Simulating For inside For in SQL - sql

Question

I'm using HP Vertica 7 as data warehouse DB.
I need to update two tables at the same time.
foreach (row r in table_1)
{
foreach (row r2 in table_2)
{
if(r.key1 == r2.key1 && r.key2 ==r2.key2 && r.soldQuantity > r2.Quantity)
{
updateRowInT1(); //r.Partner = r2.Partner;
updateRowInT2(); //r2.Quantity = r2.Quantity - r.soldQuantity;
break;
}
}
}
This part of code best describes what I need to do.
Is there any way to do this in SQL (with user-defined functions).
Because of changes in second table I can't use update and join. Or I can?
Also, this is part of ETL process done by Pentaho Kettle. Is perhaps way to do this inside this tool.?
Thanks
Edit:
I have changed code above.
What I need to achieve?
Regular Sales by Partner.
What's wrong right now?
When some SKU is sold in retail shop, I can see SKU LAST Partner. But, company I work for is buying same SKU from more Partners.
Slowly Changing Dimension is not applicable because when we switch partner on SKU, there is still some quantity of that SKU on shelf.
Assumption:
FIFO - first In first Out.
So, for initial matching, I need to compare (in date order) both sales and buying and to decrement bought quantity.

Answer 1

You probably should be thinking in a more set oriented way.
Assuming I understand what you are trying to do properly, I would do this in two update statements:
-- Update table_1's Partner based on a row in table_2.
UPDATE table_1 r
SET Partner = r2.Partner
FROM table_2 r2
WHERE r2.key1 = r.key1
AND r2.key2 = r.key2;
-- Reduce the Quantity in table_2 based on table_1's soldQuantity
UPDATE table_2 r2
SET Quantity = r2.Quantity - r.soldQuantity
FROM table_1 r
WHERE r.key1 = r2.key1
AND r.key2 = r2.key2
AND r2.Quantity >= r.soldQuantity;
-- Both of these should be done in a transaction so it is all or none. Commit the work.
COMMIT;
You can see the SQLFiddle here.