JOINS

SQL

● Welcome to this section on JOINS.

● JOINS will allow us to combine information from
multiple tables!
● Let’s see what we will learn in this section
 SQL

● Section Overview
○ Creating an alias with the AS clause
○ Understanding different kinds of JOINs
■ INNER JOINS
■ OUTER JOINS
■ FULL JOINS
■ UNIONS
○ Challenge Tasks
AS
 SQL

● Before we learn about JOINs , let’s quickly cover the

AS clause which allows us to create an “alias” for a
column or result.
● Let’s see the example syntax
 SQL

● SELECT column AS new_name

FROM table
 SQL

● SELECT SUM(column) AS new_name

FROM table
 SQL

● Example
 SQL

● Example
 SQL

● Example
 SQL

● The AS operator gets executed at the very end of a

query, meaning that we can not use the ALIAS inside
a WHERE operator.
● Let’s walk through a few examples in pgAdmin!
INNER JOIN
 SQL

● There are several types of JOINs, in this lecture we

will go through the simplest JOIN type, which is an
INNER JOIN
 SQL

● What is a JOIN operation?

● JOINs allow us to combine multiple tables together.
● The main reason for the different JOIN types is to
decide how to deal with information only present in
one of the joined tables.
 SQL

● Let’s imagine a simple example.

● Our company is holding a conference for people in
the movie rental industry.
● We’ll have people register online beforehand and then
login the day of the conference.
 SQL

● After the conference we have these tables

REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● The respective id columns indicate what order they

registered or logged in on site.
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● For the sake of simplicity, we will assume the names

are unique.
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● To help you keep track, Registrations names’ first

letters go A,B,C,D
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● An INNER JOIN will result with the set of records

that match in both tables.
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● An INNER JOIN will result with the set of records

that match in both tables.
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● SELECT * FROM TableA

INNER JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL

● SELECT * FROM TableA

INNER JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL

● SELECT * FROM TableB

INNER JOIN TableA
ON TableA.col_match = TableB.col_match
 SQL

● SELECT * FROM Registrations

INNER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● SELECT * FROM Registrations

INNER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● SELECT * FROM Registrations

INNER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● SELECT * FROM Registrations

INNER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS

reg_id name RESULTS log_id name

1 Andrew reg_id name log_id name 1 Xavier

2 Bob 1 Andrew 2 Andrew 2 Andrew

3 Charlie 2 Bob 4 Bob 3 Yolanda

4 David 4 Bob
 SQL

● SELECT reg_id,[Link],log_id
FROM Registrations
INNER JOIN Logins
ON [Link] = [Link]
RESULTS

reg_id name log_id

1 Andrew 2

2 Bob 4
 SQL

● SELECT reg_id,[Link],log_id
FROM Registrations
INNER JOIN Logins
ON [Link] = [Link]
RESULTS

reg_id name log_id

1 Andrew 2

2 Bob 4
 SQL

● Remember that table order won’t matter in an INNER

JOIN
● Also if you see just JOIN without the INNER,
PostgreSQL will treat it as an INNER JOIN.
● Let’s see some examples in pgAdmin!
OUTER JOINS
 SQL

● There are few different types of OUTER JOINs

● They will allow us to specify how to deal with values
only present in one of the tables being joined.
● These are the more complex JOINs, take your time
when trying to understand them!
 SQL

● In these lectures we will explain:

○ FULL OUTER JOIN
■ Clarifying WHERE null
○ LEFT OUTER JOIN
■ Clarifying WHERE null
○ RIGHT OUTER JOIN
■ Clarifying WHERE null
FULL OUTER JOIN
 SQL

● Let’s review our two example tables from the

previous lectures.
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● We know we would join these tables together on the

name column
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● Recall we match Andrew and Bob in both tables

REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● But we have names that only appear in one table!

REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● Let’s see how the different OUTER JOINs deal with

this discrepancy.
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

We will first take a look at the simplest, which is a

FULL OUTER JOIN
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● SELECT * FROM TableA

FULL OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL

● SELECT * FROM TableB

FULL OUTER JOIN TableA
ON TableA.col_match = TableB.col_match
 SQL

● SELECT * FROM Registrations

FULL OUTER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL
● SELECT * FROM Registrations FULL OUTER JOIN Logins
ON [Link] = [Link]

REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL
● SELECT * FROM Registrations FULL OUTER JOIN Logins
ON [Link] = [Link]
RESULTS

reg_id name log_id name

REGISTRATIONS LOGINS
1 Andrew
reg_id name log_id name
2 Bob
1 Andrew 1 Xavier
3 Charlie
2 Bob 2 Andrew

