I am in a situation where I need to find-out total time spent in office for some internal application.
I have sample data like this:
Id EmployeeId ScanDateTime Status
7 87008 2018-08-02 16:03:00.227 1
8 87008 2018-08-02 16:06:17.277 2
9 87008 2018-08-02 16:10:37.107 3
10 87008 2018-08-02 16:20:17.277 2
11 87008 2018-08-02 16:30:37.107 3
12 87008 2018-08-02 20:06:00.000 4
Here Status have different meanings:
1- Start
2- Pause
3- Resume
4- End
Means Employees start their work at ScanDateTime when status is 1. They can go for break(status 2) and come back and resume their work(Status 3) and with status 4 means they are ending their job.
Note: There could be multiple breaks during work hours.
Expected Output:
EmployeeId StartTime EndTime BreakInMins
87008 2018-08-02 16:03:00.227 2018-08-02 20:06:00.000 14
I have tried to follow some example to calculate the expected result set but not helping.
I could not find any such example where this similar example available.
Any help would be appreciated.
Please try this. Handles multiple breaks/employees and cases, when break is still in progress or session is not finished
select
[EmployeeId] = [s].[EmployeeId]
,[StartTime] = [s].[ScanDateTime]
,[EndTime] = [et].[ScanDateTime]
,[BreakInMins] = [b].[BreakInMins]
from
[Scans] as [s] -- here is your table
outer apply
(
select top 1 [ScanDateTime], [Id] from [Scans] where [Id] > [s].[Id] and [EmployeeId] = [s].[EmployeeId] and [Status] = 4 order by [ScanDateTime] asc
) as [et]
outer apply
(
select
[BreakInMins] = sum(isnull([r].[mins], datediff(mi, [sp].[ScanDateTime], getdate())))
from
[Scans] as [sp]
outer apply
(
select top 1 [mins] = datediff(mi, [sp].[ScanDateTime], [ScanDateTime]) from [Scans] where [Id] > [sp].[Id] and [EmployeeId] = [sp].[EmployeeId] and [Status] IN (3, 4) order by [ScanDateTime] asc
) as [r]
where
[sp].[id] > [s].[id] and [sp].[id] < isnull([et].[id], [id] + 1)
and [sp].[EmployeeId] = [s].[EmployeeId]
and [sp].[Status] = 2
) as [b]
where
[Status] = 1;
Here is test-friendly script: script
i consider multiple breaks per day of employee you can check below i also provided fiddle link
select t1.*,t5.breakmins from
(
select EmployeeId,min(StartTime) as StartTime,max(EndTime) as EndTime from
(
select EmployeeId,(case when status=1 then ScanDateTime end) as StartTime,
(case when status=4 then ScanDateTime end) as EndTime,
case when status=3 then ScanDateTime end as ResumeWork,
case when status=2 then ScanDateTime end as pauseTime
from emp
) as t group by EmployeeId
) t1
inner join
(
select EmployeeId, convert(date,ResumeWork) as day ,
sum(case when status=2 then datediff(minute,ResumeWork,res) end ) as breakmins from
(
select EmployeeId,ResumeWork,status ,
lag(ResumeWork) over(PARTITION BY EmployeeId order by ResumeWork desc) as res from
(
select * from
(
select EmployeeId, case when status=3 then ScanDateTime end as ResumeWork,status from emp
) as t1 where ResumeWork is not null
union all
select * from
(
select EmployeeId,case when status=2 then ScanDateTime end as pauseTime,status from emp
) as t2 where pauseTime is not null
) as t3 group by EmployeeId,ResumeWork,status
) t4 group by EmployeeId, convert(date,ResumeWork)
)t5 on t1.EmployeeId=t5.EmployeeId
and convert(date,t1.StartTime)=t5.day
EmployeeId StartTime EndTime breakmins
87008 2018-08-02T16:03:00.227Z 2018-08-02T20:06:00Z 12
http://sqlfiddle.com/#!18/ae60f/6
You can try this.
make a row_number in CTE by Status, because we need to know which Pause time correspond which Resume time. then self join in the CTE by EmployeeId
CREATE TABLE T(
Id INT,
EmployeeId INT,
ScanDateTime DATETIME,
Status INT
);
INSERT INTO T VALUES (7 ,87008 ,'2018-08-02 16:03:00.227',1);
INSERT INTO T VALUES (8 ,87008 ,'2018-08-02 16:06:17.277',2);
INSERT INTO T VALUES (9 ,87008 ,'2018-08-02 16:10:37.107',3);
INSERT INTO T VALUES (10,87008 ,'2018-08-02 16:20:17.277',2);
INSERT INTO T VALUES (11,87008 ,'2018-08-02 16:30:37.107',3);
INSERT INTO T VALUES (12,87008 ,'2018-08-02 20:06:00.000',4);
Query 1:
;with cte as(
SELECT *,
MIN(ScanDateTime) over(partition by EmployeeId order by EmployeeId) StartTime,
MAX(ScanDateTime) over(partition by EmployeeId order by EmployeeId) EndTime,
ROW_NUMBER() OVER(PARTITION BY Status order by id) rn
FROM t
)
select t1.EmployeeId,
t1.StartTime,
t1.EndTime,
SUM(datediff(minute,t1.ScanDateTime,t2.ScanDateTime)) BreakInMins
from
cte t1
inner join cte t2
on t1.rn =t2.rn and t1.Status = 2 and t2.Status = 3 and t1.EmployeeId = t2.EmployeeId
group by t1.EmployeeId,
t1.StartTime,
t1.EndTime
Results:
| EmployeeId | StartTime | EndTime | BreakInMins |
|------------|----------------------|--------------------------|-------------|
| 87008 | 2018-08-02T20:06:00Z | 2018-08-02T16:03:00.227Z | 14 |
EDIT
you can try this query if there are different day in your data. just group by the date.
