From: | Frank Hagstrom <frank(dot)hagstrom(at)gmail(dot)com> |
---|---|
To: | pgsql-sql(at)postgresql(dot)org |
Subject: | Possible to use a table to tell what table to select from? |
Date: | 2005-07-08 20:32:31 |
Message-ID: | 28b4448e05070813326d98e574@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
Hello
I've been thinking on a potential problem I might get in a distant
future, but once I started thinking on it I just as well had to
check...
Is it possible to have say ~ 6 tables, where the first one is a
'reference' to all the others (they all are with the same structure).
The other 5 or so have information stored in importance/access/age,
where one table will be only recent/important things, the other with a
bit lower priority and so on until the last table that can be really
huge and slow and used more like a storage ... Is it possible to write
an SQL query to let the first table figure out what table one will do
the actual select from?
like the pseudo code:
SELECT name,content FROM <some command> (SELECT tablename FROM table1
WHERE item_id="123456")
Or is this a totally pointless thing, as the first table would need to
be so large either way? (or only good for storing large blobs and
such?)
/Frank H
From | Date | Subject | |
---|---|---|---|
Next Message | CG | 2005-07-08 20:55:03 | Clustering problem |
Previous Message | Alvaro Herrera | 2005-07-08 19:51:25 | Re: Make COUNT(*) Faster? |