Query to match Sales Analysis output exactly

Question

Hello,
I have tried several queries to match the output of the Sales Analysis report.  I have done so by customer and by sales employee.  I have failed to create a query that always matches the outcome of the Sales Analysis query exactly.  I would greatly appreciate it if someone could provide such a query even if it just matches net A/R invoice total by date range so that it can be compared with Sales Analysis tool.  From there I can use joins to see other results, e.g., net sales by sales employee by customer.
Thank you,
Lang

Answer 1

Thanks for your response.  Yes, we need to account for credit memos and have already done so.  I had not thought to check for canceled state on invoices though, so thanks for that.  I did a quick query and there were no canceled invoices in the database, however, it will be good to incorporate regardless.  My annual net is close to being correct, but must be exact.  I don't need any other information to be output except net balance, i.e., balance after credit memos, document discounts, miscellaneous expenses recorded in freight field, and ?.  Using doctotal on documents takes into account the document discounts and miscellaneous expenses.  Another way to go is work at the line item level, e.g., INV1 or JDT1, and then account for the document discount and freight (totalexpns).  Approaches that use INV1 linetotal naturally fail on their own since they don't account for document discounts and freight expenses.  At the risk of being redundant, any approaches that don't include credit memos will naturally fail as well.  Here is a few approaches I have tried (that have been almost correct, but not exact, i.e., off by hundreds or a few thousand with net being in the single digit millions):
Approach #1:
declare @start datetime
declare @end datetime
set @start = '01/01/2008'
set @end = '12/31/2008'
select sum(t2.subtotal) from (select t0.doctotal as subtotal
from oinv t0
where t0.docdate between @start and @end
union all
select -1*t0.doctotal as subtotal
from orin t0
where t0.docdate between @start and @end) t2
Note: this approach was off by a couple hundred dollars or so.  I would like to run this query in B1, but it summarizes a union and that seems to cause B1 to not be able to handle it. It runs fine in any other SQL Server client.  I would rather run it in B1 with dynamically selectable start and end dates, via '[%0]' etc.  The same is true for Approach #2 below.
Approach #2:
declare @start datetime
declare @end datetime
set @start = u201801/02/2008u2019
set @end = u201812/31/2008u2019
Select sum(t5.subtotal) from (Select SUM as subtotal
From INV1 T0
Inner JOIN OINV T1 ON T1.DocEntry = T0.DocEntry
Where T1.DocDate BETWEEN @start AND @end
Union All
Select -SUM as subtotal
From RIN1 T0
Inner JOIN ORIN T1 ON T1.DocEntry = T0.DocEntry
Where T1.DocDate BETWEEN @start AND @end
union all
select sum(t0.totalexpns) as subtotal
from oinv t0
where t0.docdate between @start and @end union all
select sum(t0.totalexpns) as subtotal
from orin t0
where t0.docdate between @start and @end) t5
Note: the outcome of this is identical to Approach #1. 
Approach #3:
SELECT SUM as u2018Net Balanceu2019
from JDT1 T0
Inner Join OJDT T1 ON T1.TransId = T0.TransID and
(T1.TransType = u201813u2019 or T1.TransType = u201814u2019)
Inner Join OCRD T2 ON T2.CardCode = T0.ShortName
LEFT JOIN OINV T3 ON T3.TransId = T1.TransId
LEFT JOIN ORIN T4 ON T4.TransID = T1.TransId
Where T1.refdate Between u2018[%0]u2019 and u2018[%1]u2019 and T2.CardType = u2018Cu2019
Note: This query fails because it does not account for miscellaneous expenses recorded in the freight field of invoices and credit memos.  Adding those in is possible, but fails due to the 'Inner Join OCRD T2 ON T2.CardCode = T0.ShortName.  This is because in the case of freight expense journal line items the shortname does not match the cardcode of the BP, but rather the system account it references.  I haven't spent more time to rework the logic of the join to overcome this, but it could possibly be added back in via a union which is subsequently totaled.
I think I tried a few other approaches as well, but to no avail. 
A related question/annoyance: I have noticed that certain queries in B1 cannot be executed correctly (they throw an error or have strange behavior like returning only the first row returned) though they can run fine in SQL Server Manager or other SQL Server clients.  Why is that?  It is annoying and unrealistic to have clients have to use another tool for certain queries.  Examples of such queries are those involving cursors, certain summarizing selects that work on a union as input, and queries that contain certain elements of TSQL, like variable declarations (not sure about this last one, but it seems to sometimes cause the query to fail in B1).  I wish I knew how to overcome this so that all queries could run in B1.
Edited by: Lang Riley on Jan 22, 2009 8:08 PM
Edited by: Lang Riley on Jan 22, 2009 8:15 PM