Grouping date ranges into rolling group of 15 days - sql

I have data with start and end datetime on each row, I'm trying to group all rows that fall within 15days of the end datetime, reset and begin new group using the end datetime when the start date doesn't fall in 15 days.
I tried different solutions like using logic to self join and try to pick min end date after looking for start date time between end datetime and end datetime + 30, but I'm not able to identify a way to continue using end datetime from the start/previous row when the start datetime is within 15days till it finds a row that doesn't.
with cte as
(
SELECT 1 id, '2018-12-25' col1, '2019-01-05' col2
UNION ALL
SELECT 1, '2019-03-01' col1, '2019-03-10' col2
UNION ALL
SELECT 1, '2019-03-15' col1, '2019-03-19' col2
UNION ALL
SELECT 1, '2019-03-22' col1, '2019-03-28' col2
UNION ALL
SELECT 1, '2019-03-30' col1, '2019-04-02' col2
UNION ALL
SELECT 1, '2019-04-10' col1, '2019-04-15' col2
UNION ALL
SELECT 1, '2019-04-18' col1, '2019-04-25' col2
), STG AS
(
SELECT A.*, MIN(B.COL2) AS GRP_COL2
FROm CTE A
LEFT OUTER JOIN CTE B ON A.col1 BETWEEN B.col2 AND DATEADD(day, 15, B.col2)
GROUP BY A.id, A.col1, A.col2
)
SELECT A.id, A.col1, A.col2, COALESCE(GRP_COL2, A.COL2) AS GRP_COL2
FROM STG A
ORDER By 1,2
--this returns wrong result for 04/10 and 04/18 dated rows in this case
For example I have following data:
ProductId ProductStartDt ProductEndDt
1 2018-12-25 2019-01-05
1 2019-03-01 2019-03-10
1 2019-03-15 2019-03-19
1 2019-03-22 2019-03-28
1 2019-03-30 2019-04-02
1 2019-04-10 2019-04-15
1 2019-04-18 2019-04-25
So what I'm trying to do is start from min end date and group all rows with start date that fall in end date + 15 days range as one group and start next row as new group and continue using the end date of that row as new start for group.
Expected Output
ProductId ProductStartDt ProductEndDt GroupNo
1 2018-12-25 2019-01-05 1
1 2019-03-01 2019-03-10 2
1 2019-03-15 2019-03-19 2
1 2019-03-22 2019-03-28 2
1 2019-03-30 2019-04-02 3
1 2019-04-10 2019-04-15 3
1 2019-04-18 2019-04-25 4
For group 2, the start date doesn't fall within 15days of the end date from Group 1 or start row and this will trigger start of new rows. We see that 03/15 and 03/22 are within 15 days of 03/10 and we group them into single group. 03/30 doesn't fall in the previous group and this will cause start of new group 3 till we find all start dates within the end datetime + 15days.

With recursive CTE we can iterate over initial table, and decide whether current row belongs an existing group or it's a new group:
;with Product as (
select * from (
VALUES
(1, '2018-12-25','2019-01-05' ),
(1, '2019-03-01','2019-03-10' ),
(1, '2019-03-15','2019-03-19' ),
(1, '2019-03-22','2019-03-28' ),
(1, '2019-03-30','2019-04-02' ),
(1, '2019-04-10','2019-04-15' ),
(1, '2019-04-18','2019-04-25' )
) as a1 (ProductId ,ProductStartDt ,ProductEndDt)
), OrderedProduct as (
select *, ROW_NUMBER() over (order by ProductStartDt) as RowNum
from Product
), DateGroupsInterim (RowNum, GroupNum, GrpStartDt) as (
select RowNum, 1, ProductEndDt
from OrderedProduct
where RowNum=1
union all
select OrderedProduct.RowNum,
CASE WHEN OrderedProduct.ProductStartDt <= dateadd(day, 15, dgi.GrpStartDt)
THEN dgi.GroupNum
ELSE dgi.GroupNum + 1
END,
CASE WHEN OrderedProduct.ProductStartDt <= dateadd(day, 15, dgi.GrpStartDt)
THEN dgi.GrpStartDt
ELSE OrderedProduct.ProductEndDt
END
from DateGroupsInterim dgi
join OrderedProduct on OrderedProduct.RowNum=dgi.RowNum+1
) select OrderedProduct.ProductId, OrderedProduct.ProductStartDt, OrderedProduct.ProductEndDt, DateGroupsInterim.GroupNum as GroupNo
from DateGroupsInterim
JOIN OrderedProduct on OrderedProduct.RowNum = DateGroupsInterim.RowNum;

