SQL逗号分隔列表作为单个字符串

Question

在我们的业务中，我们有一个主要基本帐户，然后在该基本帐户下有一个附属帐户。
1.）如何将所有帐户（包括基本帐户（以逗号分隔））放入一个列中？

我以前在其他数据集上使用过此代码，并且效果很好。 我只是想不通如何使用所有多个联接来实现此目的。

    SELECT DISTINCT
    A.acctnbr as baseacctnbr, 
        STUFF((SELECT ', '+c1.ACCTNBR
FROM [USBI].[vw_FirmAccount] a1
    inner join [USBI].[vw_RelatedAccount] b1 on a1.firmaccountid = b1.basefirmaccountid 
    inner join [USBI].[vw_FirmAccount] a2 on a2.firmaccountid = b1.relatedfirmaccountid
    inner join USBI.vw_NameAddressBase c1 on b1.relatedfirmaccountid = c1.firmaccountid
        where c1.AcctNbr = c.ACCTNBR
        FOR XML PATH ('')),1,1, '') AS ALLACCTS

FROM [USBI].[vw_FirmAccount] a
    inner join [USBI].[vw_RelatedAccount] b on a.firmaccountid = b.basefirmaccountid 
    inner join [USBI].[vw_FirmAccount] a1 on a1.firmaccountid = b.relatedfirmaccountid
    inner join USBI.vw_NameAddressBase c on b.relatedfirmaccountid = c.firmaccountid

where a.acctnbr = '11727765'
  and c.restrdate <> '99999999'
  and c.closerestrictind <> 'c'
  and c.iscurrent = '1'
  and b.iscurrent = '1'

我的输出：

我希望看到这样的逗号分隔列表： 11727765, 11727799, 11783396, 12192670

我已经解决了将数据添加到单个字符串中的所有其他问题，但在这里找不到解决方案。 不能重复。

Answer 1

您可以右键单击左上角的框（baseacctnbr的左侧和上方的1），然后将结果另存为csv文件，然后用记事本打开它以这种方式查看。

就个人而言，我经常在这种类型的事情上使用python。 使用pyodbc（ https://code.google.com/archive/p/pyodbc/wikis/GettingStarted.wiki ），您可以执行查询并直接访问数据。 对于以下代码，您要做的就是在连接字符串中键入服务器和数据库名称

import pyodbc

connection = pyodbc.connect('DRIVER={SQL Server Native Client11.0};SERVER=YOURSERVERHERE;DATABASE=YOURDATABASENAMEHERE;TRUSTED_CONNECTION=yes')

cursor = connection.cursor()

query = """
SELECT DISTINCT
A.acctnbr as baseacctnbr, 
    STUFF((SELECT ', '+c1.ACCTNBR
FROM [USBI].[vw_FirmAccount] a1
    inner join [USBI].[vw_RelatedAccount] b1 on a1.firmaccountid = b1.basefirmaccountid 
inner join [USBI].[vw_FirmAccount] a2 on a2.firmaccountid = b1.relatedfirmaccountid
inner join USBI.vw_NameAddressBase c1 on b1.relatedfirmaccountid = c1.firmaccountid
    where c1.AcctNbr = c.ACCTNBR
    FOR XML PATH ('')),1,1, '') AS ALLACCTS

FROM [USBI].[vw_FirmAccount] a
inner join [USBI].[vw_RelatedAccount] b on a.firmaccountid = b.basefirmaccountid 
inner join [USBI].[vw_FirmAccount] a1 on a1.firmaccountid = b.relatedfirmaccountid
inner join USBI.vw_NameAddressBase c on b.relatedfirmaccountid = c.firmaccountid

where a.acctnbr = '11727765'
  and c.restrdate <> '99999999'
  and c.closerestrictind <> 'c'
  and c.iscurrent = '1'
  and b.iscurrent = '1'
"""
cursor.execute(query)
resultList = [row[0] for row in cursor.fetchall()]
print ", ".join(str(i) for i in resultList)