CREATE TABLE T(
Id INT,
EmployeeId INT,
ScanDateTime DATETIME,
Status INT
);
INSERT INTO T VALUES (7 ,87008 ,'2018-08-02 16:03:00.227',1);
INSERT INTO T VALUES (8 ,87008 ,'2018-08-02 16:06:17.277',2);
INSERT INTO T VALUES (9 ,87008 ,'2018-08-02 16:10:37.107',3);
INSERT INTO T VALUES (10,87008 ,'2018-08-02 16:20:17.277',2);
INSERT INTO T VALUES (11,87008 ,'2018-08-02 16:30:37.107',3);
INSERT INTO T VALUES (12,87008 ,'2018-08-02 20:06:00.000',4);
INSERT INTO T VALUES (27 ,87008 ,'2018-08-03 16:03:00.227',1);
INSERT INTO T VALUES (28 ,87008 ,'2018-08-03 16:06:17.277',2);
INSERT INTO T VALUES (29 ,87008 ,'2018-08-03 16:11:37.107',3);
INSERT INTO T VALUES (210,87008 ,'2018-08-03 16:20:17.277',2);
INSERT INTO T VALUES (211,87008 ,'2018-08-03 16:30:37.107',3);
INSERT INTO T VALUES (212,87008 ,'2018-08-03 20:06:00.000',4);
Query 1:
;with cte as(
SELECT EmployeeId,
MAX(CASE WHEN Status = 1 then ScanDateTime end) StartTime,
MIN(CASE WHEN Status = 4 then ScanDateTime end) EndTime,
CAST(ScanDateTime as date) dt
FROM t
GROUP BY EmployeeId,CAST(ScanDateTime as date)
)
,cte2 as(
SELECT t2.*,
Row_number() over(partition by t2.EmployeeId,t2.Status order by Id) rn,
t1.StartTime,
t1.EndTime,
t1.dt
FROM cte t1
INNER JOIN T t2 ON t1.EmployeeId = t2.EmployeeId and Status in (2,3) and t1.dt = CAST(t2.ScanDateTime as date)
)
select t1.EmployeeId,
t1.StartTime,
t1.EndTime,
SUM(datediff(minute,t1.ScanDateTime,t2.ScanDateTime)) BreakInMins
from cte2 t1
inner join cte2 t2 on
t1.rn = t2.rn
and
t1.EmployeeId = t2.EmployeeId
and t1.Status = 2 and t2.Status =3
group by t1.EmployeeId,
t1.StartTime,
t1.EndTime
Results:
| EmployeeId | StartTime | EndTime | BreakInMins |
|------------|--------------------------|----------------------|-------------|
| 87008 | 2018-08-02T16:03:00.227Z | 2018-08-02T20:06:00Z | 14 |
| 87008 | 2018-08-03T16:03:00.227Z | 2018-08-03T20:06:00Z | 15 |
Try below query: http://sqlfiddle.com/#!18/6fe11/3
select id,min(case when status=1 then stattime end) as starttime,
min(case when status=4 then stattime end) as endtime,
sum(case when status=2 then minute end) as breakinmin
from
(
select id,stattime,status,
DATEdiff(minute,stattime,lead(stattime,1,NULL)
over (partition by id ORDER BY stattime)) as minute
from ForgeRock)a
group by id
id starttime endtime breakinmin
87008 2018-08-02T16:03:00.227Z 2018-08-02T20:06:00Z 14
Related
There is a table with three columns:
CREATE TABLE #t1 ( Id INT
,VisitDate DATE
,Counter INT)
AND test data:
INSERT INTO #t1 VALUES (1,'2019-01-01', 50)
INSERT INTO #t1 VALUES (2,'2019-01-02', 15)
INSERT INTO #t1 VALUES (3,'2019-01-03', 7)
INSERT INTO #t1 VALUES (4,'2019-01-04', 7)
INSERT INTO #t1 VALUES (5,'2019-01-05', 18)
INSERT INTO #t1 VALUES (6,'2019-01-06', 19)
INSERT INTO #t1 VALUES (7,'2019-01-07', 11)
INSERT INTO #t1 VALUES (8,'2019-01-08', 1)
INSERT INTO #t1 VALUES (9,'2019-01-09', 19)
Need to find three and more consecutive days where Counter more or equal ten:
Id VisitDate Counter
5 2019-01-05 18
6 2019-01-06 19
7 2019-01-07 11
My SELECT statement is
;WITH cte AS
(
SELECT *
,IIF(Counter > 10, 1,0) AS MoreThanTen
FROM #t1
), lag_lead_cte AS
(
SELECT *
,LAG(MoreThanTen) OVER (ORDER BY VisitDate) AS LagShift
,(LAG(MoreThanTen) OVER (ORDER BY VisitDate) + MoreThanTen ) AS LagMoreThanTen
,LEAD(MoreThanTen) OVER (ORDER BY VisitDate) AS LeadShift
,(LEAD(MoreThanTen) OVER (ORDER BY VisitDate) + MoreThanTen ) AS LeadMoreThanTen
FROM cte
)
SELECT *
FROM lag_lead_cte
WHERE LagMoreThanTen = 2 OR LeadMoreThanTen = 2
But the result is not fully consistent
Id VisitDate Counter
1 2019-01-01 50
2 2019-01-02 15
5 2019-01-05 18
6 2019-01-06 19
7 2019-01-07 11
It looks like a gaps-and-islands problem.
