Determine time elapased by hour

Question

Determine time elapased by hour

Matthew24

SSCommitted

Points: 1736
More actions
October 30, 2013 at 4:24 am

#395745

Hi,
Wondering if anyone has an elegant solution to finding out how much time has elapsed for a specific hour.
declare @t as table (ptr int identity, START time, ENDD time)
insert into @t (START, ENDD)
SELECT '09:30', '10:30'
union all
SELECT '10:00', '11:00'
union all
SELECT '10:30', '11:30'
Taking the example table above, I am looking to output the amount of time elapsed for each hour block. So:
09 - 0.5
10 - 2
11 - 0.5
I can think of complicated and ugly ways using cursors and many lines of code, but I thought someone might have a more user friendly solution!
Many thanks for your time.
Matt
Forum Etiquette: How to post Reporting Services problems

[/url]

Forum Etiquette: How to post data/code on a forum to get the best help - by Jeff Moden
[/url]

How to Post Performance Problems - by Gail Shaw
[/url]

Sean Lange SSC Guru Points: 286590 More actions · Answer 1

m.dunster (10/30/2013)
Hi,
Wondering if anyone has an elegant solution to finding out how much time has elapsed for a specific hour.
declare @t as table (ptr int identity, START time, ENDD time)
insert into @t (START, ENDD)
SELECT '09:30', '10:30'
union all
SELECT '10:00', '11:00'
union all
SELECT '10:30', '11:30'
Taking the example table above, I am looking to output the amount of time elapsed for each hour block. So:
09 - 0.5
10 - 2
11 - 0.5
I can think of complicated and ugly ways using cursors and many lines of code, but I thought someone might have a more user friendly solution!
Many thanks for your time.
Matt

Thank you for providing ddl and sample data. I am confused by your expected output. Can you try to explain what you want? The business rules for the output is lost on me.

_______________________________________________________________

Need help? Help us help you.

Read the article at http://www.sqlservercentral.com/articles/Best+Practices/61537/ for best practices on asking questions.

Need to split a string? Try Jeff Modens splitter http://www.sqlservercentral.com/articles/Tally+Table/72993/.

Cross Tabs and Pivots, Part 1 – Converting Rows to Columns - http://www.sqlservercentral.com/articles/T-SQL/63681/
Cross Tabs and Pivots, Part 2 - Dynamic Cross Tabs - http://www.sqlservercentral.com/articles/Crosstab/65048/
Understanding and Using APPLY (Part 1) - http://www.sqlservercentral.com/articles/APPLY/69953/
Understanding and Using APPLY (Part 2) - http://www.sqlservercentral.com/articles/APPLY/69954/

Matthew24 SSCommitted Points: 1736 More actions · Answer 2

Hi,

To explain further. Let us assume that table @t represents light bulbs being switched on and off. So the first light bulb is switched on at 9.30am and off at 10.30am, the second on at 10.00am and off at 11.00am, the third on at 10.30 am and off at 11.30am.

I need to know how long, within hour blocks, the total time expended for the sum of all light bulbs being on.

So in our example:

Between 9am to 10am, light bulb 1 was on for 0.5 hours, light bulb 2&3 were both off, so a total of half an hour (0.5)

Between 10am to 11am, light bulb 1 was on for 0.5 hours, light bulb 2 for an hour, and light bulb 3 on for 0.5 hours, so a total of 2 hours (2.0)

Between 11.00am to 12.00pm, light bulb 1 was off, light bulb 2 was off, hour, and light bulb 3 on for 0.5 hours, so a total of half an hour (0.5)

Mark Cowne One Orange Chip Points: 26952 More actions · Answer 3

Not quite a complete solution, it doesn't handle spanning midnight

WITH Hours(hrStart,hrEnd) AS (

SELECT CAST(hrStart AS TIME),CAST(hrEnd AS TIME)

FROM (

VALUES ('00:00','01:00'),('01:00','02:00'),('02:00','03:00'),('03:00','04:00'),

('04:00','05:00'),('05:00','06:00'),('06:00','07:00'),('07:00','08:00'),

('08:00','09:00'),('09:00','10:00'),('10:00','11:00'),('11:00','12:00'),

('12:00','13:00'),('13:00','14:00'),('14:00','15:00'),('15:00','16:00'),

('16:00','17:00'),('17:00','18:00'),('18:00','19:00'),('19:00','20:00'),

('20:00','21:00'),('21:00','22:00'),('22:00','23:00'),('23:00','00:00')) h(hrStart,hrEnd))

SELECT h.hrStart AS [Hour],

SUM(DATEDIFF(minute,CASE WHEN t.START < h.hrStart THEN h.hrStart ELSE t.START END,

CASE WHEN t.ENDD > h.hrEnd THEN h.hrEnd ELSE t.ENDD END)) AS minutes

FROM Hours h

INNER JOIN @t t ON t.ENDD > h.hrStart AND t.START < h.hrEnd

GROUP BY h.hrStart

ORDER BY h.hrStart;

____________________________________________________

Deja View - The strange feeling that somewhere, sometime you've optimised this query before

How to get the best help on a forum

http://www.sqlservercentral.com/articles/Best+Practices/61537

Matthew24 SSCommitted Points: 1736 More actions · Answer 4

Thanks Mark. I'll give it a go. The midnight issue shouldn't be a problem.

