> The problem I see with that is that any value of accountnumber in > public.commontable would need to be in both company1.chartoffaccounts and > company2.chartoffaccounts. One key referencing two completely sets of > data? That sounds broken. Can you give a more detailed example of how > you want to use this?
public.commontable describes rules how invoices are posted to general ledger. It seems not reasonable to duplicate this table in every company schema since it contains usually same data for every company. public.commontable contians fixed account numbers which I think I must force to be present in chart of accounts in every company schema to use this kind of foreign key relationship. > Off hand, it sounds like what you may want is to have accountnumber as the > primary key of public.commontable with the accountnumber columns in the > schema specific tables referencing it instead. public commontable contains also a lot of other columns describing invoice posting rules. accountnumber cannot be primary key of public.commontable since it can contain same account numbers for different types of documents. Yes, it seems to be possible to create a new table public.commonaccountnumbers with accountnumber as primary key. So the question seems to be: can I use duplicate foreign keys without problems or must I waste resources to create new table commonaccountnumbers ? Andrus. ---------------------------(end of broadcast)--------------------------- TIP 6: explain analyze is your friend