Here is one way to do it.
I'm assuming SQL Server based on the T-SQL tag.
Run this query CTE-by-CTE and examine intermediate results to understand how it works.
Query
WITH
CTE_rn
AS
(
SELECT *
,CASE WHEN Counter>10 THEN 1 ELSE 0 END AS MoreThanTen
,ROW_NUMBER() OVER (ORDER BY VisitDate) AS rn1
,ROW_NUMBER() OVER (PARTITION BY CASE WHEN Counter>10 THEN 1 ELSE 0 END ORDER BY VisitDate) AS rn2
FROM #t1
)
,CTE_Groups
AS
(
SELECT
*
,rn1-rn2 AS Diff
,COUNT(*) OVER (PARTITION BY MoreThanTen, rn1-rn2) AS GroupLength
FROM CTE_rn
)
SELECT
ID
,VisitDate
,Counter
FROM CTE_Groups
WHERE
GroupLength >= 3
AND Counter > 10
ORDER BY VisitDate
;
Result
+----+------------+---------+
| ID | VisitDate | Counter |
+----+------------+---------+
| 5 | 2019-01-05 | 18 |
| 6 | 2019-01-06 | 19 |
| 7 | 2019-01-07 | 11 |
+----+------------+---------+
Try this:
select Id, VisitDate, Counter from (
select Id, VisitDate, Counter, count(*) over (partition by grp) cnt from (
select *,
-- here I used difference between row number and day to group consecutive days
row_number() over (order by visitDate) - day(visitDate) grp
from #t1
where [Counter] > 10
) a
) a where cnt >= 3 --where group count is greater or equal to three
Based on the comment that days does not need to be consecutive, just rows have to be consecutive, here is updated query, which uses similair technique:
select id, visitdate, counter from (
select id, visitdate, counter, count(*) over (partition by grp) cnt from (
select *, rn - row_number() over (order by visitDate) grp from (
select *,
case when (Counter > 10) or (lag(Counter) over (order by visitDate) > 10 and Counter > 10) then
row_number() over (order by visitdate) end rn
from #t1
) a where rn is not null
) a
) a where cnt >= 3
I think this might be most simply handled by just looking at the sequences using lead() and lag():
select id, visitdate, counter
from (select t1.*,
lag(counter, 2) over (order by visitdate) as counter_2p,
lag(counter, 1) over (order by visitdate) as counter_1p,
lead(counter, 1) over (order by visitdate) as counter_1l,
lead(counter, 2) over (order by visitdate) as counter_2l
from t1
) t1
where counter >= 10 and
((counter_2p >= 10 and counter_1p >= 10) or
(counter_1p >= 10 and counter_1l >= 10) or
(counter_1l >= 10 and counter_2l >= 10)
);
Cross apply also works for this Question
with result as (
select
t.Id as Id1,t.VisitDate as VisitDate1,t.Counter as Counter1
,tt.Id as Id2,tt.VisitDate as VisitDate2,tt.Counter as Counter2
from #t1 t cross join #t1 tt where DATEDIFF(Day,t.VisitDate,tt.visitDate)=1
and t.Counter>10 and tt.Counter>10
)
select Id1 as Id,VisitDate1 as VisitDate ,Counter1 as [Counter] from result
union
select Id2 as Id,VisitDate2 as VisitDate,Counter2 as [Counter] from result
I have a sample table
CREATE TABLE [dbo].[wt](
[id] [int] NULL,
[dt] [datetime] NULL,
[txt] [nvarchar](50) NULL
) ON [PRIMARY]
GO
INSERT INTO [dbo].[wt]
([id]
,[dt]
,[txt])
VALUES
(1, '2017-01-01 00:01:00.000', 't1'),
(2, '2017-01-01 00:03:00.000', 't1'),
(3, '2017-01-01 00:02:00.000', 't1'),
(4, '2017-01-01 01:04:00.000', 't1'),
(5, '2017-01-01 02:10:00.000', 't1'),
(6, '2017-01-01 00:01:00.000', 't1'),
(7, '2017-01-01 01:05:00.000', 't1'),
(8, '2017-01-01 02:10:00.000', 't2'),
(9, '2017-01-01 00:03:00.000', 't2'),
(10,'2017-01-01 01:04:00.000', 't2'),
(11,'2017-01-01 00:52:00.000', 't1')
I would like to have a list of txt code and dt date grouped by txt code where interval beetwen txt occurrence is at least one hour and nothing in-between.
