Thanks Sean - I guess my initial (negative) reaction to a single database was that "it's different" from what I am used to doing. But there does not appear to be a reason NOT to use a single database at the moment. Now to go and move stuff out of the public schema.... which several different sites recommended.
Mark
On Wed, Nov 19, 2008 at 1:17 PM, Mark Wimer <mwimer@usgs.gov> wrote:
In MSSQL, I have a dozen or so databases that rely on some shared tables (states, species lists, for example). I put the shared tables in a db called Common. To query them, mssql allows:
select statename, stateabbrev from common.dbo.statelist S inner join mytable on S.stateabbrev = mytable.stateabbrev
I'm not asking for PostgreSQL to be more like mssql, but I would like to share some lookup tables among applications & databases. In looking through the help, it appears from the "Schemas" help page (http://www.postgresql.org/docs/8.3/static/ddl-schemas.html) that PostgreSQL uses a single database per connection (pgAdmin client, web client). Is there a best practice or recommendation page for how to handle this sharing of a common lookup table? Or a help page I missed that discusses it?
Right now I see my options as: a) put all applications for which I use PostgreSQL into one database, but different schemas.
The standard solution is (a).
b) use a client tool to make a separate connection and create a join afterwards
Postgres has other options. You can look at the dblink contrib module to connect between postgres databases on the same server. Also, a project (perl-based) called DBI-Link allows you to query from other databases (mysql, mssql, etc.) directly from within postgresql.
If you have total control of the database, though, I would say stick with (a).