Subject: | Re: ODBC to an Amazon WS Postgressql db.
| Date: | Thu, 27 Feb 2020 11:07:19 +1100
| From: | Tom Krieg <REMOVEtomkCAPITALS@sdassociates.com.au>
| Newsgroups: | pnews.paradox-client_server
|
I had a next number procedure, but then I switched over to using
PostgreSQL sequences. Much easier than rolling your own. The line number
you noticed is just a sort field to sort lines into order within an
order number. I've now started using Universal Unique IDs as primary
keys (generated by Postgres) which means you can merge the same table
from different databases or schemas without having to worry about
duplicate primary keys. The Postgres docs say that a UUID is better than
a GUID because one is global, the other encompasses the whole universe.
On 27/02/2020 1:20 am, Mark Bannister wrote:
> Thanks for the code! Makes sense.
> So when do you generate your key field numbers? Is that liLineNo in the
> code below? I realize that you have a sql built in procedure that
> generates them.
>
|