SQL Server : 'Not valid identifier' when using PIVOT - sql

Question

I have a temp table #temproles with data like this:
ID
Dept
RoleCode
RoleName
NameOf
123456
65576
1
CLP
Frank
123456
65576
2
SUH
Susan
234567
65578
2
SUH
Susan
234567
65576
13
CLH
Alison
I am trying to use PIVOT to return a dataset like this
ID
Dept
CLP
SUH
CLH
123456
65576
Frank
Susan
NULL
234567
65578
NULL
Susan
Alison
This is what I have so far:
DECLARE #DynamicPivotQuery AS NVARCHAR(MAX)
DECLARE #ColumnName AS NVARCHAR(MAX)
SELECT
#ColumnName = STUFF((SELECT DISTINCT ',' + QUOTENAME(rolename, '''')
FROM #temproles
GROUP BY id, rolename
FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'),1,1,'')
SET #DynamicPivotQuery =
N'SELECT id, dept,' + #ColumnName + N' FROM
(
SELECT id, dept, rolename, nameOf
FROM #temproles
) x
PIVOT
(
ISNULL(nameOf, '''')
FOR rolename IN (' + #ColumnName + N')
) p'
EXEC sp_executesql #DynamicPivotQuery
This returns the following error:
Msg 102, Level 15, State 1, Line 9
Incorrect syntax near 'BDA'.
I previously was using EXEC #DynamicPivotQuery but got an error where the:
name of N'SELECT id, dept, etc.. was not a valid identifier.
My understanding is this was not correct so I changed it to EXEC sp_executesql #DynamicPivotQuery. Any help would be greatly appreciated!
For reference
SELECT STUFF((SELECT DISTINCT ',' + QUOTENAME(cdservicingrolecode, '''')
FROM #JLtemproles
GROUP BY uniqpolicy, cdservicingrolecode
FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'),1,1,'')
returns
'BDA','BDP','CMW','BRC','CXA','CXH','BRP','SUH','CXP','BDH','CLP','BDL','PLH','CLH','SUP','PLP','BIP','BDS','CMP','CLA'

Answer 1

I was able to fix the query. The PIVOT needs an aggregate function, so ISNULL() will not work. I replaced it with MIN(). The role names also need to be enclosed in brackets rather than quotes. Here is the code along with my temp table test.
DROP TABLE IF EXISTS #temproles
CREATE TABLE #temproles (ID INT, Dept INT, RoleCode INT, RoleName VARCHAR(3), NameOf VARCHAR(10))
INSERT INTO #temproles
SELECT 123456, 66576, 1, 'CLP', 'Frank'
UNION ALL SELECT 123456, 66576, 2, 'SUH', 'Susan'
UNION ALL SELECT 234567, 65578, 2, 'SUH', 'Susan'
UNION ALL SELECT 234567, 65576, 13, 'CLH', 'Alison'
DECLARE #DynamicPivotQuery AS NVARCHAR(MAX)
DECLARE #ColumnName AS NVARCHAR(MAX)
SELECT #ColumnName = STUFF((SELECT DISTINCT ',[' + rolename + ']'
FROM #temproles
GROUP BY id, rolename
FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'),1,1,'')
SELECT #ColumnName
SET #DynamicPivotQuery =
N'SELECT id, dept,' + #ColumnName + N' FROM
(
SELECT id, dept, rolename, nameOf
FROM #temproles
) x
PIVOT
(
MIN(nameOf)
FOR rolename IN (' + #ColumnName + N')
) p'
EXEC sp_executesql #DynamicPivotQuery