3 Charlie 4 David
3 Yolanda
4 David 4 Bob
 SQL
● SELECT * FROM Registrations FULL OUTER JOIN Logins
ON [Link] = [Link]
RESULTS

reg_id name log_id name

REGISTRATIONS LOGINS
1 Andrew 2 Andrew
reg_id name log_id name
2 Bob 4 Bob
1 Andrew 1 Xavier
3 Charlie
2 Bob 2 Andrew
4 David
3 Charlie 3 Yolanda
4 David 4 Bob
 SQL
● SELECT * FROM Registrations FULL OUTER JOIN Logins
ON [Link] = [Link]
RESULTS

reg_id name log_id name

REGISTRATIONS LOGINS
1 Andrew 2 Andrew
reg_id name log_id name
2 Bob 4 Bob
1 Andrew 1 Xavier
3 Charlie null null
2 Bob 2 Andrew
4 David null null
3 Charlie 3 Yolanda
4 David 4 Bob
 SQL
● SELECT * FROM Registrations FULL OUTER JOIN Logins
ON [Link] = [Link]
RESULTS

reg_id name log_id name

REGISTRATIONS LOGINS
1 Andrew 2 Andrew
reg_id name log_id name
2 Bob 4 Bob
1 Andrew 1 Xavier
3 Charlie null null
2 Bob 2 Andrew
4 David null null
3 Charlie 3 Yolanda
null null 1 Xavier
4 David 4 Bob
null null 3 Yolanda
FULL OUTER JOIN
with WHERE
Get rows unique to either table
(rows not found in both tables)
 SQL
● SELECT * FROM TableA
FULL OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
WHERE [Link] IS null OR
[Link] IS null
 SQL
● SELECT * FROM TableB
FULL OUTER JOIN TableA
ON TableA.col_match = TableB.col_match
WHERE [Link] IS null OR
[Link] IS null
 SQL
SELECT * FROM Registrations FULL OUTER JOIN
Logins
ON [Link] = [Link]
WHERE Registrations.reg_id IS null OR
Logins.log_id IS null
 SQL
● SELECT * FROM Registrations FULL OUTER JOIN Logins
ON [Link] = [Link]
RESULTS

reg_id name log_id name

REGISTRATIONS LOGINS
1 Andrew 2 Andrew
reg_id name log_id name
2 Bob 4 Bob
1 Andrew 1 Xavier
3 Charlie null null
2 Bob 2 Andrew
4 David null null
3 Charlie 3 Yolanda
null null 1 Xavier
4 David 4 Bob
null null 3 Yolanda
 SQL
● SELECT * FROM Registrations FULL OUTER JOIN Logins
ON [Link] = [Link]
WHERE Registrations.reg_id IS null OR
Logins.log_id IS null
REGISTRATIONS RESULTS LOGINS

reg_id name reg_id name log_id name log_id name

1 Andrew 3 Charlie null null 1 Xavier

2 Bob 4 David null null 2 Andrew

3 Charlie 3 Yolanda
null null 1 Xavier
4 David 4 Bob
null null 3 Yolanda
 SQL

● Let’s see a quick example in pgAdmin of FULL

OUTER JOINs!
LEFT OUTER JOIN
 SQL

● A LEFT OUTER JOIN results in the set of records

that are in the left table, if there is no match with the
right table, the results are null.
● Later on we will learn how to add WHERE statements
to further modify a LEFT OUTER JOIN
 SQL

● SELECT * FROM TableA

LEFT OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL

● SELECT * FROM TableA

LEFT JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL

● SELECT * FROM TableA

LEFT OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL ORDER
MATTERS FOR
● SELECT * FROM TableA LEFT OUTER JOIN!
LEFT OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL ORDER
MATTERS FOR
● SELECT * FROM TableA LEFT OUTER JOIN!
LEFT OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
 SQL

Let’s explore a LEFT OUTER JOIN with our two

example tables.
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● SELECT * FROM Registrations

LEFT OUTER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS

reg_id name log_id name

1 Andrew 1 Xavier

2 Bob 2 Andrew

3 Charlie 3 Yolanda
4 David 4 Bob
 SQL

● SELECT * FROM Registrations

LEFT OUTER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS
RESULTS
reg_id name log_id name
reg_id name log_id name
1 Andrew 1 Xavier
1 Andrew 2 Andrew
2 Bob 2 Andrew
2 Bob 4 Bob
3 Charlie 3 Yolanda
3 Charlie null null
4 David 4 Bob
4 David null null
 SQL

