c



 
 
The nature of relational database design means that we will often have related data that is
stored in different tables. To retrieve data from two or more tables in one query we use the
c
 statement.

The JOIN clause tells the database how the data in the two tables is related so that it can
return a correct representation of the related data.

The c
 clause tells the database to only return rows where there is a match
found between table1 and table2. An INNER JOIN is most often (but not always) created
between the primary key column of one table and the foreign key column of another table.

 

Let us assume we have an Employee table and a TrainingTaken table. When an employee
goes on a training course they have the details of the training recorded in the TrainingTaken
table.

Some example data is shown below:

 

     c  

1 Bob Marley 0222 00000 10/01/2005
2 John Lennon 0222 00050 02/05/2003
3 Ralph Kimball 0222 03307 01/04/2004
4 Bill Gates 0222 03307 01/04/2004

  

          

1 2 Sales Training 10/01/07
2 1 Risk Management 05/02/07
3 2 First Aid 01/03/07
4 3 Sales Training 10/01/07

The Employee table has a primary key column called r
 which relates to the
foreign key column in the TrainingTaken table called r
.

Now that we know how these two tables relate to each other we can write a query that
correctly 'joins' or 'matches' related data from these two tables. To do this we must specify in
our INNER JOIN clause, the relationship between the r
 column in the Employee
table and the r
 column in the TrainingTakentable.
Lets write a query that returns a list of employee names along with the title and date of any
training they have been on.

c
 

 
  
 

  
  

 

 
    
 

 


  
 

 

Returns:

       

Bob Marley Risk Management 05/02/07
John Lennon Sales Training 10/01/07
John Lennon First Aid 01/03/07
Ralph Kimball Sales Training 10/01/07

The above query demonstrates the  clause which specifies the two tables that
we are using and then uses the  keyword to define the relationship or 'joining points'
between the two tables.

We can see that columns are identified by using 2
 

   syntax so that the
query knows which table to find the column we are referencing. This is because the same
column name may be present in more than one table (e.g. the column name r

appears in both tables in our example).

The INNER JOIN clause in the example above can be rewritten in an alternative format (a
JOIN condition) by defining the relationship between the two tables in the WHERE clause.

c
 

 
   
 

  
  

 

  

 
   

 

  
 

 

From the above example we can see that the JOIN keyword can be used as shorthand for
INNER JOIN and the ON section of the INNER JOIN clause can be replaced by specifying
the relationship between the two tables in the WHERE clause.

c

It is important to note that the INNER JOIN Only returns data where a match is found.
Therefore no information for Bill Gates was returned from this query as there is no record in
the TrainingTaken table that contains Bill Gates EmployeeID.

If we wanted to return all rows from one table regardless of whether they had any related data
in the table they were joining to we would need to use an OUTER JOIN.

    !" 

In order to make our queries more readable, quicker to write and debug we can use table
aliases. This is where we give the table a shorthand name (after its definition in the FROM
clause). We can then use this shorthand name in the rest of the query when referencing the
table. Using table aliases we can rewrite our initial INNER JOIN example as follows:

c
  
 
  

 

 
    
   

 

 

In the above example we have given the Employee table an alias of r and the TrainingTaken
table an alias of 22. You can see how this helps the readability of queries, especially if you
have a query which had 3 or more tables joined together.

   #

We can create more than one join in a query as follows:

c
  ! " 

 !
#   
  !
$! #!#  
  !
"!$"$ 
 $%   ! 

The same query could be rewritten as:

c
  ! " 

 !
#  !
$! !
" 
   #!# 
 #!$"$ 
 $%  ! 

‰$" 

In this article we have demonstrated the concept of the INNER JOIN and how it is used to
match data from related tables. We can see that the JOIN clause is an essential tool when
querying well designed, relational databases. Without the JOIN clause we would not be able
to define meaningful relationships between tables in our database.

We have seen how the join relationship can be specified either after the ON keyword in the
JOIN clause or in the WHERE clause. We have also seen how table aliases can help make
writing, reading and maintaining our SQL statements easier.

We have seen how the INNER JOIN only returns data from a table when a match is found in
the other table specified in the JOIN. When data is required from one of the tables in the
JOIN regardless of whether a match is found in the other table then an OUTER JOIN must be
used.