Hello, list. First post here, I think. So I've worked with several *lousy* DB engines and *poorly* designed databases behind crappy applications over these many years. I look forward to moving everything to PG. First issue is addressed.
Now with the grand opportunity to be hoist upon my own petard on issues 2 and 3, DB design-wise, I have a question. Or several, but I'll try to stick to one. ;-) I have a situation where a location may belong to a larger corporation, for instance. In some cases, they may belong to more than one corporation. Corporation is someone we-wee can bill or pay. Everyone else is not belonging to this set. There's probably a name for this kind of model and it's probably trivial but I'm always in the dark when it comes to CS terms. Anyway, my thought is to make a table with two columns (loc_id, corp_id) and unique ID would be the combination of the two. Is that about right? I guess the question is answered by knowing how I need to use the information..?? Well, to find possible 'billable' locations, i.e. those with a 'corp' ID (this ID relates to a certain structure of charges, for example). The corp table will have a default location for AR/AP, which could be the same ID as the location searched for. Presenting a corp should give a list of possibly multiple locations. And so on… So, pretty much I would use a join with this table depending on how I want to use the data..?? I'm trying to be flexible but combine four different tables of addresses, keeping the functions they represent. Or, like that. Anyway, I'm just asking generally if I could be painting myself into a corner here. And hoping I'll eventually find a way to ask the right questions. TIA for any help or suggestions. '-) Cheers, -- |\ /| | | ~ ~ | \/ | |---| `|` ? | |ichael | |iggins \^ / michael.higgins[at]evolone[dot]org -- Sent via pgsql-sql mailing list (pgsql-sql@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-sql
