Something like this

DECLARE@startdate DATE

DECLARE@enddate DATE

SET@startdate = '20130701'

SET@enddate = '20130703'

SELECTT.ITEM, DATEADD(DAY, D.number, @startdate) AS Dt, SUM(T.QTY) AS QTY

FROM@t AS T

CROSS APPLY(

SELECT*

FROMmaster.dbo.spt_values AS sv -- You can use a Tally table instead of this

WHEREsv.type = 'P' AND sv.number <= DATEDIFF(DAY, @startdate, @enddate)

AND(

( DATEADD(DAY, sv.number, @startdate) >= T.CREATED AND T.ENDS IS NULL ) OR

( DATEADD(DAY, sv.number, @startdate) >= T.CREATED AND DATEADD(DAY, sv.number, @startdate) < T.ENDS )

)

) AS D

GROUP BY T.ITEM, DATEADD(DAY, D.number, @startdate)

ORDER BY Dt, ITEM

You can use Tally tables instead of the spt_values tables that I have used

Please check the link below for more information on Tally Tables

http://www.sqlservercentral.com/scripts/Advanced+SQL/62486/

I think the results that you have given are incorrect

There are 2 rows for ITEM5 in your sample data, one of them gets discontinued on 20130630

INSERT INTO @t (ITEM, QTY, CREATED, ENDS, USERID) SELECT 'ITEM 5',1,'28-05-2013 13:17:00',NULL,2

INSERT INTO @t (ITEM, QTY, CREATED, ENDS, USERID) SELECT 'ITEM 5',1,'28-05-2013 10:10:00','30-06-2013 11:45:00',4

So, its incorrect when you said that it has been discontinued completely.