I have a SQL table with multiple rows that I want into one row with multiple columns - sql

Question

I have a SQL table that has mulitiple rows of data for a user. I want to query that data and return one row for each user. So I want to take the multiple rows and combine them into one row with multiple columns. Is this possible?
Here is what I currently have
UserID Value
8111 396285
8111 812045789854
8111 Secretary
Here is what I am after
UserID Column1 Column2 Column3
8111 396285 812045789854 Secretary

Answer 1

You can use the PIVOT function to get the result. I used the row_number() function to generate the values that will be converted to columns.
If you know how many values you will have ahead of time, then you can hard-code the query:
select userid, Col1, Col2, Col3
from
(
select userid, value,
'Col'+cast(row_number() over(partition by userid
order by (select 1)) as varchar(10)) rn
from yt
) d
pivot
(
max(value)
for rn in (Col1, Col2, Col3)
) piv;
See SQL Fiddle with Demo.
If you have an unknown number of values, then you can use dynamic SQL:
DECLARE #cols AS NVARCHAR(MAX),
#query AS NVARCHAR(MAX)
select #cols = STUFF((SELECT distinct ',' + QUOTENAME('Col'+cast(row_number() over(partition by userid
order by (select 1)) as varchar(10)))
from yt
FOR XML PATH(''), TYPE
).value('.', 'NVARCHAR(MAX)')
,1,1,'')
set #query = 'SELECT userid,' + #cols + '
from
(
select userid, value,
''Col''+cast(row_number() over(partition by userid
order by (select 1)) as varchar(10)) rn
from yt
) x
pivot
(
max(value)
for rn in (' + #cols + ')
) p '
execute(#query);
See SQL Fiddle with Demo. Both give the result:
| USERID | COL1 | COL2 | COL3 |
----------------------------------------------
| 8111 | 396285 | 812045789854 | Secretary |

Answer 2

If you cannot or don't want to use PIVOT/UNPIVOT another option would be to join the columns one by one to the users:
DECLARE #Users TABLE(
UserID int NOT NULL
)
INSERT INTO #Users (UserID) VALUES (1)
INSERT INTO #Users (UserID) VALUES (2)
INSERT INTO #Users (UserID) VALUES (3)
DECLARE #AnyTable TABLE(
UserID int NOT NULL,
FieldNo int NOT NULL,
Value varchar(50) NULL
)
INSERT INTO #AnyTable (UserID, FieldNo, Value) VALUES (1, 1, 'abc')
INSERT INTO #AnyTable (UserID, FieldNo, Value) VALUES (1, 2, 'def')
INSERT INTO #AnyTable (UserID, FieldNo, Value) VALUES (1, 3, 'ghi')
INSERT INTO #AnyTable (UserID, FieldNo, Value) VALUES (2, 1, '123')
INSERT INTO #AnyTable (UserID, FieldNo, Value) VALUES (2, 3, '789')
SELECT u.UserID,
col1.Value as Column1,
col2.Value as Column2,
col3.Value as Column3
FROM #Users u
LEFT JOIN #AnyTable col1
ON col1.UserID = u.UserID
AND col1.FieldNo = 1
LEFT JOIN #AnyTable col2
ON col2.UserID = u.UserID
AND col2.FieldNo = 2
LEFT JOIN #AnyTable col3
ON col3.UserID = u.UserID
AND col3.FieldNo = 3
The result would be:
UserID Column1 Column2 Column3
1 abc def ghi
2 123 NULL 789
3 NULL NULL NULL