Is there a way to make two primary keys, with only one foreign key that references another tables primary key

Question

I'm new to SQL and trying to make it so that there are two primary keys and one foreign key that references other tables primary key.

I've tried adding the attribute that is a primary key that is missing the table that needs referencing and then making that a foreign key but still getting the message that "number of referencing columns must match references columns".

if there is a solution to what I'm trying to achieve it would be much appreciated.

CREATE TABLE Next_of_Kin
 (
   Employee_No       VARCHAR2(8)  NOT NULL,
   kin_Number        VARCHAR2(8)  NOT NULL,
   Name              VARCHAR2(40) NOT NULL,
   relationship      VARCHAR2(40) NOT NULL,
   contact_number    VARCHAR2(11) NOT NULL,
   PRIMARY KEY (Employee_No, Kin_Number),
   FOREIGN KEY (Employee_No, Kin_Number) REFERENCES Employee(Employee_No) 
 );

CREATE TABLE Employee 
( 
  Employee_No   VARCHAR2(8)  NOT NULL, 
  family_Name   VARCHAR2(40) NOT NULL, 
  given_Name    VARCHAR2(40) NOT NULL, 
  address       VARCHAR2(80) NOT NULL, 
  date_of_Birth DATE         NOT NULL, 
  date_Hired    DATE         NOT NULL, 
  supervisor    VARCHAR2(40) NULL, 
  PRIMARY KEY (Employee_No, Supervisor), 
  FOREIGN KEY (Employee_No,Supervisor) 
  REFERENCES Employee(Employee_No, Supervisor) 
);

Answer 1

• I don't know your company, but I'm pretty sure that the employee number is unique. Thus, the primary key of your employee table should be Employee_No , without the supervisor.

• The foreign key in your employee table does not make sense. Remove it.

• In your next of kin table, leave the primary key as it is, but make only the employee number a foreign key:

   FOREIGN KEY (Employee_No) REFERENCES Employee(Employee_No)

Answer 2

Presumably supervisor references an employee number (because the supervisor is also an employee), so it should only be one column.

Also, there's no reason why a next of kin entry should refer to another next of kin. All you need to enforce is that the employee number refers to an existing employee.

Single-column keys can be declared as part of the column definition, which simplifies the syntax and also allows the datatype of foreign key columns to be inherited from the parent.

create table employee 
( employee_no       varchar2(8)  primary key
, family_name       varchar2(40) not null
, given_name        varchar2(40) not null
, address           varchar2(80) not null
, date_of_birth     date         not null
, date_hired        date         not null
, supervisor        references employee(employee_no)
);

create table next_of_kin
( employee_no       references employee (employee_no) not null
, kin_number        varchar2(8)  not null
, name              varchar2(40) not null
, relationship      varchar2(40) not null
, contact_number    varchar2(11) not null
, primary key (employee_no, kin_number)
);

A table can have as many unique constraints as you like, but only one primary key. In your example though, you are trying to define one primary key with two columns, which is allowed (although not needed here). You can also have a foreign key with more than one column, as long as it matches a corresponding primary or unique constraint in the specified table.

Answer 3

It looks like you want to have EMPLOYEE_NO be the primary key on EMPLOYEE, but you also want (EMPLOYEE_NO, SUPERVISOR) to be unique. Note that this unique key is unnecessary - becaue EMPLOYEE_NO is the primary key on EMPLOYEE it's guaranteed to be unique, and thus any combination of (EMPLOYEE_NO, <some other column on EMPLOYEE) will automatically be unique - but sometimes I throw in a unique key like this just as documentation, or to make it referenceable in a foreign key constraint from another table.

So anyways, consider using:

 (
   Employee_No       VARCHAR2(8)  NOT NULL,
   kin_Number        VARCHAR2(8)  NOT NULL,
   Name              VARCHAR2(40) NOT NULL,
   relationship      VARCHAR2(40) NOT NULL,
   contact_number    VARCHAR2(11) NOT NULL,
   PRIMARY KEY (Employee_No, Kin_Number),
   FOREIGN KEY (Employee_No) REFERENCES Employee(Employee_No) 
 );

CREATE TABLE Employee 
( 
  Employee_No   VARCHAR2(8)  NOT NULL, 
  family_Name   VARCHAR2(40) NOT NULL, 
  given_Name    VARCHAR2(40) NOT NULL, 
  address       VARCHAR2(80) NOT NULL, 
  date_of_Birth DATE         NOT NULL, 
  date_Hired    DATE         NOT NULL, 
  supervisor    VARCHAR2(40) NULL, 
  PRIMARY KEY (Employee_No),
  UNIQUE KEY  (Employee_No, Supervisor)
);

Since (EMPLOYEE_NO, KIN_NUMBER) is the primary key on NEXT_OF_KIN it makes no sense to have a self-referencing foreign key utilizing (EMPLOYEE_NO, KIN_NUMBER) as the row would always-and-only refer to itself.