It's the first time here so I don't know if it's the right place to ask.. I'm not good in database design, just learning, so I'd like some advice if possible.
I'm building a really small application, a plugin, who's tables will likely have not too much records in real life (less than 50 in extreme situations), so performance is not crucial.
The user can create n Groups. A Group should have a 'hook' field.
For each Group, the user can create n BannerBoxes. So, a Group can have many BannerBoxes, a BannerBox can have only one group. Therefore a 1-many relation.
There can be n languages, and the Language table is already in place and has id_lang field. The language-group relation is 1 (language)-many (groups)
I'll often have to ask the db for all BannerBoxes saved for a certain language, grouped by Group. So, first find all Groups of a language, then for each Group, find its BanneBoxes.
I've thought about 2 solutions, but I cannot decide which could be better, and since I'm definitly a newbe, both could be just wrong.
a. Group entity with [hook, id_lang] fields; BannerBox entity with an [hook] field (desides the others). So when selecting, I can join the Group table with Language table on id_lang and then join with BannerBoxes on hook field.
b. Only the BannerBox entity, in this case with both [hook, id_lang] fields. I could join with Language by id_lang and then group by hook.
I also guess that a Group_BannerBox table [id_group, id_bannerbox] would be useless since I think it's a solution for many-many relations, am I right?
So, what would you do? a. or b.?