Tracking depth question

From: "Zot O'Connor" <zot(at)zotconsulting(dot)com>
To: postgres sql <pgsql-sql(at)hub(dot)org>
Subject: Tracking depth question
Date: 1999-11-04 17:39:59
Message-ID: 3821C4EF.CE5484E@zotconsulting.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-sql

I have a shopping cart (FishCartSQL) and I have a client with many
layers of sub categories (at least 4, maybe more).

The current system only has 1 subcategory, I am extending it.

I am planning on having subcats reference themselves with a scatscat
value which will need to point to scatval (a subcat ID).

I know that foreign key does not work, and I will accept that for now,
but my issue is tracking the depth of the subcat within the DB.

I would rather not rely on external SQL (php) to do the job. The
problem is that I am not sure I can reference the depth in the table
definition. I would like something like this:

CREATE TABLE scattab
scatval integer DEFAULT NEXTVAL('scatid_SEQ') PRIMARY KEY,
scatscat integer FOREIGN KEY(scattab.scatval),
scatdepth integer DEFAULT (SELECT scatval from scattab where scatval =
scatscat) + 1,
...

Now there are a few problems that I see

1) scatscat is not defined yet
2) I don't think I can default to a select
3) I am assuming an atomic value for the SELECT statement

Is there a better Clean way?

I was thinking a trigger that fires a separate depth table and adds the
values of scatval and its depth.

Any ideas?

--
Zot O'Connor

www.ZotConsulting.com
www.WhiteKnightHackers.com

Responses

Browse pgsql-sql by date

  From Date Subject
Next Message Drinks, Ivan - ITD 1999-11-04 22:19:04 Remove
Previous Message Stoyan Genov 1999-11-04 12:59:03 Re: [SQL] Optimizing a query through its syntax and indices