，需要在having子句中，而不是where子句中。从您需要的订单所在的组中进行选择

Braiam提供的最终脚本：

select a.ParentId as [Post Link]
 , count(DISTINCT a.id) as [Answers after edition]
 --, q.LastEditDate as [Last Edition]
 --, a.creationdate as [Answered On]
 --, DATEDIFF(hour, q.LastEditDate, a.creationdate) as [Hours after edition]
from posts a 
inner join posts q on q.id = a.ParentID
where q.posttypeid = 1 -- Q
and a.posttypeid = 2 -- A
and q.LastEditorUserId = ##userid##
and q.AnswerCount > 1
group by a.ParentId
having max(q.LastEditDate) < min(a.creationdate)

---

如果一个父ID找到多个LastEditDate或creationdate，您是否有任何逻辑可以使用哪一个LastEditDate或creationdate？它返回的“posts.lastedDate”错误无效，因为没有关于如何处理它的多个值的说明。第一个查询的简单解决方案是使用maxlasteditdate和mincreationdate。其中，列上有一个筛选器，聚合上有一个筛选器。在group by parentID下，输入mina.creationdate>maxq.lastedDate。它是从groupby Having的位置选择的我想你一定是在MSSQL上这里不，使用SEDE和T-SQL，是的，在阅读后消除了错误。啊…是猜测给定的tsql，而不是mysql。我们应该清理这个问题…介意我编辑我的问题，包括你的最后脚本，我们会留下来作为你问题的答案吗？