# SQL - How distribute x amount [of money] between multiple rows

Hello everyone,

This is a Microsoft SQL Server 2005 question:

Suppose I have the following 2 tables:

``````create table tblPurchaseOrders (
[PoNumber] varchar(20),
[PODate] datetime,
[POAmount] money,
[AllocatedAmount] money
)

create table tblAllocatedAmounts (
[PoNumber] varchar(20),
[AllocatedAmount] money
)

insert into tblPurchaseOrders values ('ABC', '2012-04-01 11:30:00', 1000, 0)
insert into tblPurchaseOrders values ('ABC', '2012-04-03 13:30:00', 500, 0)
insert into tblPurchaseOrders values ('DEF', '2012-04-02 12:00:00', 2000, 0)
insert into tblPurchaseOrders values ('GHI', '2012-04-03 12:15:00', 300, 0)

insert into tblAllocatedAmounts values ('ABC', 500)
insert into tblAllocatedAmounts values ('DEF', 1000)
insert into tblAllocatedAmounts values ('GHI', 300)
insert into tblAllocatedAmounts values ('ABC', 500)
insert into tblAllocatedAmounts values ('ABC', 100)
``````

How do I distribute the sum of tblAllocatedAmounts.AllocatedAmount to tblPurchaseOrders.AllocatedAmount based on PO Number and PO Date?

This is the desired output:

``````select * from tblPurchaseOrders order by ponumber, PODate asc

[PoNumber]    [PODate]                     [POAmount]  [AllocatedAmount]
ABC          2012-04-01 11:30:00.000        1000.00        1000.00
ABC          2012-04-03 13:30:00.000        500.00          100.00
DEF          2012-04-02 12:00:00.000        2000.00        1000.00
GHI          2012-04-03 12:15:00.000        300.00          300.00
``````

Thank you!
LVL 13
###### Who is Participating?

x
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Commented:
``````select PoNumber, PODate, POAmount,
case
when AllocatedAmount >= AccAmount then AccAmount
when AllocatedAmount < AccAmount then POAmount - AccAmount + AllocatedAmount
else 0
end as AllocatedAmount
from (
select o.PoNumber, o.PODate, o.POAmount, a.AllocatedAmount,
(select sum(i.poamount) from tblPurchaseOrders i where i.ponumber = o.ponumber and i.id <= o.id) AccAmount
from tblPurchaseOrders o
join (select PONumber, Sum(AllocatedAmount) AllocatedAmount from tblAllocatedAmounts group by PONumber) a on a.PONumber = o.PONumber) t
order by ponumber, PODate asc
``````
0
Commented:
Here's my go at it:

``````;with CTE as
(select poNumber, sum(AllocatedAmount) Allocated
from tblAllocatedAmounts group by poNumber)
update tblPurchaseOrders
set AllocatedAmount =
case
when
cte.allocated -
isnull((select sum(POAmount) from tblPurchaseOrders p
where p.PoNumber = PO.PoNumber
and p.PODate < PO.PODate),0)
>= PO.POAmount
then PO.POAmount
else
cte.allocated -
isnull((select sum(POAmount) from tblPurchaseOrders p
where p.PoNumber = PO.PoNumber
and p.PODate < PO.PODate),0)
end
from tblPurchaseOrders PO inner join CTE on CTE.PoNumber = PO.PONumber
``````
0
Author Commented:
wdosanjos,

I removed the "and i.id <= o.id" on line 9 because there there is no column named "id".
This is giving me:

[PoNumber]    [PODate]                     [POAmount]  [AllocatedAmount]
ABC          2012-04-01 11:30:00.000        1000.00        600.00   <==== Should be 1,000 here ...
ABC          2012-04-03 13:30:00.000        500.00          100.00
DEF          2012-04-02 12:00:00.000        2000.00        1000.00
GHI          2012-04-03 12:15:00.000        300.00          300.00
0
Commented:
p.s.  Sometimes we get questions that are a bit contrived for the sake of simplification or privacy.  I hope that is the case here, because your tables lack the keys necessary for entity integrity.

Isn't it out of the ordinary to have the same PO on two different dates?  I should also point out that the logic fails if you get two rows for the same PO on the same date.  Note, I believe wdosanjos's introduction of an ID column is an attempt to circumvent some of these issues.
0
Author Commented:
dqmq,

It seems like your solution is going to work for me.
Let me check...
0
Author Commented:
@37825395:

>> your tables lack the keys necessary for entity integrity

Sorry, I was in a bit of a hurry... : )
0
Commented:
Here is the corrected query. Sorry, I was experimenting a couple things and I forgot to clean up the query.
``````select PoNumber, PODate, POAmount,
case
when AllocatedAmount >= AccAmount then AccAmount
when AllocatedAmount < AccAmount then POAmount - AccAmount + AllocatedAmount
else 0
end as AllocatedAmount
from (
select o.PoNumber, o.PODate, o.POAmount, a.AllocatedAmount,
(select sum(i.poamount) from tblPurchaseOrders i where i.ponumber = o.ponumber and i.PODate <= o.PODate) AccAmount
from tblPurchaseOrders o
join (select PONumber, Sum(AllocatedAmount) AllocatedAmount from tblAllocatedAmounts group by PONumber) a on a.PONumber = o.PONumber) t
``````
0

Experts Exchange Solution brought to you by