Databases: Quick Question

Endurance Man

Right, just a quick question here with regards to the whole normalisation thing, its becoming the bain of my life.
Here is one entity and its set of attributes:

Project ( P_code, P_type, P_name, Start_date, End_date, No_contractors, Cust_num)

Cust_num is the foreign key. Now are there any repeating groups in these attributes . If someone could explain this one im sure i could figure the rest out.

Any help muchio appreciated lads
Endo

scojones

I haven't done this in so long so do not assume what I say is right, because chances are it is not.

Project( P_code, Contractor.Con_num, Customer.Cust_num)

Project_details( Project.P_code, P_type, P_name, Start_date, End_date)

Contractor(Con_num, Con_fname, Con_lname, Con_rank)

Customer(Cust_num, Cust_fname, Cust_lname)

Does that make any sense?

Check this out http://www.serverwatch.com/tutorials/article.php/1549781

Civilian_Target

Endurance Man wrote:

Project ( P_code, P_type, P_name, Start_date, End_date, No_contractors, Cust_num)

Sounds to me like a college exercise, reminds me of Data Processing at college.

How you normalise is a matter of opinion, you have to minimise the possibilities of putting in the same data twice, but there's a form called 3NF which is the most sensible and is quite achieveable using formulaic methods, but in the real world, intuition works better.

I would say:
Project(P_code,P_typeID,P_propsID,Cust_num)
ProjectType(typeID,P_type,...)
ProjectProps(propsID,P_name,Start_date,End_date,No_contractors)

However, like sjones mentioned, you could have a seperate table for contractors. Note that in Project TypeID and PropsID are foreign keys connecting to typeID and propsID which are primary keys in their relevant tables