To clarify when t1 first occures at '2017-01-01 00:01:00.000'
then next occurrence I am looking for is after at least one hour
which will be '2017-01-01 01:04:00.000'
third occurrence I am looking for is after at least one hour from '2017-01-01 01:04:00.000' and so on.
After some searching I found something like this
;with a as (
select txt, dt,
rn = row_number() over (partition by txt order by dt asc)
from [wt]),
b as (
select txt, dt, dt as dt2, rn, null tm, 0 recurrence
from a
where rn = 1
union all
select a.txt, a.dt, a.dt,
a.rn, datediff(MINUTE,a.dt,b.dt) tm,
case when dateadd(MINUTE,-60,a.dt) < b.dt then recurrence + 1 else 0 end
from b join a
on b.rn = a.rn - 1 and b.txt = a.txt
)
select txt, dt, rn, tm, recurrence
from b
where recurrence = 0
order by txt, dt
but this wasn't good because the interval isn't counted from first occurrence but from last, so I got
txt dt rn tm recurrence
t1 2017-01-01 00:01:00.000 1 NULL 0
t1 2017-01-01 02:10:00.000 8 -65 0
t2 2017-01-01 00:03:00.000 1 NULL 0
t2 2017-01-01 01:04:00.000 2 -61 0
t2 2017-01-01 02:10:00.000 3 -66 0
I think I found a workaround because in this case I could group record within same hour but I am not happy with that solution.
select txt, min(dt) dt
into #ttwt
from [wt]
group by txt, substring(convert(varchar,dt,120),1,14)+'00:00.000'
;with a as (
select txt, dt,
rn = row_number() over (partition by txt order by dt asc)
from #ttwt),
b as (
select txt, dt, dt as dt2, rn, null tm, 0 recurrence
from a
where rn = 1
union all
select a.txt, a.dt, a.dt,
a.rn, datediff(MINUTE,a.dt,b.dt) tm,
case when dateadd(MINUTE,-60,a.dt) < b.dt then recurrence + 1 else 0 end
from b join a
on b.rn = a.rn - 1 and b.txt = a.txt
)
select txt, dt, rn, tm, recurrence
from b
where recurrence = 0
order by txt, dt
drop table #ttwt
txt dt rn tm recurrence
t1 2017-01-01 00:01:00.000 1 NULL 0
t1 2017-01-01 01:04:00.000 2 -63 0
t1 2017-01-01 02:10:00.000 3 -66 0
t2 2017-01-01 00:03:00.000 1 NULL 0
t2 2017-01-01 01:04:00.000 2 -61 0
t2 2017-01-01 02:10:00.000 3 -66 0
Any suggestions to improve the script so it will let the interval be any entered value in minutes would be appreciated.
If I have understood correctly I think the following does what you need.
CREATE TABLE #T (id INT , rn INT, txt VARCHAR(10), dt DATETIME, lagDiff INT, runningDiff INT)
INSERT INTO #T (id, rn, txt, dt, lagDiff, runningDiff)
SELECT id
, ROW_NUMBER() OVER( PARTITION BY txt ORDER BY dt, id) -1 rn
, txt
, dt
, DATEDIFF(MINUTE, COALESCE(LAG(dt) OVER( PARTITION BY txt ORDER BY dt, id), dt), dt) Diff
, DATEDIFF(MINUTE, COALESCE(FIRST_VALUE(dt) OVER( PARTITION BY txt ORDER BY dt, id), dt), dt) RunningDiff
FROM wt
; WITH CTE AS (
SELECT *, 1 AS Level
FROM #T
WHERE rn = 0
UNION ALL
SELECT T.*, CTE.Level + 1
FROM #T T
INNER JOIN CTE ON CTE.txt = T.txt AND CTE.rn < T.rn AND T.runningDiff - 60 > CTE.runningDiff
WHERE T.rn > 0
)
, X AS (
SELECT txt
, Level
, MIN(rn) rn
FROM CTE
GROUP BY txt, Level
)
SELECT #T.*
FROM X
INNER JOIN #T ON #T.txt = X.txt AND #T.rn = X.rn
Output
+----+----+-----+-------------------------+---------+-------------+
| id | rn | txt | dt | lagDiff | runningDiff |
+----+----+-----+-------------------------+---------+-------------+
| 1 | 0 | t1 | 2017-01-01 00:01:00.000 | 0 | 0 |
| 4 | 5 | t1 | 2017-01-01 01:04:00.000 | 12 | 63 |
| 5 | 7 | t1 | 2017-01-01 02:10:00.000 | 65 | 129 |
| 9 | 0 | t2 | 2017-01-01 00:03:00.000 | 0 | 0 |
| 10 | 1 | t2 | 2017-01-01 01:04:00.000 | 61 | 61 |
| 8 | 2 | t2 | 2017-01-01 02:10:00.000 | 66 | 127 |
+----+----+-----+-------------------------+---------+-------------+
I kind of like a method that is a bubble sort. The problem I have found when doing recursive operations is they work great for small sets(think less than 5 or 10k), then behave horrid when you get larger. For this reason I like a cursor approach were you are essentially saying: "Are you larger than a criteria? Yes, No. Insert or Ignore, Delete, move on." This way you are evaluating over every item once and once only, not every variation of a theme of recursion.
DECLARE #Temp TABLE
(
id INT
, dt DATETIME
, txt VARCHAR(8)
, rwn INT
)
DECLARE #Holder TABLE
(
id INT
, dt DATETIME
, txt VARCHAR(8)
, Dif int
)
INSERT INTO #Temp
SELECT *, row_number() over (partition by txt order by dt, id) AS rn
From wt
WHILE EXISTS (SELECT 1 FROM #Temp)
BEGIN
DECLARE
#CurId INT
, #CurDt DATETIME
, #Curtxt VARCHAR(8)
, #LastDate DATETIME
;
SELECT TOP 1 #CurId = Id, #CurDt = Dt, #Curtxt = txt FROM #Temp ORDER BY txt, rwn
--If there is not entry you need a single entry
IF NOT EXISTS (SELECT TOP 1 * FROM #Holder)
BEGIN
INSERT INTO #Holder VALUES (#CurId, #CurDt, #curtxt, null)
END
ELSE
--if you reset the grouping you need to reset and begin anew
IF (SELECT rwn FROM #Temp WHERE Id = #CurId) = 1
BEGIN
INSERT INTO #Holder VALUES (#CurId, #CurDt, #curtxt, null)
END
--if you are going along check the logic for the difference of what the last was compared to the current
ELSE
BEGIN
SELECT TOP 1 #LastDate = dt FROM #Holder ORDER BY id desc
IF DATEDIFF(HOUR, #LastDate, #CurDt) >= 1
BEGIN
INSERT INTO #Holder VALUES (#CurId, #CurDt, #curtxt, DATEDIFF(MINUTE, #LastDate, #CurDt))
END
END
--Delete the running values and loop again
DELETE #Temp WHERE Id = #CurId
END
Select *
From #Holder
ID LogDate LogTime InoutMode
1 2017-02-23 19:30:00.0000000 1
2 2017-02-23 20:00:00.0000000 0
3 2017-02-23 20:30:00.0000000 1
4 2017-02-23 21:00:00.0000000 0
5 2017-02-23 21:30:00.0000000 1
6 2017-02-24 08:00:00.0000000 0
The above result got from a temp table.now i want to insert LogTime to another temp table like below.
InOutMode-1 =>InTime
InOutMode-0 =>OutTime
InTime OutTime
19:30:00.0000000 20:00:00.0000000
20:30:00.0000000 21:00:00.0000000
21:30:00.0000000 08:00:00.0000000
To insert the row of one table into another table you can write this SQL query:
INSERT INTO anotherTemp (InTime)
SELECT LogTime
FROM Temp
WHERE InoutMode = 1
INSERT INTO anotherTemp (OutTime)
SELECT LogTime
FROM Temp
WHERE InoutMode = 0
See this link for more http://www.dofactory.com/sql/insert-into
Use 2 cte's, one for InoutMode with 1 and other for 0 and also give row-number based on the order of Id column or date and time columns. Then join these 2 cte's.
Query
;with cte as(
select [rn] = row_number() over(
order by [Id] -- also order by [LogDate], [LogTime]
), *
from #your_temp_table_name
where [InoutMode] = 1
),
cte2 as(
select [rn] = row_number() over(
order by [Id] -- also order by [LogDate], [LogTime]
), *
from #your_temp_table_name
where [InoutMode] = 0
)
select
t1.[LogTime] [InTime],
t2.[LogTime] [OutTime]
from cte t1
left join cte2 t2
on t1.[rn] = t2.[rn];
If you want the result set into a new temp table. Then last part can be modified as
select
t1.[LogTime] [InTime],
t2.[LogTime] [OutTime]
into #your_new_temp_table_name
from cte t1
left join cte2 t2
on t1.[rn] = t2.[rn];
Find demo here
Use INSERT INTO .. SELECT FROM construct like
insert into #temp2
select ID, LogDate,
case when InoutMode = 1 then LogTime end as InTime,
case when InoutMode = 0 then LogTime end as OutTime,
InoutMode
from #temp1
I have table the following data structure in SQL Server:
ID Date Allocation
1, 2012-01-01, 0
2, 2012-01-02, 2
3, 2012-01-03, 0
4, 2012-01-04, 0
5, 2012-01-05, 0
6, 2012-01-06, 5
etc.
What I need to do is get all consecutive day periods where Allocation = 0, and in the following form:
Start Date End Date DayCount
2012-01-01 2012-01-01 1
2012-01-03 2012-01-05 3
etc.
Is it possible to do this in SQL, and if so how?
In this answer, I'll assume that the "id" field numbers the rows consecutively when sorted by increasing date, like it does in the example data. (Such a column can be created if it does not exist).
This is an example of a technique described here and here.
1) Join the table to itself on adjacent "id" values. This pairs adjacent rows. Select rows where the "allocation" field has changed. Store the result in a temporary table, also keeping a running index.
SET #idx = 0;
CREATE TEMPORARY TABLE boundaries
SELECT
(#idx := #idx + 1) AS idx,
a1.date AS prev_end,
a2.date AS next_start,
a1.allocation as allocation
FROM allocations a1
JOIN allocations a2
ON (a2.id = a1.id + 1)
WHERE a1.allocation != a2.allocation;
This gives you a table having "the end of the previous period", "the start of the next period", and "the value of 'allocation' in the previous period" in each row:
+------+------------+------------+------------+
| idx | prev_end | next_start | allocation |
+------+------------+------------+------------+
| 1 | 2012-01-01 | 2012-01-02 | 0 |
| 2 | 2012-01-02 | 2012-01-03 | 2 |
| 3 | 2012-01-05 | 2012-01-06 | 0 |
+------+------------+------------+------------+
2) We need the start and end of each period in the same row, so we need to combine adjacent rows again. Do this by creating a second temporary table like boundaries but having an idx field 1 greater:
+------+------------+------------+
| idx | prev_end | next_start |
+------+------------+------------+
| 2 | 2012-01-01 | 2012-01-02 |
| 3 | 2012-01-02 | 2012-01-03 |
| 4 | 2012-01-05 | 2012-01-06 |
+------+------------+------------+
Now join on the idx field and we get the answer:
SELECT
boundaries2.next_start AS start,
boundaries.prev_end AS end,
allocation
FROM boundaries
JOIN boundaries2
USING(idx);
+------------+------------+------------+
| start | end | allocation |
+------------+------------+------------+
| 2012-01-02 | 2012-01-02 | 2 |
| 2012-01-03 | 2012-01-05 | 0 |
+------------+------------+------------+
** Note that this answer gets the "internal" periods correctly but misses the two "edge" periods where allocation = 0 at the beginning and allocation = 5 at the end. Those can be pulled in using UNION clauses but I wanted to present the core idea without that complication.
Following would be one way to do it. The gist of this solution is
Use a CTE to get a list of all consecutive start and enddates with Allocation = 0
Use the ROW_NUMBER window function to assign rownumbers depending on both start- and enddates.
Select only those records where both ROW_NUMBERS equal 1.
Use DATEDIFFto calculate the DayCount
SQL Statement
;WITH r AS (
SELECT StartDate = Date, EndDate = Date
FROM YourTable
WHERE Allocation = 0
UNION ALL
SELECT r.StartDate, q.Date
FROM r
INNER JOIN YourTable q ON DATEDIFF(dd, r.EndDate, q.Date) = 1
WHERE q.Allocation = 0
)
SELECT [Start Date] = s.StartDate
, [End Date ] = s.EndDate
, [DayCount] = DATEDIFF(dd, s.StartDate, s.EndDate) + 1
FROM (
SELECT *
, rn1 = ROW_NUMBER() OVER (PARTITION BY StartDate ORDER BY EndDate DESC)
, rn2 = ROW_NUMBER() OVER (PARTITION BY EndDate ORDER BY StartDate ASC)
FROM r
) s
WHERE s.rn1 = 1
AND s.rn2 = 1
OPTION (MAXRECURSION 0)
Test script
;WITH q (ID, Date, Allocation) AS (
SELECT * FROM (VALUES
(1, '2012-01-01', 0)
, (2, '2012-01-02', 2)
, (3, '2012-01-03', 0)
, (4, '2012-01-04', 0)
, (5, '2012-01-05', 0)
, (6, '2012-01-06', 5)
) a (a, b, c)
)
, r AS (
SELECT StartDate = Date, EndDate = Date
FROM q
WHERE Allocation = 0
UNION ALL
SELECT r.StartDate, q.Date
FROM r
INNER JOIN q ON DATEDIFF(dd, r.EndDate, q.Date) = 1
WHERE q.Allocation = 0
)
