I have a table that contains data as follows:
========= ============ ========= ==============
| num | module-id | page-no | txt |
========= ============ ========= ==============
| 3177564 | 1 | 1 |text string 1 |
--------- ------------ --------- --------------
| 3177564 | 1 | 2 |text string 2 |
--------- ------------ --------- --------------
| 3177564 | 1 | 3 |text string 3 |
--------- ------------ --------- --------------
| 3177564 | 2 | 1 |text string 1 |
--------- ------------ --------- --------------
| 3177564 | 2 | 2 |text string 2 |
--------- ------------ --------- --------------
| 3177564 | 3 | 1 |text string 1 |
--------- ------------ --------- --------------
| 3177564 | 3 | 2 |text string 2 |
--------- ------------ --------- --------------
| 3177564 | 3 | 3 |text string 3 |
--------- ------------ --------- --------------
I can then use the PIVOT function to produce the following:
========= ============ ============== ============== ==============
| num | module-id | page-no-1 | page-no-2 | page-no-3 |
========= ============ ============== ============== ==============
| 3177564 | 1 |text string 1 |text string 2 |text string 3 |
--------- ------------ -------------- -------------- --------------
| 3177564 | 2 |text string 1 |text string 2 |NULL |
--------- ------------ -------------- -------------- --------------
| 3177564 | 3 |text string 1 |text string2 |text string 3 |
--------- ------------ -------------- -------------- --------------
The code used to produce the pivot is:
SELECT *
FROM
(
SELECT
j.num
,jpad.[module-id]
,jpad.[page-no]
,jpad.txt
FROM
job j
LEFT OUTER JOIN jobpad jpad ON jpad.num = j.num
WHERE
(j.num IN
('3177564'))
AND (jpad.[page-no] IN (1,2,3,4,5,6,7,8,9,10)) a
PIVOT
(
MAX(a.txt) FOR a.[page-no] IN ([1],[2],[3],[4],[5],[6],[7],[8],[9],[10])
) p
ORDER BY num
This is great however, what I would like to do is concatenate the data as per the following example if at all possible:
Please note that p-1, p-2 etc. refers to the page numbers.
========= ========================================================= ============= =============
| num | module-id-1 | module-id-2 | module-id-3 |
========= ========================================================= ============= =============
| 3177564 |p-1 text string 1 & p-2 text string 2 & p-3 text string 3|per module 1 |per module 1 |
--------- --------------------------------------------------------- ------------- -------------
I've been pondering this for a while now but can't quite get it right in my head.
I am currently on SQL Server 2014.
CodePudding user response:
To get the first column you're after:
; -- see sqlblog.org/cte
WITH src AS
(
SELECT num, [module-id], txt = STUFF(
(SELECT CONCAT(' & ', 'p-', [page-no], ' ', txt)
FROM dbo.StuffAndThings WHERE num = s.num
AND [module-id] = s.[module-id]
FOR XML PATH(''),
TYPE).value(N'./text()[1]', N'varchar(max)'), 1, 3, '')
FROM dbo.StuffAndThings AS s
WHERE [page-no] = 1
)
SELECT num,
[module-id-1] = [1],
[module-id-2] = [2],
[module-id-3] = [3]
FROM src PIVOT
(MAX(txt) FOR [module-id] IN ([1],[2],[3])) AS p;
- Example db<>fiddle