Calculate Running count in SQL Server excluding certain rows - sql

Question

This is my table/ source data,
|---------------------------------------------------|
|ID | DT | DAY | ATTENDANCE |
|-------+---------------+-----------+---------------|
|89 | 2014-08-23 | NULL | 1 |
|90 | 2014-08-24 | Sunday | NULL |
|91 | 2014-08-25 | NULL | 1 |
|92 | 2014-08-26 | NULL | 1 |
|93 | 2014-08-27 | NULL | 0 |
|94 | 2014-08-28 | NULL | 1 |
|95 | 2014-08-29 | NULL | 0 |
|96 | 2014-08-30 | NULL | 1 |
|97 | 2014-08-31 | Sunday | NULL |
|98 | 2014-08-01 | NULL | 1 |
|99 | 2014-08-02 | NULL | 1 |
|100 | 2014-08-03 | NULL | 0 |
|101 | 2014-08-04 | NULL | 0 |
|102 | 2014-08-05 | NULL | 1 |
|103 | 2014-08-06 | NULL | 1 |
|104 | 2014-08-07 | Sunday | NULL |
|105 | 2014-08-08 | NULL | 1 |
|106 | 2014-08-09 | NULL | 1 |
|107 | 2014-08-10 | NULL | 1 |
|---------------------------------------------------|
I want a result as given. The 5th column [Streak] is what I am trying to calculate. It is a value calculated based on attendance. At any day, If [ATTENDANCE] = 0, [Streak] becomes reset to 0.
|-------------------------------------------------------------|
|ID | DT | DAY | ATTENDANCE | Streak |
|-------+---------------+-----------+---------------+---------|
|89 | 2014-08-23 | NULL | 1 | 1 |
|90 | 2014-08-24 | Sunday | NULL | |
|91 | 2014-08-25 | NULL | 1 | 2 |
|92 | 2014-08-26 | NULL | 1 | 3 |
|93 | 2014-08-27 | NULL | 0 | 0 |
|94 | 2014-08-28 | NULL | 1 | 1 |
|95 | 2014-08-29 | NULL | 0 | 0 |
|96 | 2014-08-30 | NULL | 1 | 1 |
|97 | 2014-08-31 | Sunday | NULL | |
|98 | 2014-08-01 | NULL | 1 | 2 |
|99 | 2014-08-02 | NULL | 1 | 3 |
|100 | 2014-08-03 | NULL | 0 | 0 |
|101 | 2014-08-04 | NULL | 0 | 0 |
|102 | 2014-08-05 | NULL | 1 | 1 |
|103 | 2014-08-06 | NULL | 1 | 2 |
|104 | 2014-08-07 | Sunday | NULL | |
|105 | 2014-08-08 | NULL | 1 | 3 |
|106 | 2014-08-09 | NULL | 1 | 4 |
|107 | 2014-08-10 | NULL | 1 | 5 |
|-------------------------------------------------------------|
This is what I so far did. For me Sunday is also getting incremented as a Day.
Any help, in solving it..
SQL
SELECT X.*, X.ID - LU.FROMID + 1
FROM #TAB X LEFT JOIN
(
SELECT (SELECT MIN(ID) FROM #TAB) FROMID,MIN(ID) TOID FROM #TAB WHERE ATTENDANCE = 0
UNION
SELECT A.ID FROMID,B.ID TOID
FROM (SELECT ID,ROW_NUMBER() OVER (ORDER BY ID) R FROM #TAB WHERE ATTENDANCE = 0) A CROSS JOIN
(SELECT ID,ROW_NUMBER() OVER (ORDER BY ID) R FROM #TAB WHERE ATTENDANCE = 0) B
WHERE A.R = (B.R - 1)
UNION
SELECT MAX(ID),(SELECT MAX(ID) FROM #TAB) FROM #TAB WHERE ATTENDANCE = 0
UNION
SELECT MAX(ID),MAX(ID) + 1 FROM #TAB
) LU
ON X.ID >= LU.FROMID AND X.ID < LU.TOID
Source Data for testing:
SET DATEFORMAT DMY
DECLARE #TAB TABLE (ID INT IDENTITY(89,1),DT DATE,DAY VARCHAR(15),ATTENDANCE BIT)
INSERT INTO #TAB VALUES
('23-08-2014',Null,1),
('24-08-2014','Sunday',Null ),
('25-08-2014',Null ,1),
('26-08-2014',Null ,1),
('27-08-2014',Null ,0),
('28-08-2014',Null ,1),
('29-08-2014',Null ,0),
('30-08-2014',Null ,1),
('31-08-2014','Sunday',Null ),
('01-08-2014',Null ,1),
('02-08-2014',Null ,1),
('03-08-2014',Null ,0),
('04-08-2014',Null ,1),
('05-08-2014',Null ,0),
('06-08-2014',Null ,1),
('07-08-2014','Sunday',Null ),
('08-08-2014',Null ,1),
('09-08-2014',Null ,1),
('10-08-2014',Null ,1)
Thanks in Advance.

