Python SQLite – JOIN Clause

In this article, we discuss the JOIN clause in SQLite using the sqlite3 module in Python. But at first let’s see a brief about join in SQLite.

Join Clause

 A JOIN clause combines the records from two tables on the basis of common attributes. The different types of joins are as follows:

• INNER JOIN (OR JOIN) – Gives the records that have common attributes in both tables.
• LEFT JOIN – Gives all records from the left table and only the common records from the right table.
• RIGHT JOIN – Gives all records from the right table and only the common records from the left table.
• FULL OUTER JOIN – Gives all records when there is a common attribute in either the left or the right table.
• CROSS JOIN – Gives records of one table with all other records of another table.

Note: 

• Unlike other types of joins, it does not include a join condition.
• SQLite does not directly support the RIGHT JOIN and FULL OUTER JOIN. 

Creating a Database

Here, we will create a simple database having two tables Advisor(AdvisorID, AdvisorName) and Student(StudentID, StudentName, AdvisorID) where AdvisorID of the Student table is the foreign key referencing AdvisorID of the Advisor table.

Python3

# Import required libraries
import sqlite3

# Connect to SQLite database
# New file created if it doesn't already exist
conn = sqlite3.connect(r'C:\Users\SQLite\Geeks.db')

# Create cursor object
cursor = conn.cursor()

# Create and populate tables
cursor.executescript('''
CREATE TABLE Advisor(
AdvisorID INTEGER NOT NULL,
AdvisorName TEXT NOT NULL,
PRIMARY KEY(AdvisorID)
);

CREATE TABLE Student(
StudentID NUMERIC NOT NULL,
StudentName NUMERIC NOT NULL,
AdvisorID INTEGER,
FOREIGN KEY(AdvisorID) REFERENCES Advisor(AdvisorID),
PRIMARY KEY(StudentID)
);

INSERT INTO Advisor(AdvisorID, AdvisorName) VALUES
(1,"John Paul"),
(2,"Anthony Roy"),
(3,"Raj Shetty"),
(4,"Sam Reeds"),
(5,"Arthur Clintwood");

INSERT INTO Student(StudentID, StudentName, AdvisorID) VALUES
(501,"Geek1",1),
(502,"Geek2",1),
(503,"Geek3",3),
(504,"Geek4",2),
(505,"Geek5",4),
(506,"Geek6",2),
(507,"Geek7",2),
(508,"Geek8",3),
(509,"Geek9",NULL),
(510,"Geek10",1);

''')

#Commit changes to database
conn.commit()

# Closing the connection
conn.close()

Tables Created:

Advisor Table

Student Table

Now, let’s perform different types of join on the above-created database.

INNER JOIN 

Inner join also represented as join which gives the records that have common attributes in both tables.

Syntax:

   SELECT columns

   FROM table1

   [INNER] JOIN table2

   ON table1.column = table2.column;

INNER keyword is optional

Python3

# Import required libraries
import sqlite3

# Connect to SQLite database
conn = sqlite3.connect(r'C:\Users\SQLite\Geeks.db')

# Create cursor object
cursor = conn.cursor()

# Query for INNER JOIN
sql = '''SELECT StudentID, StudentName, AdvisorName
FROM Student
INNER JOIN Advisor
ON Student.AdvisorID = Advisor.AdvisorID;'''

# Executing the query
cursor.execute(sql)

# Fetching rows from the result table
result = cursor.fetchall()
for row in result:
    print(row)

# Closing the connection
conn.close()

Output:

LEFT JOIN 

Gives all records from the left table, and only the common records from the right table.

Syntax:

  SELECT columns

  FROM table1

  LEFT [OUTER] JOIN table2 

  ON table1.column = table2.column;

OUTER keyword is optional

Python3

# Import required libraries
import sqlite3

# Connect to SQLite database
conn = sqlite3.connect(r'C:\Users\SQLite\Geeks.db')

# Create cursor object
cursor = conn.cursor()

# Query for LEFT JOIN
sql = '''SELECT StudentID, StudentName, AdvisorName
FROM Student
LEFT JOIN Advisor
USING(AdvisorID) ;'''

# Executing the query
cursor.execute(sql)

# Fetching rows from the result table
result = cursor.fetchall()
for row in result:
    print(row)

# Closing the connection
conn.close()

Since the column name (AdvisorID) of joined tables is same, the clause USING(AdvisorID) can be used instead of ON Student.AdvisorID = Advisor.AdvisorID.

Output:

RIGHT JOIN

Gives all records from the right table, and only the common records from the left table. As mentioned before, SQLite does not directly support RIGHT JOIN. However, it can be emulated using LEFT JOIN by switching the positions of the student and advisor table.

Syntax:

  SELECT columns

  FROM table1

  RIGHT [OUTER] JOIN table2

  ON table1.column = table2.column;

OUTER keyword is optional

Python3

# Import required libraries
import sqlite3

# Connect to SQLite database
conn = sqlite3.connect(r'C:\Users\SQLite\Geeks.db')

# Create cursor object
cursor = conn.cursor()

# Query for RIGHT JOIN
sql = '''SELECT StudentID, StudentName, AdvisorName
FROM Advisor
LEFT JOIN Student
USING(AdvisorID);'''

# Executing the query
cursor.execute(sql)

# Fetching rows from the result table
result = cursor.fetchall()
for row in result:
    print(row)

# Closing the connection
conn.close()

Output:

FULL OUTER JOIN

Gives all records when there is a common attribute in either left or the right table. As mentioned before, SQLite does not directly support FULL OUTER JOIN. However, it can be emulated using LEFT JOIN. In this query, the second SELECT statement has the positions of the student and advisor table switched. The UNION ALL clause retains the duplicate rows from the result of both SELECT queries. And the WHERE clause in the second SELECT statement removes rows that already included in the result set of the first SELECT statement.

Syntax:

  SELECT columns

  FROM table1

  FULL [OUTER] JOIN table2

  ON table1.column = table2.column;

OUTER keyword is optional

Python3

# Import required libraries
import sqlite3

# Connect to SQLite database
conn = sqlite3.connect(r'C:\Users\SQLite\Geeks.db')

# Create cursor object
cursor = conn.cursor()

# Query for FULL OUTER JOIN
sql = '''SELECT StudentID, StudentName, AdvisorName
FROM Student
LEFT JOIN Advisor
USING(AdvisorID)
UNION ALL
SELECT StudentID, StudentName, AdvisorName
FROM Advisor
LEFT JOIN Student
USING(AdvisorID)
WHERE Student.AdvisorID IS NULL;'''

# Executing the query
cursor.execute(sql)

# Fetching rows from the result table
result = cursor.fetchall()
for row in result:
    print(row)

# Closing the connection
conn.close()

Output:

CROSS JOIN 

It combines all records of one table with all other records of another table, that is, it creates a Cartesian product of records from the join tables.

Syntax:

  SELECT columns 

  FROM table1

  CROSS JOIN table2;

Python3

# Import required libraries
import sqlite3

# Connect to SQLite database
conn = sqlite3.connect(r'C:\Users\SQLite\Geeks.db')

# Create cursor object
cursor = conn.cursor()

# Query for CROSS JOIN
sql = '''SELECT StudentID, StudentName, AdvisorName
FROM Student
CROSS JOIN Advisor;'''

# Executing the query
cursor.execute(sql)

# Fetching rows from the result table
result = cursor.fetchall()
for row in result:
    print(row)

# Closing the connection
conn.close()

Output: