[RESOLVED] How should I design this?

[Piranha]

I am building a database to store code that are commonly used. One part of it is that I want to be able to see codes that do simular things. For example I can have a function that checks if a date is correct, and it takes the date as yymmdd. Another one can use yyddmm, yet another yyyymmdd and the last one yyyyddmm. I would like to be able to set all of them as simular to each other. The first database structure I thought of was this:

table code
id
code
// other variables

table simular
id
// no other variables

table simular_code
code_id
simular_id

I would then make a new simular and add all code_id and that simular_id to the simular_code table to make the codes simular. But do I really need the simular table? It will only hold an id and no description, would it be better to drop that table since it will basically be of no use?

[leatherback]

If your table similar only hold one colum it is pretty much useless, I'd say. What would it do?

Think about how you define similar. Will it just be a grouping of functionalities? In that case you could have a lookup table functional types:

<ft_id><ft_label>
[1][date checking]
[2][email validation]
[3][array sorting]

and a table with your code, as defined in your post

and in between a table holding the relations (Assuming a many-tomay relation: Every snippet can belong to multiple functional types, and every type can contain multiple snippets):
<snippet_nr><type_nr>

[Piranha]

Ty leatherback. I was not sure if I should use a name/label, but it seems like it is a good idea. And since I will do that there basically are only one way to do it 😉