What is the proper way to fill a form with multiple data fields? - vb.net

Question

I have been asked to create a vb.net WinForms app that automates the creation of this spreadsheet:
Obviously, creating that spreadsheet manually is a back breaking task.
I am very comfortable writing SQL to pull each individual field of data, however I've got to believe that there is a better way. Such a method would require 144 queries !!
I've considered creating a form with individual labels for each field, or using a gridview. Either would be acceptable to the end user, but I'm really not sure how I would write a query to produce a dataset that look like the end product anyway.
I'm not asking anyone to write any code for me, what I'm asking for help with is the concept of how I should attack this task.
Here is a query that I wrote which returns the fields for Business Unit #1 for the first week on the spreadsheet. I don't think it helps much, but I'm adding it to show that I've put some effort into this.
DECLARE #WeekEnding AS DATE
DECLARE #DIV AS INT
SET #WeekEnding = '11/13/2015'
SET #DIV = 20
-- A/R downpayments OPCH
SELECT
SUM(OPCH.DocTotal - OPCH.VatSum - OPCH.TotalExpns) AS 'ARDownPaymentInvoice'
FROM OPCH
LEFT JOIN INV1 ON OPCH.DocEntry = INV1.DocEntry AND INV1.VisOrder = 0
LEFT JOIN OITM ON INV1.ItemCode = OITM.ItemCode
LEFT JOIN OITB ON OITM.ItmsGrpCod = OITB.ItmsGrpCod
LEFT JOIN OACT AS CurrCode (NOLOCK) ON OITB.RevenuesAc = CurrCode.AcctCode
WHERE DATEPART(WEEK, OPCH.DocDate) = DATEPART(WEEK, #WeekEnding) AND YEAR(OPCH.DocDate) = YEAR(#WeekEnding)
AND CurrCode.Segment_4 = #DIV
-- Credits ORIN
SELECT
SUM(ORIN.DocTotal - ORIN.VatSum - ORIN.TotalExpns) * -1 AS 'Credit'
FROM ORIN
LEFT JOIN INV1 ON ORIN.DocEntry = INV1.DocEntry AND INV1.VisOrder = 0
LEFT JOIN OITM ON INV1.ItemCode = OITM.ItemCode
LEFT JOIN OITB ON OITM.ItmsGrpCod = OITB.ItmsGrpCod
LEFT JOIN OACT AS CurrCode (NOLOCK) ON OITB.RevenuesAc = CurrCode.AcctCode
WHERE DATEPART(WEEK, ORIN.DocDate) = DATEPART(WEEK, #WeekEnding) AND YEAR(ORIN.DocDate) = YEAR(#WeekEnding)
AND CurrCode.Segment_4 = #DIV
--Invoices
SELECT
SUM(OINV.DocTotal - OINV.VatSum - OINV.TotalExpns) AS 'Invoice'
FROM OINV
LEFT JOIN INV1 ON OINV.DocEntry = INV1.DocEntry AND INV1.VisOrder = 0
LEFT JOIN OITM ON INV1.ItemCode = OITM.ItemCode
LEFT JOIN OITB ON OITM.ItmsGrpCod = OITB.ItmsGrpCod
LEFT JOIN OACT AS CurrCode (NOLOCK) ON OITB.RevenuesAc = CurrCode.AcctCode
WHERE DATEPART(WEEK, OINV.DocDate) = DATEPART(WEEK, #WeekEnding) AND YEAR(OINV.DocDate) = YEAR(#WeekEnding)
AND CurrCode.Segment_4 = #DIV`
Thanks for any ideas.