Answer 2

尝试创建一个标量函数，将所有需要的参数传递给它，以查找下级帐户。 选择@ResultVar = ResultVar +'，'+ ..将运行所有行并增加@ResultVar上的内容

CREATE FUNCTION ALLACCTS
(
    @baseacctnbr as varchar(100) 
)
RETURNS varchar(max)
AS
BEGIN
    -- Declare the return variable here
    DECLARE @ResultVar as varchar(max)
    set @ResultVar  =''
    -- Add the T-SQL statements to compute the return value here
    SELECT @ResultVar =@ResultVar+','+ 
        STUFF((SELECT ', '+c1.ACCTNBR
    FROM [USBI].[vw_FirmAccount] a1
    inner join [USBI].[vw_RelatedAccount] b1 on a1.firmaccountid =     b1.basefirmaccountid 
    inner join [USBI].[vw_FirmAccount] a2 on a2.firmaccountid =     b1.relatedfirmaccountid
    inner join USBI.vw_NameAddressBase c1 on b1.relatedfirmaccountid = c1.firmaccountid
        where c1.AcctNbr = c.ACCTNBR
        FOR XML PATH ('')),1,1, '') AS ALLACCTS

    FROM [USBI].[vw_FirmAccount] a
    inner join [USBI].[vw_RelatedAccount] b on a.firmaccountid = b.basefirmaccountid 
    inner join [USBI].[vw_FirmAccount] a1 on a1.firmaccountid = b.relatedfirmaccountid
    inner join USBI.vw_NameAddressBase c on b.relatedfirmaccountid = c.firmaccountid

    if(len(@ResultVar)>1)
        set @ResultVar = SUBSTRING(@ResultVar,1,len(@ResultVar)-1)
    -- Return the result of the function
    RETURN @ResultVar

END
GO

Answer 3

这是我为在单个字符串或单个列中创建逗号分隔列表的查询而解决的问题。 我必须使用XML路径在CTE中用子字符串编写代码。 它可以满足我的需要，我将在SSRS报告中使用代码。

 ;with t1 as
(
SELECT b.relatedfirmaccountid, b.basefirmaccountid, c.firmaccountid, a.acctnbr as baseacctnbr, a1.acctnbr as relatedacct,
rn = row_number() over (partition by a.acctnbr order by a.acctnbr),
    substring((select ', '+c.acctnbr as 'data()' 
FROM [USBI].[vw_FirmAccount] a
    inner join [USBI].[vw_RelatedAccount] b on a.firmaccountid = b.basefirmaccountid 
    inner join [USBI].[vw_FirmAccount] a1 on a1.firmaccountid = b.relatedfirmaccountid
    inner join USBI.vw_NameAddressBase c on b.relatedfirmaccountid = c.firmaccountid
where a.acctnbr = '11727765'
    and c.restrdate <> '99999999'
    and c.closerestrictind <> 'c'
    and c.iscurrent = '1'
    and b.iscurrent = '1'
for xml path('')),2,255) as "AllRelatedAccts"

FROM [USBI].[vw_FirmAccount] a
    inner join [USBI].[vw_RelatedAccount] b on a.firmaccountid = b.basefirmaccountid 
    inner join [USBI].[vw_FirmAccount] a1 on a1.firmaccountid = b.relatedfirmaccountid
    inner join USBI.vw_NameAddressBase c on b.relatedfirmaccountid = c.firmaccountid
where a.acctnbr = '11727765'
    and c.restrdate <> '99999999'
    and c.closerestrictind <> 'c'
    and c.iscurrent = '1'
    and b.iscurrent = '1'
 ),
 t2 as
 (
select distinct a.acctnbr, a.name
from USBI.vw_NameAddressBase a
where a.acctnbr = '11727765'
    and a.restrdate <> '99999999'
    and a.closerestrictind <> 'c'
    and a.iscurrent = '1'
group by a.acctnbr, a.name
  )
select 
    a.baseacctnbr,
    a.relatedacct,
    a.allrelatedaccts,
    a.rn
from t1 a
inner join t2 b on a.baseacctnbr = b.acctnbr
--where rn = 1

我的结果：

更改rn = 1后的结果