Dwain Camps SSC Guru Points: 86948 More actions · Answer 5

Another take (also not handling crossing midnight), with some sample data added.

declare @t as table (ptr int identity, START time, ENDD time)

insert into @t (START, ENDD)

SELECT '09:30', '10:30' union all

SELECT '10:00', '11:00' union all

SELECT '10:30', '11:30' union all

SELECT '09:30', '11:30' union all

SELECT '11:00', '11:30';

WITH Tally (n) AS

(

SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL))

FROM (VALUES (0),(0),(0),(0),(0),(0)) a(a)

CROSS JOIN (VALUES (0),(0),(0),(0)) b(b)

)

SELECT Hour=RIGHT(100+d.n-1,2) + ':00'

,LightBulbsShining=SUM(

CASE WHEN StartHr=c.n AND EndHr=c.n THEN DATEDIFF(minute, START, ENDD)/60.

WHEN StartHr=c.n AND EndHr<>c.n THEN DATEDIFF(minute, START, DATEADD(hour, 1, hr))/60.

ELSE DATEDIFF(minute, hr, ENDD)/60. END)

FROM @t a

CROSS APPLY

(

SELECT StartHr=DATEPART(hour, START), EndHr=DATEPART(hour, ENDD)

) b

CROSS APPLY

(

SELECT n, hr=RIGHT(100+n,2) + ':00'

FROM Tally

WHERE n BETWEEN StartHr AND EndHr

) c

RIGHT JOIN Tally d ON d.n-1 = c.n

GROUP BY RIGHT(100+d.n-1,2) + ':00';

My mantra: No loops! No CURSORs! No RBAR! Hoo-uh![/I]
My thought question: Have you ever been told that your query runs too fast?

My advice:
INDEXing a poor-performing query is like putting sugar on cat food. Yeah, it probably tastes better but are you sure you want to eat it?
The path of least resistance can be a slippery slope. Take care that fixing your fixes of fixes doesn't snowball and end up costing you more than fixing the root cause would have in the first place.

Need to UNPIVOT? Why not CROSS APPLY VALUES instead?[/url]
Since random numbers are too important to be left to chance, let's generate some![/url]
Learn to understand recursive CTEs by example.[/url]
[url url=http://www.sqlservercentral.com/articles/St

Matthew24 SSCommitted Points: 1736 More actions · Answer 6

Hi Mark,

Your solution is good, but it has a limitation in that if a period start and end more than one hour apart, it doesn't get calculated.

Dwain: Yours, I can't really understand (my limitation), but there is something wrong, if you alter the first end time to 13:30, then it counts all the extra hours in the 10:00 slot.

Thanks for your efforts and help. I'll have to keep going until I beat it.

Matt

brad.mason5 Hall of Fame Points: 3299 More actions · Answer 7

Here is my shot at this using Mark-101232 derived table for times.

--use any database which has this numbers table.

declare @t as table (ptr int identity, START time, ENDD time)

insert into @t (START, ENDD)

SELECT '09:30', '10:00'

union all

SELECT '10:00', '11:00'

union all

SELECT '10:30', '11:30'

union all

SELECT '13:30', '15:30'

union all

SELECT '18:30', '19:00'

union all

SELECT '10:00', '10:30'

union all

SELECT '03:00', '08:30'

union all

SELECT '19:00', '19:45'

--uses number table

--http://sqlblog.com/blogs/adam_machanic/archive/2006/07/12/you-require-a-numbers-table.aspx

/**************************************

Most of this code is because of expanding the

times in the start and end. If times span more than

1 hour this will generate a new record for this.

*******************************************/

IF OBJECT_ID('tempdb..#LightHour','u') IS NOT NULL

DROP TABLE #LightHour

;WITH ExpandHours

AS

(

SELECT t.ptr

,e.RowNum

,StartTime = CAST(e.StartTime AS TIME)

,EndTime = CAST(e.EndTime AS TIME)

,OriginalStartTime = t.START

,OriginalEndTime = t.ENDD

FROM @t t

CROSS APPLY (

SELECT RowNum = ROW_NUMBER() OVER (PARTITION BY t.ptr ORDER BY prev.Number)

,DATEADD(hh, prev.Number, dateadd(hour, datediff(hour, 0, t.START), 0)) AS StartTime

,DATEADD(hh, curr.Number, dateadd(hour, datediff(hour, 0, t.START), 0)) AS EndTime

