Select Statement to show missing records (Easy Question) - sql-server

Question

I need some T-SQL that will show missing records.
Here is some sample data:
Emp 1
01/01/2010
02/01/2010
04/01/2010
06/01/2010
Emp 2
02/01/2010
04/01/2010
05/01/2010
etc...
I need to know
Emp 1 is missing
03/01/2010
05/01/2010
Emp 2 is missing
01/01/2010
03/01/2010
06/01/2010
The range to check will start with todays date and go back 6 months.
In this example, lets say today's date is 06/12/2010 so the range is going to be 01/01/2010 thru 06/01/2010.
The day is always going to be the 1st in the data.
Thanks a bunch. :)
Gerhard Weiss
Secretary of Great Lakes Area .NET Users Group
GANG Upcoming Meetings | GANG LinkedIn Group

Answer 1

Try This:
DECLARE #Employees table (DateOf datetime, EmployeeID int)
INSERT #Employees VALUES ('01/01/2010',1)
INSERT #Employees VALUES ('02/01/2010',1)
INSERT #Employees VALUES ('04/01/2010',1)
INSERT #Employees VALUES ('06/01/2010',1)
INSERT #Employees VALUES ('02/01/2010',2)
INSERT #Employees VALUES ('04/01/2010',2)
INSERT #Employees VALUES ('05/01/2010',2)
--I was unsure of the data in the question
--this gives first day of each month for last six months
DECLARE #StartDate datetime
,#EndDate datetime
SELECT #StartDate=DATEADD(month,-6,DATEADD(month,DATEDIFF(month,0,GETDATE()),0) )
,#EndDate=GETDATE()
;with AllDates AS
(
SELECT #StartDate AS DateOf
UNION ALL
SELECT DateAdd(month,1,DateOf)
FROM AllDates
WHERE DateOf<#EndDate
)
SELECT
dt.DateOf,dt.EmployeeID
FROM (SELECT DISTINCT
a.DateOf,e.EmployeeID
FROM AllDates a
CROSS JOIN (SELECT DISTINCT EmployeeID FROM #Employees) e
) dt
LEFT OUTER JOIN #Employees ee ON dt.EmployeeID=ee.EmployeeID AND dt.DateOf=ee.DateOf
WHERE ee.EmployeeID IS NULL
ORDER BY dt.EmployeeID,dt.DateOf
OUTPUT:
DateOf EmployeeID
----------------------- -----------
2009-10-01 00:00:00.000 1
2009-11-01 00:00:00.000 1
2009-12-01 00:00:00.000 1
2010-03-01 00:00:00.000 1
2010-05-01 00:00:00.000 1
2009-10-01 00:00:00.000 2
2009-11-01 00:00:00.000 2
2009-12-01 00:00:00.000 2
2010-01-01 00:00:00.000 2
2010-03-01 00:00:00.000 2
(10 row(s) affected)
this will do every day for last six months, just incorporate this in the above if that is what you want:
DECLARE #StartDate datetime
,#EndDate datetime
SELECT #StartDate=DATEADD(month,-6,GETDATE())
,#EndDate=GETDATE()
;with AllDates AS
(
SELECT #StartDate AS DateOf
UNION ALL
SELECT DateOf+1
FROM AllDates
WHERE DateOf<#EndDate
)
SELECT * FROM AllDates
--OPTION (MAXRECURSION 500) --uncomment and increase if the date range needs more rows

Answer 2

fill a temp table with the date ranges and outer join the temp table to your Emp* table and only return records from your temp table that have null in the corresponding row of the Emp* table

Answer 3

If you're only going back a fixed # of months, you can precalc those "first of month" dates and left join to your employee data:
SELECT d.DT, CASE WHEN e.DT IS NULL THEN 1 ELSE 0 END AS IsMissing
FROM (
SELECT DATEADD(m, DATEDIFF(m, 0, CURRENT_TIMESTAMP), 0) AS DT
UNION
SELECT DATEADD(m, DATEDIFF(m, 0, CURRENT_TIMESTAMP) - 1, 0)
UNION
SELECT DATEADD(m, DATEDIFF(m, 0, CURRENT_TIMESTAMP) - 2, 0)
UNION
SELECT DATEADD(m, DATEDIFF(m, 0, CURRENT_TIMESTAMP) - 3, 0)
UNION
SELECT DATEADD(m, DATEDIFF(m, 0, CURRENT_TIMESTAMP) - 4, 0)
UNION
SELECT DATEADD(m, DATEDIFF(m, 0, CURRENT_TIMESTAMP) - 5, 0)
) AS d
LEFT JOIN EmployeeDates e ON d.DT = e.DT AND e.EmpID = 1