Related

SQL : create intermediate data from date range

I have a table as shown here:
USER
ROI
DATE
1
5
2021-11-24
1
4
2021-11-26
1
6
2021-11-29
I want to get the ROI for the dates in between the other dates, expected result will be as below
From 2021-11-24 to 2021-11-30
USER
ROI
DATE
1
5
2021-11-24
1
5
2021-11-25
1
4
2021-11-26
1
4
2021-11-27
1
4
2021-11-28
1
6
2021-11-29
1
6
2021-11-30
You may use a calendar table approach here. Create a table containing all dates and then join with it. Sans an actual table, you may use an inline CTE:
WITH dates AS (
SELECT '2021-11-24' AS dt UNION ALL
SELECT '2021-11-25' UNION ALL
SELECT '2021-11-26' UNION ALL
SELECT '2021-11-27' UNION ALL
SELECT '2021-11-28' UNION ALL
SELECT '2021-11-29' UNION ALL
SELECT '2021-11-30'
),
cte AS (
SELECT USER, ROI, DATE, LEAD(DATE) OVER (ORDER BY DATE) AS NEXT_DATE
FROM yourTable
)
SELECT t.USER, t.ROI, d.dt
FROM dates d
INNER JOIN cte t
ON d.dt >= t.DATE AND (d.dt < t.NEXT_DATE OR t.NEXT_DATE IS NULL)
ORDER BY d.dt;

Incremental business day column that resets each month

I need to create a table that contains records with 1) all 365 days of the year and 2) a counter representing which business day of the month the day is. Non-business days should be represented with a 0. For example:
Date | Business Day
2019-10-01 1
2019-10-02 2
2019-10-03 3
2019-10-04 4
2019-10-05 0 // Saturday
2019-10-06 0 // Sunday
2019-10-07 5
....
2019-11-01 1
2019-11-02 0 // Saturday
2019-11-03 0 // Sunday
2019-11-04 2
So far, I've been able to create a table that contains all dates of the year.
CREATE TABLE ${TMPID}_days_of_the_year
(
`theDate` STRING
);
INSERT OVERWRITE TABLE ${TMPID}_days_of_the_year
select
dt_set.theDate
from
(
-- last 0~99 months
select date_sub('2019-12-31', a.s + 10*b.s + 100*c.s) as theDate
from
(
select 0 as s union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9
) a
cross join
(
select 0 as s union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9
) b
cross join
(
select 0 as s union all select 1 union all select 2 union all select 3
) c
) dt_set
where dt_set.theDate between '2019-01-01' and '2019-12-31'
order by dt_set.theDate DESC;
And I also have a table that contains all of the weekend days and holidays (this data is loaded from a file, and the date format is YYYY-MM-DD)
CREATE TABLE ${TMPID}_company_holiday
(
`holidayDate` STRING
)
;
LOAD DATA LOCAL INPATH '${FILE}' INTO TABLE ${TMPID}_company_holiday;
My question is.... how do I join these tables together while creating the business day counter column shown as in the sample data above?
You can use row_number() for the enumeration. This is a little tricky, because it needs to be conditional, but the information you need is provided by a left join:
select dy.*,
(case when ch.holiday_date is null
then row_number() over (partition by trunc(dy.date, 'MONTH'), ch.holiday_date
order by dy.date
)
else 0
end) as business_day
from days_of_the_year dy left join
company_holiday ch
on dy.date = ch.holiday_date;