SELECT s.StartDate, s.EndDate, DATEDIFF(dd, s.StartDate, s.EndDate) + 1
FROM (
SELECT *
, rn1 = ROW_NUMBER() OVER (PARTITION BY StartDate ORDER BY EndDate DESC)
, rn2 = ROW_NUMBER() OVER (PARTITION BY EndDate ORDER BY StartDate ASC)
FROM r
) s
WHERE s.rn1 = 1
AND s.rn2 = 1
OPTION (MAXRECURSION 0)
Alternative way with CTE but without ROW_NUMBER(),
Sample data:
if object_id('tempdb..#tab') is not null
drop table #tab
create table #tab (id int, date datetime, allocation int)
insert into #tab
select 1, '2012-01-01', 0 union
select 2, '2012-01-02', 2 union
select 3, '2012-01-03', 0 union
select 4, '2012-01-04', 0 union
select 5, '2012-01-05', 0 union
select 6, '2012-01-06', 5 union
select 7, '2012-01-07', 0 union
select 8, '2012-01-08', 5 union
select 9, '2012-01-09', 0 union
select 10, '2012-01-10', 0
Query:
;with cte(s_id, e_id, b_id) as (
select s.id, e.id, b.id
from #tab s
left join #tab e on dateadd(dd, 1, s.date) = e.date and e.allocation = 0
left join #tab b on dateadd(dd, -1, s.date) = b.date and b.allocation = 0
where s.allocation = 0
)
select ts.date as [start date], te.date as [end date], count(*) as [day count] from (
select c1.s_id as s, (
select min(s_id) from cte c2
where c2.e_id is null and c2.s_id >= c1.s_id
) as e
from cte c1
where b_id is null
) t
join #tab t1 on t1.id between t.s and t.e and t1.allocation = 0
join #tab ts on ts.id = t.s
join #tab te on te.id = t.e
group by t.s, t.e, ts.date, te.date
Live example at data.SE.
Using this sample data:
CREATE TABLE MyTable (ID INT, Date DATETIME, Allocation INT);
INSERT INTO MyTable VALUES (1, {d '2012-01-01'}, 0);
INSERT INTO MyTable VALUES (2, {d '2012-01-02'}, 2);
INSERT INTO MyTable VALUES (3, {d '2012-01-03'}, 0);
INSERT INTO MyTable VALUES (4, {d '2012-01-04'}, 0);
INSERT INTO MyTable VALUES (5, {d '2012-01-05'}, 0);
INSERT INTO MyTable VALUES (6, {d '2012-01-06'}, 5);
GO
Try this:
WITH DateGroups (ID, Date, Allocation, SeedID) AS (
SELECT MyTable.ID, MyTable.Date, MyTable.Allocation, MyTable.ID
FROM MyTable
LEFT JOIN MyTable Prev ON Prev.Date = DATEADD(d, -1, MyTable.Date)
AND Prev.Allocation = 0
WHERE Prev.ID IS NULL
AND MyTable.Allocation = 0
UNION ALL
SELECT MyTable.ID, MyTable.Date, MyTable.Allocation, DateGroups.SeedID
FROM MyTable
JOIN DateGroups ON MyTable.Date = DATEADD(d, 1, DateGroups.Date)
WHERE MyTable.Allocation = 0
), StartDates (ID, StartDate, DayCount) AS (
SELECT SeedID, MIN(Date), COUNT(ID)
FROM DateGroups
GROUP BY SeedID
), EndDates (ID, EndDate) AS (
SELECT SeedID, MAX(Date)
FROM DateGroups
GROUP BY SeedID
)
SELECT StartDates.StartDate, EndDates.EndDate, StartDates.DayCount
FROM StartDates
JOIN EndDates ON StartDates.ID = EndDates.ID;
The first section of the query is a recursive SELECT, which is anchored by all rows that are allocation = 0, and whose previous day either doesn't exist or has allocation != 0. This effectively returns IDs: 1 and 3 which are the starting dates of the periods of time you want to return.
The recursive part of this same query starts from the anchor rows, and finds all subsequent dates that also have allocation = 0. The SeedID keeps track of the anchored ID through all the iterations.
The result so far is this:
ID Date Allocation SeedID
----------- ----------------------- ----------- -----------
1 2012-01-01 00:00:00.000 0 1
3 2012-01-03 00:00:00.000 0 3
4 2012-01-04 00:00:00.000 0 3
5 2012-01-05 00:00:00.000 0 3
The next sub query uses a simple GROUP BY to filter out all the start dates for each SeedID, and also counts the days.
The last sub query does the same thing with the end dates, but this time the day count isn't needed as we already have this.
The final SELECT query joins these two together to combine the start and end dates, and returns them along with the day count.
Give it a try if it works for you
Here SDATE for your DATE remains same as your table.
