需要SQL Server查询协助 [英] SQL Server query assistance needed

问题描述

我正在视图中存储一堆数据(将MS Access查询转换为视图).现在，我想做的是编写一个存储过程，以根据添加数据的时间提取数据.这是我正在运行的查询

I'm storing a bunch of data in a view (converted MS Access queries to views). Now what I'm trying to do is write a stored procedure to pull data based on when the data was added. Here is the query that I'm running

Select 
    Name, PlanID, ApptDate, 1stAppt, rn, 
from 
    (Select 
         *,   
         row_number() over (partition by PlanID Order BY AddedonDate desc) as rn 
     From vClientInfo)

因此，这使我的所有数据都可以了.我要解决的问题是，客户端实际上进入了第一个应用程序和第二个应用程序-基本上，我需要提取最新日期和最早的数据，并在查询中包含"2ndAppt".因此1stAppt和2ndAppt保留了一个字符串值，稍后我将使用...这两个值可能从1stAppt到2ndAppt不同-只有AddedOn日期才能告诉我是否有不同(最早的是1stAppt，最新的是2ndAppt)./p>

So this pulls all my data okay. The issue that I have to address is, the client actually comes in for the 1st appt and 2nd appt - Basically I need to pull the data for LATEST date, as well as EARLIEST, and include a '2ndAppt' in my query. So 1stAppt and 2ndAppt hold a string value that I will later use...The 2 values might differ from 1stAppt to 2ndAppt - and only AddedOn date will tell me if there was a different (earliest for 1stAppt, and latest for 2ndAppt).

Select 
    Name, PlanID, ApptDate, 1stAppt, 2ndAppt rn, 
From 
    (Select 
         *, 
         row_number() over (partition by PlanID Order BY AddedonDate desc) as rn 
     From  vClientInfp)

所以我做到了，我改变了...

So I did this and I changed this...

(Select *,  row_number() over (partition by PlanID Order BY AddedonDate *ASC*)

但是，由于我要处理大量数据且难以验证，因此我不确定这是否提取正确的数据.因此，基本上我需要进行合并，以便我的两个查询都能提取正确的数据.最初我获得1920条记录，所以有了工会，我应该获得的记录也一样多，但是，这个数字是两倍，所以3840，为什么?

However I'm not sure if this is pulling correct data as I'm working with a lot of data and hard to verify. So basically I need to do a union so that both my queries pull in the correct data. Originally I was getting 1920 records, so with a union i should be getting just as many too i assume, however, the number is double so 3840, why?

Select 
    Name, PlanID, ApptDate, 1stAppt, 2ndappt, rn, 
From 
    (Select 
         *,   
         row_number() over (partition by PlanID Order BY AddedonDate desc) as rn 
     From  vClientInfo

     Union All

     Select 
         Name, PlanID, ApptDate, 1stAppt, 2ndAppt, rn, 
     From 
         (Select 
              *,  
              row_number() over (partition by PlanID Order BY AddedonDate asc) as rn 
          From vClientInfo

推荐答案

您正在使用窗口函数，但没有对其进行过滤，所以有什么用呢?如果要最早和最新，请选择rn = 1:

You are using window function, but you are not filtering by it, so whats the point? If you want the earliest and the latest, choose rn = 1 :

Select Name, PlanID, ApptDate, 1stAppt,2ndappt, rn,
from (Select *,
            row_number() over (partition by PlanID Order BY AddedonDate desc) as rn
     from  vClientInfo
     Union All
     Select *, 
           row_number() over (partition by PlanID Order BY AddedonDate asc) as rn 
     from  vClientInfo)
WHERE rn = 1

如果您希望最早的应用程序是第一个，最新的应用程序是第二个:

If you want 1st appt from the earliest and 2ndappt from the latest :

Select Name, PlanID, ApptDate,
       MAX(CASE WHEN rnk = 1 then 1stAppt end) as 1stAppt,
       MAX(CASE WHEN rn = 1 THEN 2ndappt end) as 2ndAppt
from (Select *,
            row_number() over (partition by PlanID Order BY AddedonDate desc) as rn,
     0 as rnk
     from  vClientInfo
     Union All
     Select *,
           0 as rn,
           row_number() over (partition by PlanID Order BY AddedonDate asc) as rnk 
     from  vClientInfo)
GROUP BY Name,PlanID,ApptDate

这篇关于需要SQL Server查询协助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！