Frequency Distribution by Day

I have records of No. of calls coming to a call center. When a call comes into a call center a ticket is open.
So, let's say ticket 1 (T1) is open on 8/1/19 and it stays open till 8/5/19. So, if a person ran a query everyday then on 8/1 it will show 1 ticket open...same think on day 2 till day 5....I want to get records by day to see how many tickets were open for each day.....
In short, Frequency Distribution by Day.
Ticket Open_date Close_date
T1 8/1/2019 8/5/2019
T2 8/1/2019 8/6/2019
Result:
Result
Date # Tickets_Open
8/1/2019 2
8/2/2019 2
8/3/2019 2
8/4/2019 2
8/5/2019 2
8/6/2019 1
8/7/2019 0
8/8/2019 0
8/9/2019 0
8/10/2019 0
We can handle your requirement via the use of a calendar table, which stores all dates covering the full range in your data set.
WITH dates AS (
SELECT '2019-08-01' AS dt UNION ALL
SELECT '2019-08-02' UNION ALL
SELECT '2019-08-03' UNION ALL
SELECT '2019-08-04' UNION ALL
SELECT '2019-08-05' UNION ALL
SELECT '2019-08-06' UNION ALL
SELECT '2019-08-07' UNION ALL
SELECT '2019-08-08' UNION ALL
SELECT '2019-08-09' UNION ALL
SELECT '2019-08-10'
)
SELECT
d.dt,
COUNT(t.Open_date) AS num_tickets_open
FROM dates d
LEFT JOIN tickets t
ON d.dt BETWEEN t.Open_date AND t.Close_date
GROUP BY
d.dt;
Note that in practice if you expect to have this reporting requirement in the long term, you might want to replace the dates CTE above with a bona-fide table of dates.
This solution generates the list of dates from the tickets table using CTE recursion and calculates the count:
WITH Tickets(Ticket, Open_date, Close_date) AS
(
SELECT "T1", "8/1/2019", "8/5/2019"
UNION ALL
SELECT "T2", "8/1/2019", "8/6/2019"
),
Ticket_dates(Ticket, Dates) as
(
SELECT t1.Ticket, CONVERT(DATETIME, t1.Open_date)
FROM Tickets t1
UNION ALL
SELECT t1.Ticket, DATEADD(dd, 1, CONVERT(DATETIME, t1.Dates))
FROM Ticket_dates t1
inner join Tickets t2 on t1.Ticket = t2.Ticket
where DATEADD(dd, 1, CONVERT(DATETIME, t1.Dates)) <= CONVERT(DATETIME, t2.Close_date)
)
SELECT CONVERT(varchar, Dates, 1), count(*)
FROM Ticket_dates
GROUP by Dates
ORDER by Dates
A "general purpose" trick is to generate a series of numbers, which can be done using CTE's but there are many alternatives, and from that create the needed range of dates. Once that exists then you can left join your ticket data to this and then count by date.
CREATE TABLE mytable(
Ticket VARCHAR(8) NOT NULL PRIMARY KEY
,Open_date DATE NOT NULL
,Close_date DATE NOT NULL
);
INSERT INTO mytable(Ticket,Open_date,Close_date) VALUES ('T1','8/1/2019','8/5/2019');
INSERT INTO mytable(Ticket,Open_date,Close_date) VALUES ('T2','8/1/2019','8/6/2019');
Also note I am using a cross apply in this example to "attach" the min and max dates of your tickets to each numbered row. You would need to include your own logic on what data to select here.
;WITH
cteDigits AS (
SELECT 0 AS digit UNION ALL SELECT 1 UNION ALL SELECT 2 UNION ALL SELECT 3 UNION ALL SELECT 4 UNION ALL
SELECT 5 UNION ALL SELECT 6 UNION ALL SELECT 7 UNION ALL SELECT 8 UNION ALL SELECT 9
)
, cteTally AS (
SELECT
[1s].digit
+ [10s].digit * 10
+ [100s].digit * 100 /* add more like this as needed */
AS num
FROM cteDigits [1s]
CROSS JOIN cteDigits [10s]
CROSS JOIN cteDigits [100s] /* add more like this as needed */
)
select
n.num + 1 rownum
, dateadd(day,n.num,ca.min_date) as on_date
, count(t.Ticket) as tickets_open
from cteTally n
cross apply (select min(Open_date), max(Close_date) from mytable) ca (min_date, max_date)
left join mytable t on dateadd(day,n.num,ca.min_date) between t.Open_date and t.Close_date
where dateadd(day,n.num,ca.min_date) <= ca.max_date
group by
n.num + 1
, dateadd(day,n.num,ca.min_date)
order by
rownum
;
result:
+--------+---------------------+--------------+
| rownum | on_date | tickets_open |
+--------+---------------------+--------------+
| 1 | 01.08.2019 00:00:00 | 2 |
| 2 | 02.08.2019 00:00:00 | 2 |
| 3 | 03.08.2019 00:00:00 | 2 |
| 4 | 04.08.2019 00:00:00 | 2 |
| 5 | 05.08.2019 00:00:00 | 2 |
| 6 | 06.08.2019 00:00:00 | 1 |
+--------+---------------------+--------------+

