From: | Rory Campbell-Lange <rory(at)campbell-lange(dot)net> |
---|---|
To: | Ron St-Pierre <rstpierre(at)syscor(dot)com> |
Cc: | jonathon(at)octahedron(dot)com(dot)au, pgsql-novice <pgsql-novice(at)postgresql(dot)org> |
Subject: | Re: Multi Database queries |
Date: | 2004-09-22 21:24:59 |
Message-ID: | 20040922212459.GB26432@campbell-lange.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-novice |
On 22/09/04, Ron St-Pierre (rstpierre(at)syscor(dot)com) wrote:
> >Yes, but, if your databases are all on one server under one postmaster
> >you should consider using one database with multiple schemas. See:
> > http://www.postgresql.org/docs/current/static/sql-createschema.html
>
> We're using this for one of our projects and it really works well.
> Initially we had separate databases and queried them using dblink, but
> for our purposes multiple schemas works much better, and cross schema
> queries are as simple as any other queries. BTW if you are looking at
> setting up a new system, why not use 7.4.5 (or whatever the latest
> stable version is)?
I concur. Using schemas works very well. Another benefit is that you can
load your plpgsql functions into a schema and set your search path to
see it or perhaps another schema loaded with test functions.
eg "set search_path = data_schema, function_schema, global_schema;"
versus
eg "set search_path = data_schema, test_function_schema, global_schema;"
This can be a great way of testing your new functions against existing
data.
Rory
From | Date | Subject | |
---|---|---|---|
Next Message | Christian Armeanu | 2004-09-22 23:26:47 | Using PERFORM in plpgsql throws compile errors |
Previous Message | Kumar S | 2004-09-22 18:01:41 | DATE Error: COPY |