I could only find one previous post that didn't explain it for me so here it is.
It's a batch process aimed at creating a daily list of PO's per SUPPLIER to send out. Items are always sorted by supplier_id. All items without list_id are given a newly created list_id, then after gathering SQL DISTINCT on item supplier_id, new PO for each supplier_id with the list_id is created. The process is semi-transactional because a commit has to happen for the SQL part.
Is using a multi key predicate for the relationships this way instead of assigning the ITEMS a parent PO_id a good or bad idea?
At the moment the relations sort themselves automatically. For other reasons I need to show data only in List View , po unique is used as PO# for reference and has meaning ( list_id-supplier_id ).
Can anyone spot any potential pitfalls maybe.
Any comments are appreciated.