Fill up date gap by month

I have table of products and their sales quantity in months.
Product Month Qty
A 2018-01-01 5
A 2018-02-01 3
A 2018-05-01 5
B 2018-08-01 10
B 2018-10-01 12
...
I'd like to first fill in the data gap between each product's min and max dates like below:
Product Month Qty
A 2018-01-01 5
A 2018-02-01 3
A 2018-03-01 0
A 2018-04-01 0
A 2018-05-01 5
B 2018-08-01 10
B 2018-09-01 0
B 2018-10-01 12
...
Then I would need to perform an accumulation of each product's sales quantity by month.
Product Month total_Qty
A 2018-01-01 5
A 2018-02-01 8
A 2018-03-01 8
A 2018-04-01 8
A 2018-05-01 13
B 2018-08-01 10
B 2018-09-01 10
B 2018-10-01 22
...
I fumbled over the "cross join" clause, however it seems to generate some unexpected results for me. Could someone help to give a hint how I can achieve this in SQL?
Thanks a lot in advance.
I think a recursive CTE is a simple way to do this. The code is just:
with cte as (
select product, min(mon) as mon, max(mon) as end_mon
from t
group by product
union all
select product, dateadd(month, 1, mon), end_mon
from cte
where mon < end_mon
)
select cte.product, cte.mon, coalesce(qty, 0) as qty
from cte left join
t
on t.product = cte.product and t.mon = cte.mon;
Here is a db<>fiddle.
Hi i think this example can help you and perform what you excepted :
CREATE TABLE #MyTable
(Product varchar(10),
ProductMonth DATETIME,
Qty int
);
GO
CREATE TABLE #MyTableTempDate
(
FullMonth DATETIME
);
GO
INSERT INTO #MyTable
SELECT 'A', '2019-01-01', 214
UNION
SELECT 'A', '2019-02-01', 4
UNION
SELECT 'A', '2019-03-01', 50
UNION
SELECT 'B', '2019-01-01', 214
UNION
SELECT 'B', '2019-02-01', 10
UNION
SELECT 'C', '2019-04-01', 150
INSERT INTO #MyTableTempDate
SELECT '2019-01-01'
UNION
SELECT '2019-02-01'
UNION
SELECT '2019-03-01'
UNION
SELECT '2019-04-01'
UNION
SELECT '2019-05-01'
UNION
SELECT '2019-06-01'
UNION
SELECT '2019-07-01';
------------- FOR NEWER SQL SERVER VERSION > 2005
WITH MyCTE AS
(
SELECT T.Product, T.ProductMonth AS 'MMonth', T.Qty
FROM #MyTable T
UNION
SELECT T.Product, TD.FullMonth AS 'MMonth', 0 AS 'Qty'
FROM #MyTable T, #MyTableTempDate TD
WHERE NOT EXISTS (SELECT 1 FROM #MyTable TT WHERE TT.Product = T.Product AND TD.FullMonth = TT.ProductMonth)
)
-- SELECT * FROM MyCTE;
SELECT Product, MMonth, Qty, SUM( Qty) OVER(PARTITION BY Product ORDER BY Product
ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) as 'TotalQty'
FROM MyCTE
ORDER BY Product, MMonth ASC;
DROP TABLE #MyTable
DROP TABLE #MyTableTempDate
I have other way to perform this in lower SQL Server Version (like 2005 and lower)
It's a SELECT on SELECT if it's your case let me know and i provide some other example.
You can create the months with a recursive CTE
DECLARE #MyTable TABLE
(
ProductID CHAR(1),
Date DATE,
Amount INT
)
INSERT INTO #MyTable
VALUES
('A','2018-01-01', 5),
('A','2018-02-01', 3),
('A','2018-05-01', 5),
('B','2018-08-01', 10),
('B','2018-10-01', 12)
DECLARE #StartDate DATE
DECLARE #EndDate DATE
