Table/Column optimisation enquiry

[SupaMonkey]

Ok, so I am gonna jack up my product catalogue script a bit as it is getting slightly outdated. One of the new features is "Other related products"

Now, my question is, which would be better:
1) Adding a column to a products "related_products" and just put csv ID's of other products in there and just explode,etc. whenever i need them
2) Create a new table and create a new record for every related product for each product
3) Something else you guys may think of?



My worthless-based-on-nothing impressions are:
1) Would be faster, more efficient
2) Would be easier to make initially, but slower/less efficient

Any takes on this or any suggestions?

[Barand]

No contest - use a separate table.

If you have a csv field of id's you can't join those ids to the product file to get descriptions etc of the related products.

[code]
product              related_prods
----------           --------------
prod_id     <----+    rec_id
description      |
price            +--- prod_id
                 |
                 +--- rel_prod_id
[/code]

[SupaMonkey]

Couldnt I just retrieve the CSV column then explode it, then loop through it and get each id's data with a query?

[Barand]

Yes. You could go the whole hog, uninstall MySql and just use a text file.