From: | Ron Mayer <rm_pg(at)cheapcomplexdevices(dot)com> |
---|---|
To: | Ben <bench(at)silentmedia(dot)com> |
Cc: | PostgreSQL <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Schema design question |
Date: | 2008-03-31 07:38:48 |
Message-ID: | 47F09508.7070707@cheapcomplexdevices.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Ben wrote:
> I'm working on a project which requires me to keep track of objects,
> each of which can have an arbitrary number of attributes. Although there
> will be many attributes that an object can have,...
> Anyway, this seems like a common problem without a perfect solution, and
> I'm sure people must have hindsight opinions on how they solved it. Your
> thoughts?
If each attribute can only occur once for any object, and if your
queries are based more on exact matches of the values rather than
range queries, you might want to look into the hstore module:
http://www.postgresql.org/docs/current/static/hstore.html
If attributes can occur more than once, you might even want to
serialize the whole object as some text format (yaml, xml, etc)
instead.
From | Date | Subject | |
---|---|---|---|
Next Message | Stuart Brooks | 2008-03-31 08:03:23 | ANALYZE getting dead tuple count hopelessly wrong |
Previous Message | Allan Kamau | 2008-03-31 07:35:47 | Removing redundant itemsets |