SELECT #StartDate = MIN(Date), #EndDate = MAX(Date) FROM #MyTable
;WITH dates AS (
SELECT #StartDate AS Date
UNION ALL
SELECT DATEADD(Month, 1, Date)
FROM dates
WHERE Date < #EndDate
)
SELECT A.ProductID, d.Date, COALESCE(Amount,0) AS Amount, COALESCE(SUM(Amount) OVER(PARTITION BY A.ProductID ORDER BY A.ProductID, d.Date ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW),0) AS Total
FROM
(
SELECT ProductID, MIN(date) as DateStart, MAX(date) as DateEnd
FROM #MyTable
GROUP BY ProductID -- As I read in your comments that you need different min and max dates per product
) A
JOIN dates d ON d.Date >= A.DateStart AND d.Date <= A.DateEnd
LEFT JOIN #MyTable T ON A.ProductID = T.ProductID AND T.Date = d.Date
ORDER BY A.ProductID, d.Date
Try this below
IF OBJECT_ID('tempdb..#Temp') IS NOT NULL
DROP TABLE #Temp
;WITH CTE(Product,[Month],Qty)
AS
(
SELECT 'A','2018-01-01', 5 UNION ALL
SELECT 'A','2018-02-01', 3 UNION ALL
SELECT 'A','2018-05-01', 5 UNION ALL
SELECT 'B','2018-08-01', 10 UNION ALL
SELECT 'D','2018-10-01', 12
)
SELECT ct.Product,[MonthDays],ct.Qty
INTO #Temp
FROM
(
SELECT c.Product,[Month],
ISNULL(Qty,0) AS Qty
FROM CTE c
)ct
RIGHT JOIN
(
SELECT -- This code is to get month data
CONVERT(VARCHAR(10),'2018-'+ RIGHT('00'+CAST(MONTH(DATEADD(MM, s.number, CONVERT(DATETIME, 0)))AS VARCHAR),2) +'-01',120) AS [MonthDays]
FROM master.dbo.spt_values s
WHERE [type] = 'P' AND s.number BETWEEN 0 AND 11
)DT
ON dt.[MonthDays] = ct.[Month]
SELECT
MAX(Product)OVER(ORDER BY [MonthDays])AS Product,
[MonthDays],
ISNULL(Qty,0) Qty,
SUM(ISNULL(Qty,0))OVER(ORDER BY [MonthDays]) As SumQty
FROM #Temp
Result
Product MonthDays Qty SumQty
------------------------------
A 2018-01-01 5 5
A 2018-02-01 3 8
A 2018-03-01 0 8
A 2018-04-01 0 8
A 2018-05-01 5 13
A 2018-06-01 0 13
A 2018-07-01 0 13
B 2018-08-01 10 23
B 2018-09-01 0 23
D 2018-10-01 12 35
D 2018-11-01 0 35
D 2018-12-01 0 35
First of all, i would divide month and year to get easier with statistics.
I will give you an example query, not based on your table but still helpful.
--here i create the table that will be used as calendar
Create Table MA_MonthYears (
Month int not null ,
year int not null
PRIMARY KEY ( month, year) )
--/////////////////
-- here i'm creating a procedure to fill the ma_monthyears table
declare #month as int
declare #year as int
set #month = 1
set #year = 2015
while ( #year != 2099 )
begin
insert into MA_MonthYears(Month, year)
select #month, #year
if #month < 12
set #month=#month+1
else
set #month=1
if #month = 1
set #year = #year + 1
end
--/////////////////
--here you are the possible result you are looking for
select SUM(Ma_saledocdetail.taxableamount) as Sold, MA_MonthYears.month , MA_MonthYears.year , item
from MA_MonthYears left outer join MA_SaleDocDetail on year(MA_SaleDocDetail.DocumentDate) = MA_MonthYears.year
and Month(ma_saledocdetail.documentdate) = MA_MonthYears.Month
group by MA_SaleDocDetail.Item, MA_MonthYears.year , MA_MonthYears.month
order by MA_MonthYears.year , MA_MonthYears.month

