How to create a specific SQL Server stored procedure? [closed] - sql

Question

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 1 year ago.
Improve this question
I have a CONTRACTOR_SCHEDULER table. Letters in columns "Schedule" mean working time (m - 8:00-20:00, n - 20:00-8:00(next day), d - 8:00-8:00(next day), h-day off). (Name, Begin_date) is unique.
Name
Schedule
Begin_date
End_date
John
nhmh
2019-01-01
2019-01-08
John
nnh
2019-01-09
2019-01-25
Kate
dh
2019-01-01
2019-01-07
Kate
mnhdh
2019-01-08
2019-01-14
Mike
nh
2019-01-01
2019-02-01
Mike
mh
2019-02-02
2019-12-31
I need a SQL Server stored procedure that creates new table CONTRACTOR_WORK_DAY of working days using CONTRACTOR_SCHEDULER rows (days off do not appear in table).
Example:
First row - John has schedule nhm. First letter is n so begin_date - 2019-01-01 20:00, End_date - 2019-01-02 08:00. Next letter is h, skip it as a day off. Last letter is m so begin_date - 2019-01-03 08:00, End_date - 2019-01-03 20:00. Repeat schedule until 2019-01-08 - end_date in first table.
Table for the first row of CONTRACTOR_SCHEDULER would be:
Name
Begin_date
End_date
John
2019-01-01 20:00
2019-01-02 08:00
John
2019-01-03 08:00
2019-01-03 20:00
John
2019-01-05 20:00
2019-01-06 08:00
John
2019-01-08 08:00
2019-01-08 20:00
I wrote this in python using some loop over schedule string etc. but can not figure out how to do it in T-SQL for SQL Server.

Answer 1

Yet another option.
Not sure if I agree with the last record of the desired results. I have 2019-01-07 while you have 2019-01-08
Example or dbFiddle
Select A.[Name]
,[Begin_Date] = convert(datetime,left(dateadd(DAY,N ,[Begin_date]),10)+' '+BegTime)
,[End_Date] = convert(datetime,left(dateadd(DAY,N+NxtDay,[Begin_date]),10)+' '+EndTime)
From YourTable A
Cross Apply ( values ( left(replicate(Schedule,10),datediff(DAY,Begin_Date,End_Date)) ) )B(S)
Cross Apply (
Select N=N-1
,Subs=substring(B.S,N,1)
From ( Select Top (len(S)+1) N=Row_Number() Over (Order By (Select Null)) From master..spt_values n1 ) B1
) C
Join ( values ('m','08:00','20:00',0)
,('n','20:00','08:00',1)
,('d','08:00','08:00',1)
) D(SchdCd,BegTime,EndTime,NxtDay) on Subs=SchdCd
Order By [Begin_Date]
Results
Name Begin_Date End_Date
John 2019-01-01 20:00:00.000 2019-01-02 08:00:00.000
John 2019-01-03 08:00:00.000 2019-01-03 20:00:00.000
John 2019-01-05 20:00:00.000 2019-01-06 08:00:00.000
John 2019-01-07 08:00:00.000 2019-01-07 20:00:00.000

Answer 2

WITH
numbers AS
(
-- Generate a table of values (from 0 upwards)
-- Must be at least as long as the longest schedule string
SELECT
ROW_NUMBER() OVER (ORDER BY sv.number) - 1 AS id
FROM
master..spt_values sv
),
pivotted AS
(
-- Create one row for each day in the date range
-- Extract the relevant character from the schedule for each day
SELECT
c.*,
n.id AS date_offset,
SUBSTRING(c.schedule, (n.id % LEN(c.schedule)) + 1, 1) AS schedule_char
FROM
CONTRACTOR_SHERULER c
INNER JOIN
numbers n
ON n.id <= DATEDIFF(DAY, c.begin_date, c.end_date)
)
-- Add a number of days to the begin_date
-- Then add a number of hours based on the current character from the schedule
SELECT
pivotted.*,
DATEADD(
HOUR,
CASE pivotted.schedule_char WHEN 'm' THEN 8
WHEN 'n' THEN 20
WHEN 'd' THEN 8 END,
DATEADD(DAY, date_offset, pivotted.begin_date)
)
AS begin_datetime,
DATEADD(
HOUR,
CASE pivotted.schedule_char WHEN 'm' THEN 20
WHEN 'n' THEN 32
WHEN 'd' THEN 32 END,
DATEADD(DAY, date_offset, pivotted.begin_date)
)
AS end_datetime
FROM
pivotted
WHERE
pivotted.schedule_char <> 'h'
Demo : https://dbfiddle.uk/?rdbms=sqlserver_2019&fiddle=f6c52e6718329f2ae013dc93edad5d78