Find SUM of DATEDIFF on distinct pairs grouped by UserID? - sql

Question

So I have a command that looks like this:
SELECT
UserID,
FacilityMMXID,
ScheduleDate,
StartTime,
EndTime
FROM TblPASchedule
WHERE UserID = 244 AND MONTH(ScheduleDate) = 03 AND Year(ScheduleDate) = 2017
The output looks like this
UserID FacilityMMXID ScheduleDate StartTime EndTime
----------- ------------- ------------ ---------------- ----------------
244 1 2017-03-17 01:00:00 05:00:00
244 2 2017-03-17 01:00:00 05:00:00
244 3 2017-03-17 01:00:00 05:00:00
244 4 2017-03-17 01:00:00 05:00:00
244 5 2017-03-17 01:00:00 05:00:00
244 6 2017-03-17 01:00:00 05:00:00
244 7 2017-03-17 01:00:00 05:00:00
244 8 2017-03-17 01:00:00 05:00:00
244 9 2017-03-17 01:00:00 05:00:00
244 10 2017-03-17 01:00:00 05:00:00
244 11 2017-03-17 01:00:00 05:00:00
244 12 2017-03-17 01:00:00 05:00:00
244 13 2017-03-17 01:00:00 05:00:00
244 14 2017-03-17 01:00:00 05:00:00
244 15 2017-03-17 01:00:00 05:00:00
244 1 2017-03-17 05:00:00 22:00:00
244 2 2017-03-17 05:00:00 22:00:00
244 3 2017-03-17 05:00:00 22:00:00
244 4 2017-03-17 05:00:00 22:00:00
244 5 2017-03-17 05:00:00 22:00:00
244 6 2017-03-17 05:00:00 22:00:00
244 7 2017-03-17 05:00:00 22:00:00
244 8 2017-03-17 05:00:00 22:00:00
244 9 2017-03-17 05:00:00 22:00:00
244 10 2017-03-17 05:00:00 22:00:00
244 11 2017-03-17 05:00:00 22:00:00
244 12 2017-03-17 05:00:00 22:00:00
244 13 2017-03-17 05:00:00 22:00:00
244 14 2017-03-17 05:00:00 22:00:00
244 15 2017-03-17 05:00:00 22:00:00
I left out the ID row as it really isn't important in this case.
Also- yes- I realize that this table is very very redundant- It isn't something I can currently fix as I am not allowed to- I can only work on getting the aforementioned summing function working.
The end goal is to pair off the distinct StartTime and EndTime pairs and then find the date difference of those- and then, for the entire month- find the sum of all the entries.
This is as far as I have gotten:
Using:
SELECT
UserID,
DATEDIFF(HOUR, StartTime, EndTime) AS 'Hours Worked'
FROM TblPASchedule WHERE UserID = 244 AND MONTH(ScheduleDate) = 03 AND Year(ScheduleDate) = 2017
GROUP BY UserId, StartTime, EndTime
I get the output to be:
UserID Hours Worked
----------- ------------
244 4
244 17
But I am not too sure about where I should go from here.
I eventually need to make it group these sums based on the UserIDs, but one step at a time I suppose. I am using a where clause to work with a single id for now...

Answer 1

This query gets all the distinct sets of UserID, Starttime and Endtime
;WITH CTE AS
(SELECT DISTINCT UserID, StartTime, EndTime FROM [dbo].[TblPASchedule])
SELECT SUM(DATEDIFF(MINUTE, StartTime, EndTime))/60.0 AS 'Hours Worked', UserID
FROM CTE GROUP BY UserID
RESULTS look like this
Hours Worked UserID
1.666666 19
1.233333 37
0.500000 38

Answer 2

Have you tried wrapping additional sub query on top of your groups?
SELECT UserId, SUM('Hours Worked') as 'Hours Worked' FROM (
SELECT
UserID,
DATEDIFF(HOUR, StartTime, EndTime) AS 'Hours Worked'
FROM TblPASchedule WHERE UserID = 244 AND MONTH(ScheduleDate) = 03 AND Year(ScheduleDate) = 2017
GROUP BY UserId, StartTime, EndTime
) AS temp
GROUP BY UserId