Select min/max dates for periods that don't intersect

Example! I have a table with 4 columns. date format dd.MM.yy
id ban start end
1 1 01.01.15 31.12.18
1 1 02.02.15 31.12.18
1 1 05.04.15 31.12.17
In this case dates from rows 2 and 3 are included in dates from row 1
1 1 02.04.19 31.12.20
1 1 05.05.19 31.12.20
In this case dates from row 5 are included in dates from rows 4. Basically we have 2 periods that don't intersect.
01.01.15 31.12.18
and
02.04.19 31.12.20
Situation where a date starts in one period and ends in another are impossible. The end result should look like this
1 1 01.01.15 31.12.18
1 1 02.04.19 31.12.20
I tried using analitical functions(LAG)
select id
, ban
, case
when start >= nvl(lag(start) over (partition by id, ban order by start, end asc), start)
and end <= nvl(lag(end) over (partition by id, ban order by start, end asc), end)
then nvl(lag(start) over (partition by id, ban order by start, end asc), start)
else start
end as start
, case
when start >= nvl(lag(start) over (partition by id, ban order by start, end asc), start)
and end <= nvl(lag(end) over (partition by id, ban order by start, end asc), end)
then nvl(lag(end) over (partition by id, ban order by start, end asc), end)
else end
end as end
from table
Where I order rows and if current dates are included in previous I replace them. It works if I have just 2 rows. For example this
1 1 08.09.15 31.12.99
1 1 31.12.15 31.12.99
turns into this
1 1 08.09.15 31.12.99
1 1 08.09.15 31.12.99
which I can then group by all fields and get what I want, but if there are more
1 2 13.11.15 31.12.99
1 2 31.12.15 31.12.99
1 2 16.06.15 31.12.99
I get
1 2 16.06.15 31.12.99
1 2 16.06.15 31.12.99
1 2 13.11.15 31.12.99
I understand why this happens, but how do I work around it? Running the query multiple times is not an option.
This query looks promising:
-- test data
with t(id, ban, dtstart, dtend) as (
select 1, 1, date '2015-01-01', date '2015-03-31' from dual union all
select 1, 1, date '2015-02-02', date '2015-03-31' from dual union all
select 1, 1, date '2015-03-15', date '2015-03-31' from dual union all
select 1, 1, date '2015-08-05', date '2015-12-31' from dual union all
select 1, 2, date '2015-01-01', date '2016-12-31' from dual union all
select 2, 1, date '2016-01-01', date '2017-12-31' from dual),
-- end of test data
step1 as (select id, ban, dt, to_number(inout) direction
from t unpivot (dt for inout in (dtstart as '1', dtend as '-1'))),
step2 as (select distinct id, ban, dt, direction,
sum(direction) over (partition by id, ban order by dt) sm
from step1),
step3 as (select id, ban, direction, dt dt1,
lead(dt) over (partition by id, ban order by dt) dt2
from step2
where (direction = 1 and sm = 1) or (direction = -1 and sm = 0) )
select id, ban, dt1, dt2
from step3 where direction = 1 order by id, ban, dt1
step1 - unpivot dates and assign 1 for start date, -1 for end
date (column direction)
step2 - add cumulative sum for direction
step3 - filter only interesting dates, pivot second date using lead()
You can shorten this syntax, I divided it to steps to show what's going on.
Result:
ID BAN DT1 DT2
------ ---------- ----------- -----------
1 1 2015-01-01 2015-03-31
1 1 2015-08-05 2015-12-31
1 2 2015-01-01 2016-12-31
2 1 2016-01-01 2017-12-31
I assumed that for different (ID, BAN) we have to make calculations separately. If not - change partitioning and ordering in sum() and lead().
Pivot and unpivot works in Oracle 11 and later, for earlier versions you need case when.
BTW - START is reserved word in Oracle so in my example I changed slightly column names.
I like to do this by identifying the period starts, then doing a cumulative sum to define the group, and a final aggregation:
select id, ban, min(start), max(end)
from (select t.*, sum(start_flag) over (partition by id, bin order by start) as grp
from (select t.*,
(case when exists (select 1
from t t2
where t2.id = t.id and t2.ban = t.ban and
t.start <= t2.end and t.end >= t2.start and
t.start <> t2.start and t.end <> t2.end
)
then 0 else 1
end) as start_flag
from t
) t
) t
group by id, ban, grp;