Question

我希望你能帮助/指导我，我有一个工作CTE，我想再添加两个SQL查询来获得一个SQL语句，我遇到各种类型的错误，因为我尝试使用这些参数。 ..

我在CTE方面的经验较少，所以请耐心等待......我的目的是使用报告生成器创建SSRS报告，并且我希望将每个查询的输出合并到一个列栏...

这是工作CTE，这是我在SSRS报告中的第一个栏栏......

WITH Cnt AS (
select Count(Distinct UserID) as Entitled_Users, DATEFROMPARTS(YEAR(t.WhenAddedToGroup),MONTH(t.WhenAddedToGroup),1) as When_Added_To_Group
from Membership t 
where FirstName not like '%test%' and LastName not like '%test%' and FirstName not like '%user%' and LastName not like '%user%' and Account_Disabled not like 'YES' and Obj_Type not like 'NON_USER' and Region not like 'EMEA' and Region not like 'CCLA' and SecGroup IN ('SecurityGroup1', 'SecurityGroup2', 'SecurityGroup3', 'SecurityGroup4')and curr_member like 'yes' and ATTUID is not null and WhenAddedToGroup is not null
Group By DATEFROMPARTS(YEAR(t.WhenAddedToGroup),MONTH(t.WhenAddedToGroup),1)
)
Select When_Added_To_Group, Entitled_Users, (Select SUM(t2.Entitled_Users) as Entitled_Users
from Cnt T2
where T2.When_Added_To_Group <=T1.When_Added_To_Group) as Running_Total
from Cnt T1

下面是第二个SQL查询，我希望我可以合并或加入上面的CTE，这将是我在SSRS中的Column Report的第二个栏栏。

SELECT c.EventType, DATEFROMPARTS(YEAR(c.Event_Date),MONTH(c.Event_Date),1) as Concurrent_Date, MAX(c.MAX_Concurrent_Users) as Peak_Concurrent_Users, c.Hub
FROM vNon_Concurrent_Users c
where c.EventType like 'Broker_Daily_Max_Users' and c.Hub like 'TOK Hub'
group by DATEFROMPARTS(YEAR(c.Event_Date),MONTH(c.Event_Date),1), c.Hub, c.EventType

下面是我的第三个SQL查询，它将是我的SSRS报告的第三个栏栏...

select hs.hub, hs.NAME, DATEFROMPARTS(YEAR(hs.BOOT_TIME),MONTH(hs.BOOT_TIME),1) as Host_Boot_Time,
ROUND(CAST(hs.CPU_CORE_COUNT as FLOAT)*hs.CPU_Hz*count(cast(hs.HOSTID as BIGINT))/800000000,0) as Host_Capacity
from HVD_VPXV_HOSTS as hs WITH (NOLOCK,NOWAIT)
where hs.hub like 'TOK Hub'
group by hs.hub, hs.CPU_CORE_COUNT, hs.CPU_Hz, hs.NAME, hs.BOOT_TIME
order by hs.BOOT_TIME desc

三个查询的where子句是：

When_Added_To_Group=Concurrent_date, 
When_Added_To_Group=Host_Boot_time

如上所述，我试着玩这些参数，但我得到了不同的错误......下面是我试过的一个（很多）陈述，我可以举一个例子：

WITH Cnt AS (
select Count(Distinct UserID) as Entitled_Users, DATEFROMPARTS(YEAR(t.WhenAddedToGroup),MONTH(t.WhenAddedToGroup),1) as When_Added_To_Group
from HVDMembership t 
where FirstName not like '%test%' and LastName not like '%test%' and FirstName not like '%user%' and LastName not like '%user%' and Account_Disabled not like 'YES' and Obj_Type not like 'NON_USER' and Region not like 'EMEA' and Region not like 'CCLA' and SecGroup IN ('SecurityGroup1', 'SecurityGroup2', 'SecurityGroup3', 'SecurityGroup4')and curr_member like 'yes' and ATTUID is not null and WhenAddedToGroup is not null
Group By DATEFROMPARTS(YEAR(t.WhenAddedToGroup),MONTH(t.WhenAddedToGroup),1)
)
, Concurrent
as (
SELECT c.EventType, DATEFROMPARTS(YEAR(c.Event_Date),MONTH(c.Event_Date),1) as Concurrent_Date, MAX(c.MAX_Concurrent_Users) as Peak_Concurrent_Users, c.Hub
FROM vNon_Concurrent_Users c
where c.EventType like 'Broker_Daily_Max_Users' and c.Hub like 'TOK Hub'
group by DATEFROMPARTS(YEAR(c.Event_Date),MONTH(c.Event_Date),1), c.Hub, c.EventType
)
, Capacity
as (
select hs.hub, hs.NAME, DATEFROMPARTS(YEAR(hs.BOOT_TIME),MONTH(hs.BOOT_TIME),1) as Host_Boot_Time,
ROUND(CAST(hs.CPU_CORE_COUNT as FLOAT)*hs.CPU_Hz*count(cast(hs.HOSTID as BIGINT))/800000000,0) as Host_Capacity
from HVD_VPXV_HOSTS as hs WITH (NOLOCK,NOWAIT)
where hs.hub like 'TOK Hub'
group by hs.hub, hs.CPU_CORE_COUNT, hs.CPU_Hz, hs.NAME, hs.BOOT_TIME
)
Select When_Added_To_Group, Entitled_Users, (Select SUM(t2.Entitled_Users) as Entitled_Users
from Cnt T2
where T2.When_Added_To_Group <=T1.When_Added_To_Group) as Running_Total,
Select Peak_concurrent_users, concurrent_date from Concurrent,
select SUM(Host_Capacity), Host_Boot_Time from Capacity
where When_Added_To_Group=Concurrent_date, When_Added_To_Group=Host_Boot_time
from Cnt T1

提前致谢。

Answer 1

CTE必须仅在CTE之后的第一个SQL语句中引用。你有多个选择。这是你的第一个问题。

第二个问题是您的CTE似乎没有彼此相关的数据。如果不知道您对样本数据的结果会有什么期望，那么建议您如何进行查询几乎是不可能的。我无法从你想要做的事情中做出任何意义，所以我不知道你想写什么查询。我可以随机建议3-4种不同的替代查询，但这对您来说可能没有用处，而是要求您提供有关您正在尝试的内容以及业务规则应该是什么以及样本数据的信息。如果不理解数据的基本含义，就无法有效地编写查询。

如何创建或组合三个CTE

1 个答案: