Structure of the DB
I am trying to structure a 开发者_JS百科database for optimal use. But I have no idea how to make that happen. Here is what i will have
A category which will have many products and a count for each on of those products
For example
Restaurants will be the category and the products will be the computer system, cash register and each one of these products has a count so maybe there will be 2 cash registers and 3 computer systems and so on. Another catagory might be a bar, which might or might not have the same products. Then the user has the option of choosing a 2 restarant package which might have a discount rate for getting all the products.
I just need to structure the ERB if i can get some help, Thanks
PRODUCTS
- PRODUCT_ID (primary key)
- PRODUCT_NAME
CATEGORIES
- CATEGORY_ID (primary key)
- CATEGORY_NAME
PRODUCT_CATEGORIES_MAP
- PRODUCT_ID (primary key, foreign key to
PRODUCTS
) - CATEGORY_ID (primary key, foreign key to
CATEGORIES
) - QUANTITY
This'll allow you to associate a product to more than one category, but not allow duplicates.
Establishment
- ID
- Name
- <address info>
Product
- ID
- Name
Estab_Prod_Bridge
- ID_Product
- ID_Establishment
- Quantity
Discounts
- ID_Establishment
- Percent_Purchase
- Discount
Note:
Establishment is a long word, I just used it here for example. You could easily replace it with location/loc, venue, or anything else you see fit.
Depending on how the data changes, I added the discount table, thinking there would be different discounts for variant amount of products purchased.
Example: 80% of products purchased would yield a 5% discount, but 100% of purchases might yield a 10% discount.
精彩评论