FROM dbo.Numbers curr

JOIN dbo.Numbers prev

ON curr.Number = prev.Number + 1

WHERE curr.Number <= DATEDIFF(hh, t.START, t.ENDD)

) e

),

Final

AS

(

SELECT ptr

,StartTime = CASE WHEN StartTime < OriginalStartTime

THEN OriginalStartTime

ELSE StartTime

END

,EndTime = CASE WHEN OriginalEndTime < EndTime

THEN OriginalEndTime

ELSE EndTime

END

FROM ExpandHours

UNION

SELECT ptr

,StartTime = (SELECT MAX(c.EndTime) FROM ExpandHours c WHERE i.ptr = c.ptr)

,EndTime = OriginalEndTime

FROM ExpandHours i

)

SELECT ptr

,Start = StartTime

,Endd = EndTime

INTO #LightHour

FROM Final

WHERE StartTime <> EndTime

UNION

--if under 1 hour then include here

SELECT *

FROM @t

WHERE DATEDIFF(mi, START, ENDD) < 60

IF OBJECT_ID('tempdb..#Final','u') IS NOT NULL

DROP TABLE #Final

SELECT LighHour =hrStart

,TotalTimeOn_MI = SUM(CASE WHEN t.Start >= hrStart AND t.Start < hrEnd

THEN DATEDIFF(mi, t.Start, t.Endd)

ELSE 0

END)

--,ptr

INTO #Final

FROM #LightHour t

CROSS APPLY

(

SELECT hrStart = CAST(hrStart AS TIME)

,hrEnd = CAST(hrEnd AS TIME)

FROM (

VALUES ('00:00','01:00'),('01:00','02:00'),('02:00','03:00'),('03:00','04:00'),

('04:00','05:00'),('05:00','06:00'),('06:00','07:00'),('07:00','08:00'),

('08:00','09:00'),('09:00','10:00'),('10:00','11:00'),('11:00','12:00'),

('12:00','13:00'),('13:00','14:00'),('14:00','15:00'),('15:00','16:00'),

('16:00','17:00'),('17:00','18:00'),('18:00','19:00'),('19:00','20:00'),

('20:00','21:00'),('21:00','22:00'),('22:00','23:00'),('23:00','24:00')) h(hrStart,hrEnd)

) h

GROUP BY h.hrStart

-- ,ptr

SELECT *

FROM #Final

WHERE TotalTimeOn_MI > 0

mister.magoo SSC-Forever Points: 47174 More actions · Answer 8

Borrowing from Dwain's sample data and Tally, here is my submission:

--== Dwain's setup ==--

declare @t as table (ptr int identity, START time, ENDD time)

insert into @t (START, ENDD)

SELECT '09:30', '10:30' union all

SELECT '10:00', '11:00' union all

SELECT '10:30', '11:30' union all

SELECT '09:30', '11:30' union all

SELECT '11:00', '11:30';

WITH Tally (n) AS

(

SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL))

FROM (VALUES (0),(0),(0),(0),(0),(0)) a(a)

CROSS JOIN (VALUES (0),(0),(0),(0)) b(b)

)

--== And MM's solution ==--

select Tally.n as [Hour],SUM(calc.UpToENDD - calc.UpToSTART) as [TotalMinutes]

from Tally

left outer join @t as source

on Tally.n between datepart(hour,convert(datetime,source.START)) and datepart(hour,convert(datetime,source.ENDD))

outer apply (

select case when source.START is null then 0

when dateadd(hour,Tally.n,0)<convert(datetime,source.START) then datediff(minute,dateadd(hour,Tally.n,0),convert(datetime,source.START))

else 0

end as UpToSTART,

case when source.ENDD is null then 0

when Tally.n=datepart(hour,convert(datetime,source.ENDD)) then datediff(minute,dateadd(hour,Tally.n,0),convert(datetime,source.ENDD))

else 60

end as UpToENDD

) calc

group by Tally.n

order by Tally.n;

MM

select geometry::STGeomFromWKB(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

brad.mason5 Hall of Fame Points: 3299 More actions · Answer 9

Monster Maghoul thank you for sharing! This is awesome now I just need to understand what is going on here. This guy will perform well under load too!

mister.magoo SSC-Forever Points: 47174 More actions · Answer 10

brad.mason5 (11/1/2013)
Monster Maghoul thank you for sharing! This is awesome now I just need to understand what is going on here. This guy will perform well under load too!

You are most welcome.

It's really quite simple in terms of "how it works".

The link from Tally (the hours in a day) to your data is where the "hour of the day (n)" falls between the start and end time in your data.

This gives us every data row for each hour, then it is simply a CASE of working out how much of that hour each start/end pair has consumed and SUMming them up.

MM

select geometry::STGeomFromWKB(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

Matthew24 SSCommitted Points: 1736 More actions · Answer 11

Hi Guys,

Thanks all for your input. I suppose the MM solution is as close as we are going to get to "eloquent", adn certainly beats mine !!

Thanks again.

Matt