SELECT SDATE,
CASE WHEN (SELECT COUNT(*)-1 FROM TABLE1 WHERE ID BETWEEN TBL1.ID AND (SELECT MIN(ID) FROM TABLE1 WHERE ID > TBL1.ID AND ALLOCATION!=0)) >0 THEN(
CASE WHEN (SELECT SDATE FROM TABLE1 WHERE ID =(SELECT MAX(ID) FROM TABLE1 WHERE ID >TBL1.ID AND ID<(SELECT MIN(ID) FROM TABLE1 WHERE ID > TBL1.ID AND ALLOCATION!=0))) IS NULL THEN SDATE
ELSE (SELECT SDATE FROM TABLE1 WHERE ID =(SELECT MAX(ID) FROM TABLE1 WHERE ID >TBL1.ID AND ID<(SELECT MIN(ID) FROM TABLE1 WHERE ID > TBL1.ID AND ALLOCATION!=0))) END
)ELSE (SELECT SDATE FROM TABLE1 WHERE ID = (SELECT MAX(ID) FROM TABLE1 WHERE ID > TBL1.ID ))END AS EDATE
,CASE WHEN (SELECT COUNT(*)-1 FROM TABLE1 WHERE ID BETWEEN TBL1.ID AND (SELECT MIN(ID) FROM TABLE1 WHERE ID > TBL1.ID AND ALLOCATION!=0)) <0 THEN
(SELECT COUNT(*) FROM TABLE1 WHERE ID BETWEEN TBL1.ID AND (SELECT MAX(ID) FROM TABLE1 WHERE ID > TBL1.ID )) ELSE
(SELECT COUNT(*)-1 FROM TABLE1 WHERE ID BETWEEN TBL1.ID AND (SELECT MIN(ID) FROM TABLE1 WHERE ID > TBL1.ID AND ALLOCATION!=0)) END AS DAYCOUNT
FROM TABLE1 TBL1 WHERE ALLOCATION = 0
AND (((SELECT ALLOCATION FROM TABLE1 WHERE ID=(SELECT MAX(ID) FROM TABLE1 WHERE ID < TBL1.ID))<> 0 ) OR (SELECT MAX(ID) FROM TABLE1 WHERE ID < TBL1.ID)IS NULL);
A solution without CTE:
SELECT a.aDate AS StartDate
, MIN(c.aDate) AS EndDate
, (datediff(day, a.aDate, MIN(c.aDate)) + 1) AS DayCount
FROM (
SELECT x.aDate, x.allocation, COUNT(*) idn FROM table1 x
JOIN table1 y ON y.aDate <= x.aDate
GROUP BY x.id, x.aDate, x.allocation
) AS a
LEFT JOIN (
SELECT x.aDate, x.allocation, COUNT(*) idn FROM table1 x
JOIN table1 y ON y.aDate <= x.aDate
GROUP BY x.id, x.aDate, x.allocation
) AS b ON a.idn = b.idn + 1 AND b.allocation = a.allocation
LEFT JOIN (
SELECT x.aDate, x.allocation, COUNT(*) idn FROM table1 x
JOIN table1 y ON y.aDate <= x.aDate
GROUP BY x.id, x.aDate, x.allocation
) AS c ON a.idn <= c.idn AND c.allocation = a.allocation
LEFT JOIN (
SELECT x.aDate, x.allocation, COUNT(*) idn FROM table1 x
JOIN table1 y ON y.aDate <= x.aDate
GROUP BY x.id, x.aDate, x.allocation
) AS d ON c.idn = d.idn - 1 AND d.allocation = c.allocation
WHERE b.idn IS NULL AND c.idn IS NOT NULL AND d.idn IS NULL AND a.allocation = 0
GROUP BY a.aDate
Example
Can I limit rows by sum of a column in a SQL Server database?
For example:
Type | Time (in minutes)
-------------------------
A | 50
B | 10
C | 30
D | 20
E | 70
...
And I want to limit the selection by sum of time. For example maximum of 100 minutes. Table must look like this:
Type | Time (in minutes)
-------------------------
A | 50
B | 10
C | 30
Any ideas? Thanks.
DECLARE #T TABLE
(
[Type] CHAR(1) PRIMARY KEY,
[Time] INT
)
INSERT INTO #T
SELECT 'A',50 UNION ALL
SELECT 'B',10 UNION ALL
SELECT 'C',30 UNION ALL
SELECT 'D',20 UNION ALL
SELECT 'E',70;
WITH RecursiveCTE
AS (
SELECT TOP 1 [Type], [Time], CAST([Time] AS BIGINT) AS Total
FROM #T
ORDER BY [Type]
UNION ALL
SELECT R.[Type], R.[Time], R.Total
FROM (
SELECT T.*,
T.[Time] + Total AS Total,
rn = ROW_NUMBER() OVER (ORDER BY T.[Type])
FROM #T T
JOIN RecursiveCTE R
ON R.[Type] < T.[Type]
) R
WHERE R.rn = 1 AND Total <= 100
)
SELECT [Type], [Time], Total
FROM RecursiveCTE
OPTION (MAXRECURSION 0);
Or if your table is small
SELECT t1.[Type],
t1.[Time],
SUM(t2.[Time])
FROM #T t1
JOIN #T t2
ON t2.[Type] <= t1.[Type]
GROUP BY t1.[Type],t1.[Time]
HAVING SUM(t2.[Time]) <=100