Answer 1

I would do a pivot rather than doing multiple queries. However I think you should ask yourself or the end user: "Do you really want to see an aging report by date as columns?" If so do I add one manually each time the report ages another increment? Here is a self extracting example FYI. The first set is just displaying my data where I mock up a user, a department they belong to, and their last access. Then I pretend I want a report of aging to see how many people access the system in total, and by each fictitious department. Doing multiple queries in SQL is expensive and can often be alleviated with a little know how of pivots, unions, and holding your data creatively with SQL.
DECLARE #Example TABLE (PersonId INT IDENTITY, PersonName VARCHAR(128), DepartmentName VARCHAR(128), AccessedOn Date);
INSERT INTO #Example (PersonName, DepartmentName, AccessedOn) VALUES ('Brett', 'Dev', '1-1-2017'), ('John', 'Dev', '1-6-2017'), ('Mark', 'Dev', '1-8-2017'), ('Shawn', 'Ops', '1-15-2017'), ('Ryan', 'Ops', '1-16-2017'), ('Kevin', 'Ops', '1-21-2017');
--Data as is
SELECT *
From #Example
--I would use Date By row as it is far easier to maintain
SELECT
DATEADD(WEEK, DATEDIFF(WEEK, 0, AccessedOn), 0) AS Grouping
, COUNT(PersonId) AS PeopleAccessedAtTime
, SUM(CASE WHEN DepartmentName = 'Dev' THEN 1 ELSE 0 END) AS DevCounts
, SUM(CASE WHEN DepartmentName = 'Ops' THEN 1 ELSE 0 END) AS OpsCounts
FROM #Example
GROUP BY DATEADD(WEEK, DATEDIFF(WEEK, 0, AccessedOn), 0)
--Aging Report you are asking for trouble as you need to manually pivot ELSE go down the road of doing dynamic sql to do a pivot which is No Bueno
; WITH x AS
(
SELECT
DATEADD(WEEK, DATEDIFF(WEEK, 0, AccessedOn), 0) AS Grouping
, COUNT(PersonId) AS PeopleAccessedAtTime
, SUM(CASE WHEN DepartmentName = 'Dev' THEN 1 ELSE 0 END) AS DevCounts
, SUM(CASE WHEN DepartmentName = 'Ops' THEN 1 ELSE 0 END) AS OpsCounts
FROM #Example
GROUP BY DATEADD(WEEK, DATEDIFF(WEEK, 0, AccessedOn), 0)
)
Select
'PeopleAccessed' AS Header
, MAX([2017-01-02]) AS [2017-01-02]
, Max([2017-01-09]) AS [2017-01-09]
, Max([2017-01-16]) AS [2017-01-16]
From x
PIVOT(MAX(PeopleAccessedAtTime) FOR Grouping IN ([2017-01-02], [2017-01-09], [2017-01-16])) AS pvt
UNION
Select
'DevDivisionAccessed'
, MAX([2017-01-02]) AS [2017-01-02]
, Max([2017-01-09]) AS [2017-01-09]
, Max([2017-01-16]) AS [2017-01-16]
From x
PIVOT(MAX(DevCounts) FOR Grouping IN ([2017-01-02], [2017-01-09], [2017-01-16])) AS pvt
UNION
Select
'OpsDivisionAccessed'
, MAX([2017-01-02]) AS [2017-01-02]
, Max([2017-01-09]) AS [2017-01-09]
, Max([2017-01-16]) AS [2017-01-16]
From x
PIVOT(MAX(OpsCounts) FOR Grouping IN ([2017-01-02], [2017-01-09], [2017-01-16])) AS pvt

Answer 2

One way to achieve this:
CREATE TABLE #tmpData
(column1 varchar(50),
column2 varchar(50),
column3 varchar(50),
column4 varchar(50),
column5 varchar(50),
column6 varchar(50),
column7 varchar(50),
column8 varchar(50),
column9 varchar(50),
column10 varchar(50),
column11 varchar(50),
column12 varchar(50))
Then you could insert a row displaying the dates of each column
INSERT INTO #tmpData
--Write your select query here for filling in all the dates
Then you would insert the data for each line item in the rows.
INSERT INTO #tmpdata
SELECT 'Business Unit 1',
-- insert remaining column data here