Refining SQL Replacement for Oracle's WM_CONCAT Function (Pivoting Rows to column)

Question

Greetings,
I am moving a query from Oracle SQL to Microsoft SQL and am having some difficulty with the transition. Part of my code (listed here) used to pivot the data and display RESULT_VALUE 
by the CPI_SEQ . In Oracle this was easy to do by using the function: WM_CONCAT.
I have been ‘Googling’ for the non-Oracle way to achieve this and stumbled on to the “STUFF((SELECT…)”
 method. This almost works for me except that it places everything into one cell, whereas
 I need to concatenate  RESULT_VALUE by CPI_SEQ (the CPI_SEQ is the unique ID). The current code gives me this:
RESULTS
Anxiety, Depression, Diabetes, 
COPD, ARDS
Whereas I want  my code to present it this way instead
CPI_SEQ                              
RESULTS
22                                          
Anxiety, Depression
44                                          
Diabetes
46                                          
COPD, ARDS
SELECT (STUFF((SELECT ',' + RESULT_VALUE
FROM
SELECT DISTINCT
C1.CPI_Seq AS CPI_SEQ,C1.Result_Value AS RESULT_VALUE
,count(FCurrent.Field_Name) "Current"
,count(FPast.Field_Name) "Past"
,Count(*) Count
,CASE
When count(FCurrent.Field_Name) > 0 and count(FPast.Field_Name) >0 and count(Fcurrent.Field_Name)+ count(FPast.Field_Name) = Count(C1.PF_RESULT_SEQ) Then CONCAT(C1.Result_Value,'(C/P)')
When count(FCurrent.Field_Name) > 0 and count(FPast.Field_Name) =0 and count(Fcurrent.Field_Name)+ count(FPast.Field_Name) = Count(C1.PF_RESULT_SEQ) Then CONCAT (C1.Result_Value,'(C)')
When count(FCurrent.Field_Name) = 0 and count(FPast.Field_Name) >0 and count(Fcurrent.Field_Name)+ count(FPast.Field_Name) = Count(C1.PF_RESULT_SEQ) Then CONCAT(C1.Result_Value,'(P)')
End Result
From
[AnalyticsDW].[dbo].[rr_stag_PF_Results] A1
join [AnalyticsDW].[dbo].[rr_stag_PF_Results] C1 on A1.PF_Result_Seq = C1.PF_Result_Seq
join [AnalyticsDW].[dbo].[rr_PF_Fields_Dept] FCurrent on A1.PF_Result_Seq = FCurrent.RES_SEQ
AND FCurrent.Field_Name = 'Current'
AND A1.Label_Seq in('187582', '187576','187612','187600','187618','187612')
left join [AnalyticsDW].[dbo].rr_PF_Fields_Dept FPast on A1.PF_Result_Seq = FPast.Res_Seq
AND FPast.Field_Name = 'Past'
AND A1.Label_Seq in('187583', '187577','187613','187601','187619','187613')
WHERE
A1.Result_Value ='Yes'
AND (C1.Result_Value in ('Dyspnea', 'Confusion','Pressure Ulcer', 'Stasis Ulcer','Depression','Anxiety'))
GROUP BY
C1.CPI_Seq,
C1.Result_Value
) L1
FOR XML PATH('')),1,2,'')) RESULTS

Answer 1

;With CTE
AS
SELECT
C1.CPI_Seq AS CPI_SEQ,
C1.Result_Value AS RESULT_VALUE
From
[AnalyticsDW].[dbo].[rr_stag_PF_Results] A1
join [AnalyticsDW].[dbo].[rr_stag_PF_Results] C1 on A1.PF_Result_Seq = C1.PF_Result_Seq
join [AnalyticsDW].[dbo].[rr_PF_Fields_Dept] FCurrent on A1.PF_Result_Seq = FCurrent.RES_SEQ
AND FCurrent.Field_Name = 'Current'
AND A1.Label_Seq in('187582', '187576','187612','187600','187618','187612')
left join [AnalyticsDW].[dbo].rr_PF_Fields_Dept FPast on A1.PF_Result_Seq = FPast.Res_Seq
AND FPast.Field_Name = 'Past'
AND A1.Label_Seq in('187583', '187577','187613','187601','187619','187613')
WHERE
A1.Result_Value ='Yes'
AND (C1.Result_Value in ('Dyspnea', 'Confusion','Pressure Ulcer', 'Stasis Ulcer','Depression','Anxiety'))
GROUP BY
C1.CPI_Seq,
C1.Result_Value
SELECT CPI_Seq,
STUFF((SELECT ',' + Result_Value
FROM CTE
WHERE CPI_Seq = c.CPI_Seq
FOR XML PATH(''),TYPE).value('.','varchar(max)'),1,1,'') AS Results
FROM (SELECT DISTINCT CPI_Seq FROM CTE)c
Please Mark This As Answer if it helps to solve the issue Visakh ---------------------------- http://visakhm.blogspot.com/ https://www.facebook.com/VmBlogs