I have the following five tables:
- ISP
- Product
- Connection
- AddOn
- AddOn/Product (pivot table for many-to-many relationship).
Each Product is linked to an ISP, each Connection is listed to a Product. Each product can have a number of add-ons, through the use of the pivot table (which simply has 2 fields, one for the product ID and one for the AddOn ID).
The result I am interested in is each connection with the addons listed (I am making use of MySQL’s GROUP_CONCAT for this, to make a comma-separated list of the addon’s name field). This works fine as is, the query looks something like this:
SELECT i.name AS ispname, i.img_link, c.download, c.upload, c.monthly_price, c.link, GROUP_CONCAT(a.name) AS addons, SUM(pa.monthly_fee) AS addon_price FROM isp i JOIN product p ON i.id = p.isp_id JOIN `connection` c ON p.id = c.product_id LEFT JOIN product_addon pa ON pa.product_id = p.id AND pa.forced = 0 LEFT JOIN addon a ON pa.addon_id = a.id GROUP BY c.id I am using LEFT JOINS as it is possible for products to have no addons at all.
My problem is that it is possible to select some addons that listed connections MUST have, presented as a list of addon IDs, like (1,14,237). If I put it in as an additional condition in the JOIN statements (AND pa.addon_id IN (…)), it will return all connections that have just one of the listed addons, but not necessarily all of them.
Is there some way to return all connections that as a minimum have all the addons (they can have additional as well) via SQL?
1 Answer