SQL Server: pivoting without aggregation on a table with two columns - sql

Question

This is a question on a test. I have a table with two columns. I want to pivot on one of them and output the other.
Table structure:
(Name varchar(10), Age int)
I need output with age values as columns and Names listed below each age value.
From searching, I only see examples where there is at least one other column that is used to "group by" for want of a better term. In other words, there is a common factor in each row of the output. My problem does not have this property.
I tried:
SELECT
[agevalue1], [agevalue2], [agevalue3], [agevalue4]
FROM
(SELECT Name, Age FROM MyClass) AS SourceTable
PIVOT
(MAX(Name)
FOR Age IN ([agevalue1], [agevalue2], [agevalue3], [agevalue4])
) AS PivotTable;
I specified agevalue* as a string, i.e. in quotes. I got the column headings alright but a row of NULLS below them.
P.S.: The solution does not need to use pivot but I couldn't think of an alternative approach.
Sample Data:
Name Age
Bob 11
Rick 25
Nina 30
Sam 11
Cora 16
Rachel 25
Desired output:
11 16 25 30
Bob Cora Rick Nina
Sam NULL Rachel NULL

Answer 1

Try this :
with tab as
(
Select 'A' Name, 10 Age union all
Select 'B',11 union all
Select 'c',10 union all
Select 'D',11 union all
Select 'E',11 union all
Select 'F',11
)
select distinct
Age
, stuff((
select ',' + g.Name
from tab g
where g.age = g1.age
order by g.age
for xml path('')
),1,1,'') as Names_With_Same_Age
from tab g1
group by g1.age,Name

Answer 2

To group these together in one row:
11 16 25 30
Bob Cora Rick Nina
and separate them from another set, like:
11 16 25 30
Sam NULL Rachel NULL
they must have something different between each row, since doing a MAX(Name) would get you only one Name for each Age.
This query creates a number that links a particular Age to a row number and then pivots the result. As you said, the PIVOT will group by all columns not referenced in the PIVOT function, so it will group by this row indexer, separating the values like you wanted.
;WITH IndexedClass AS
(
SELECT
M.Name,
M.Age,
-- The ordering will determine which person goes first for each Age
RowIndexer = ROW_NUMBER() OVER (PARTITION BY M.Age ORDER BY M.Name)
FROM
MyClass AS M
)
SELECT
P.[11],
P.[16],
P.[25],
P.[30]
FROM
IndexedClass AS I
PIVOT (
MAX(I.Name) FOR I.Age IN ([11], [16], [25], [30])
) AS P