● SELECT * FROM Registrations

LEFT OUTER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS
RESULTS
reg_id name log_id name
reg_id name log_id name
1 Andrew 1 Xavier
1 Andrew 2 Andrew
2 Bob 2 Andrew
2 Bob 4 Bob
3 Charlie 3 Yolanda
3 Charlie null null
4 David 4 Bob
4 David null null
LEFT OUTER JOIN
With WHERE
Get rows unique to left table
 SQL
● What if we only wanted entries unique to Table A?
Those rows found in Table A and not found in Table
B.
 SQL
● SELECT * FROM TableA
LEFT OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
WHERE [Link] IS null
 SQL

● SELECT * FROM Registrations

LEFT OUTER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS LOGINS
RESULTS
reg_id name log_id name
reg_id name log_id name
1 Andrew 1 Xavier
1 Andrew 2 Andrew
2 Bob 2 Andrew
2 Bob 4 Bob
3 Charlie 3 Yolanda
3 Charlie null null
4 David 4 Bob
4 David null null
 SQL
SELECT * FROM Registrations
LEFT OUTER JOIN Logins
ON [Link] = [Link]
REGISTRATIONS
WHERE Logins.log_id IS null LOGINS

reg_id name log_id name

1 Andrew 1 Xavier
RESULTS
2 Bob 2 Andrew
reg_id name log_id name
3 Charlie 3 Yolanda
3 Charlie null null
4 David 4 Bob
4 David null null
 SQL

● Let’s see some examples of LEFT OUTER JOINs in

pgAdmin!
RIGHT JOINS
 SQL

● A RIGHT JOIN is essentially the same as a LEFT

JOIN, except the tables are switched.
● This would be the same as switching the table order in
a LEFT OUTER JOIN.
● Let’s quickly see some examples of a RIGHT JOIN.
 SQL

● SELECT * FROM TableA

RIGHT OUTER JOIN TableB
ON TableA.col_match = TableB.col_match

Table A Table B
 SQL

● SELECT * FROM TableA

RIGHT JOIN TableB
ON TableA.col_match = TableB.col_match

Table A Table B
 SQL
● SELECT * FROM TableA
RIGHT OUTER JOIN TableB
ON TableA.col_match = TableB.col_match
WHERE [Link] IS null

Table A Table B
 SQL

● It is up to you and how you have the tables organized

“in your mind” when it comes to choosing a LEFT vs
RIGHT join, since depending on the table order you
specify in the JOIN, you can perform duplicate JOINs
with either method.
 JOIN
CHALLENGE TASKS
UNIONS
 SQL

● The UNION operator is used to combine the result-set

of two or more SELECT statements.
● It basically serves to directly concatenate two results
together, essentially “pasting” them together.
 SQL

● SELECT column_name(s) FROM table1

UNION
SELECT column_name(s) FROM table2;
 SQL

Let’s explore a UNION with two example tables.

Sales2021_Q1 Sales2021_Q2

name amount name amount

David 100 David 200

Claire 50 Claire 100

 SQL

● SELECT * FROM Sales2021_Q1

UNION
SELECT * FROM Sales2021_Q2;
name amount

David 100

Claire 50

David 200

Claire 100
 SQL

● SELECT * FROM Sales2021_Q1

UNION
SELECT * FROM Sales2021_Q2
ORDER BY name;
name amount

David 100

David 200

Claire 50

Claire 100
 JOIN
Challenges
 SQL

● California sales tax laws have changed and we need to

alert our customers to this through email.
● What are the emails of the customers who live in
California?
 SQL

● Expected Results
 SQL

● Hints
○ You will need to use the address and customer
tables.
○ Look at the district column
 SQL

SELECT district,email FROM address

INNER JOIN customer ON
address.address_id = customer.address_id
WHERE district = 'California'
 SQL

● A customer walks in and is a huge fan of the actor

“Nick Wahlberg” and wants to know which movies he
is in.
● Get a list of all the movies “Nick Wahlberg” has been
in.
 SQL

● Expected Results

……………...
 SQL

● HINTS
○ You will need to do 2 JOINs in a row to do this in
a single query.
○ Try using the online documentation or a simple
google search to see how this is done.
○ Tables to use: actor, film, film_actor
 SQL

SELECT title,first_name,last_name
FROM film_actor INNER JOIN actor
ON film_actor.actor_id = actor.actor_id
INNER JOIN film
ON film_actor.film_id = film.film_id
WHERE first_name = 'Nick'
AND last_name = 'Wahlberg'