Re: normalization problems

From: Karen Osmond <kosmond_at_softhome.net>
Date: Thu, 21 Dec 2000 23:08:05 GMT
Message-ID: <3a428996.34700317_at_news.ntlworld.com>


>So you are looking at something like this:
>

 <snip>
>where the prod-cat table contains all of the valid product-category
>combinations, and then all of the attributes specific to just products
>or categories listed in those seperate tables, yes?

Thanks for the reply! However that's not quite what I meant, my fault for being unclear...

I'm looking at something like this

Vendor



vendor_code
(etc)

Product



product_code
vendor_code
(etc...product stats eg release date, max users)

Product_Category



product_code
table_ref

CRM



product_code
response_cycle_analysis
kpi_trending
(other crm specific fields)

OLAP



product_code
smp_parallel_processing
(other olap specific fields)

Vendor 1-n Product
Product n-1 Product_Category

Product 1-1 CRM
Product 1-1 OLAP

CRM and OLAP are two of a number of product categories which will be needed. Depending on what type of product a product is, I need to store different information in different formats, that's why I've put CRM, OLAP etc in separate tables.

Each product will fall into at least one of the tables CRM, OLAP etc.

The Product_Category table is really a workaround so that I can tell which category table I need to look in to find more data on a product. So I would be hardcoding in something like: if (tableref='OLAP') {

	SELECT * FROM OLAP...
	} 

etc

Although I can implement what I need to with the layout above, it seems like an awful solution! Any ideas? Afraid I've only worked with 'classic' library/stock style databases before, and this has got me stumped!

-Karen. Received on Fri Dec 22 2000 - 00:08:05 CET

Original text of this message