Answer 1

#HHH, I added another temp table around #TAB. This works, Please test it and tell.
DECLARE #TAB2 TABLE (MASTERID INT IDENTITY(1,1),ID INT,DT DATE,DAY VARCHAR(15),ATTENDANCE BIT)
INSERT INTO #TAB2
SELECT * FROM #TAB WHERE DAY IS NULL
SELECT Y.*,
LU2.Streak
FROM #TAB Y LEFT JOIN (
SELECT X.ID, X.MASTERID - LU.FROMID + 1 [Streak]
FROM #TAB2 X LEFT JOIN
(
SELECT (SELECT MIN(MASTERID) FROM #TAB2) FROMID,MIN(MASTERID) TOID FROM #TAB2 WHERE ATTENDANCE = 0
UNION
SELECT A.MASTERID FROMID,B.MASTERID TOID
FROM (SELECT MASTERID,ROW_NUMBER() OVER (ORDER BY MASTERID) R FROM #TAB2 WHERE ATTENDANCE = 0) A CROSS JOIN
(SELECT MASTERID,ROW_NUMBER() OVER (ORDER BY MASTERID) R FROM #TAB2 WHERE ATTENDANCE = 0) B
WHERE A.R = (B.R - 1)
UNION
SELECT MAX(MASTERID),(SELECT MAX(MASTERID) FROM #TAB2) FROM #TAB2 WHERE ATTENDANCE = 0
UNION
SELECT MAX(MASTERID),MAX(MASTERID) + 1 FROM #TAB2
) LU
ON X.MASTERID >= LU.FROMID AND X.MASTERID < LU.TOID ) LU2
ON Y.ID = LU2.ID
RESULT:

Answer 2

Like said Gordon, you need to identify groups of consecutive attendace day. But this is another version..
select
t.*,
t2.Seq,
(case ATTENDANCE
when 1 then ROW_NUMBER() over (partition by t2.Seq,t.attendance order by t.id)
when 0 then 0
else null end) Streak
from
#TAB t
outer apply
(select
count(1) as Seq
from
#tab t2
where
t2.id < t.ID
and
t2.ATTENDANCE = 0) as t2
order by
t.id

Answer 3

You need to identify groups of consecutive attendance days. You can do this using the difference of row_numbers() approach. Subtract the row_number() by attendance (and without Sunday) from the overall row_number() (without Sunday). The difference is constant for a streak. The rest is just another row_number() call.
Here is what it looks like:
select t.id, t.dt, t.day, t.attendance,
(case when t.day = 'Sunday' then NULL
else row_number() over (partition by attendance, grp order by dt)
end) as streak
from (select t.*,
(row_number() over (partition by day order by dt) -
row_number() over (partition by day, attendance order by dt)
) as grp
from #tab t
) t;

Answer 4

According to your requirement, i think this query will be useful ,please change the fieldname and table name.it's a very simple way to get calculate simple running total in a table......
SELECT a.id,a.dateday, a.dateval, (SELECT SUM(b.hoursday) FROM #temp b
WHERE b.dateday<=a.dateday) as totalhours
FROM #temp a
Thanks