I have a table like this:

Column = NAME
Column = DATE

NAME  | DATE       | Y/N
John  | 01/01/2012 | bit
Mary  | 01/01/2012 | bit
James | 01/01/2012 | bit
John  | 01/02/2012 | bit
Mary  | 01/02/2012 | bit
James | 01/02/2012 | bit
John  | 01/03/2012 | bit
Mary  | 01/03/2012 | bit
James | 01/03/2012 | bit

I want to create some form of matrix or pivot so I end up with this:

NAME  | 01/01/2012 | 01/02/2012 | 01/03/2012
John  | bit        | bit        | bit
Mary  | bit        | bit        | bit
James | bit        | bit        | bit

I have seen some pivot examples that have a small amount of column items (like Banana, Apple, Orange) I need to have an indeterminate number of names and an indeterminate number of dates (so, no hard-coded column names). I was thinking of splitting into multiple tables but I will always need to dynamically create either date columns or name columns.

Can anyone help?

• sql
• pivot

Maybe something like this:

Test data

CREATE TABLE Table1
(
    NAME VARCHAR(100),
    [DATE] DATE,
    [Y/N] BIT
)

INSERT INTO Table1
VALUES
    ('John','01/01/2012',1),
    ('Mary','01/01/2012',0),
    ('James','01/01/2012',1),
    ('John','01/02/2012',0),
    ('Mary','01/02/2012',1),
    ('James','01/02/2012',1),
    ('John','01/03/2012',1),
    ('Mary','01/03/2012',0),
    ('James','01/03/2012',0)

Finding unique columns

DECLARE @cols VARCHAR(MAX)
;WITH CTE
AS
(
    SELECT
        ROW_NUMBER() OVER(PARTITION BY [DATE] ORDER BY [DATE]) AS RowNbr,
        convert(varchar, [DATE], 103) AS [Date]
    FROM
        Table1
)
SELECT @cols=STUFF
(
    (
        SELECT
            ',' +QUOTENAME([Date])
        FROM
            CTE
        WHERE
            CTE.RowNbr=1
    FOR XML PATH('')
    )
,1,1,'')

Declare and executing dynamic sql

DECLARE @query NVARCHAR(4000)=
N'SELECT
    *
FROM
(
    SELECT
        Table1.NAME,
        CAST(Table1.[Y/N] AS INT) AS [Y/N],
        convert(varchar, Table1.[DATE], 103) AS [Date]
    FROM
        Table1
) AS p
PIVOT
(
    MAX([Y/N])
    FOR [Date] IN ('+@cols+')
) AS pvt'

EXECUTE(@query)

Cleaning up after myself

DROP TABLE Table1

Result

Name        01/01/2012    02/01/2012     03/01/2012
James       1             1              0
John        1             0              1
Mary        0             1              0

If you have just a few columns you can use a static pivot:

create table #temp
(
    name varchar(50),
    date datetime,
    yesno bit
)

insert into #temp values('John', '01/01/2012', 1)
insert into #temp values('Mary', '01/01/2012', 1)
insert into #temp values('James', '01/01/2012', 1)
insert into #temp values('John', '01/02/2012', 0)
insert into #temp values('Mary', '01/02/2012', 0)
insert into #temp values('James', '01/02/2012', 0)
insert into #temp values('John', '01/03/2012', 1)
insert into #temp values('Mary', '01/03/2012', 0)
insert into #temp values('James', '01/03/2012', 1)

select name, [01/01/2012], [01/02/2012], [01/03/2012]
from 
(
    select name, date, cast(yesno as tinyint) as yesno
    from #temp
) x
pivot
(
    max(yesno)
    for date in ([01/01/2012], [01/02/2012], [01/03/2012])
) p

drop table #temp

But it sounds like you want a dynamic pivot for this. To do this you must first get the list of columns to pivot and then run your query:

create table test
(
    name varchar(50),
    date datetime,
    yesno bit
)

insert into test values('John', '01/01/2012', 1)
insert into test values('Mary', '01/01/2012', 1)
insert into test values('James', '01/01/2012', 1)
insert into test values('John', '01/02/2012', 0)
insert into test values('Mary', '01/02/2012', 0)
insert into test values('James', '01/02/2012', 0)
insert into test values('John', '01/03/2012', 1)
insert into test values('Mary', '01/03/2012', 0)
insert into test values('James', '01/03/2012', 1)

DECLARE @cols AS VARCHAR(MAX),
    @query  AS VARCHAR(MAX);

SELECT  @cols = STUFF(( SELECT DISTINCT TOP 100 PERCENT
                                '],[' + convert(varchar(10), t2.date, 101)
                        FROM    test AS t2
                        ORDER BY '],[' + convert(varchar(10), t2.date, 101)
                        FOR XML PATH('')
                      ), 1, 2, '') + ']'


set @query = 'select name, ' + @cols + '
            from 
            (
                select name, date, cast(yesno as tinyint) as yesno
                from test
            ) x
            pivot
            (
                max(yesno)
                for date in (' + @cols + ')
            ) p'


execute(@query)

Here are some helpful links on dynamic pivots:

Pivots with Dynamic Columns in SQL Server

Using SQL Server 2005/2008 Pivot on Unknown Number of Columns (Dynamic Pivot)

there are a lot of questions/answers on SO for dynamic pivots.