Overview
- SQLite has two databases, which are always loaded onto it. One is the “main” database and another one is the “temp” database.
- The “temp” database is for storing temporary tables.
- The “main” database and “temp” database can neither be attached nor be detached.
- A database file can be attached to an existing SQLite connection by using the ATTACH statement.

The ATTACH statement in SQLite
The ATTACH statement in SQLite is of the form,
ATTACH DATABASE <database file name> AS <database name>- If a database file with the specified file name does not exist a new file will be created.
Example 1- Attach Database
ATTACH DATABASE bill.db AS billingThe above example will create a database named billing inside the file bill.db and any tables created on the database billing will be saved under the file bill.db.
Example 2- Create a new table in the attached SQLite Database
Assuming the currency is stored in the lowest denomination such as cents and tax rate is stored as basis points the following statement will create a table with the name cashbill in the SQLite database billing.
CREATE TABLE billing.cashbill(id int, itemcode int, rate int, quantity int, price int, taxrate int, tax int, total int)The following INSRT statements will populate the newly attached SQLite database billing.cashbill with two records.
INSERT INTO billing.cashbill VALUES (1, 256,300,2,600,2,12,612)
INSERT INTO billing.cashbill VALUES (2, 272,100,3,300,2,6,306)Attaching a SQLite Database using Python
- The Sqlite3 module acts as the SQLite client using which a Python program can connect to the Sqlite3 database.
- Once Sqlite3 is imported into a python program a connection object can be created to the SQLite main database.
- Through the connection object a cursor object should be obtained using which any SQL statement including the ATTACH DATABASE statement.
- Once the ATTACH DATABASE is executed, the connection object obtained can now be used to access both the main database and the newly attached database.
- The example Python program below attaches a new SQLite database, creates a new table on the attached database and populates two records into the new table.
- The SQLite table is queried and the results are displayed before detaching the database and closing the connection.
Example3- Attaching a SQLite table using Python
# Example Python Program to attach a database file to an existing DB connection
# import the sqlite module
import sqlite3
# Create database connection to the sqlite main database
connectionObject = sqlite3.connect("primedb.db")
#Obtain a cursor object
cursorObject = connectionObject.cursor()
'''
Perform operations on the prime db
.
.
.
Done with prime db and want to work on billing database using the same database connection
'''
# Attach a database file
attachDatabaseSQL = "ATTACH DATABASE ? AS billing"
dbSpec = ("bill.db",)
cursorObject.execute(attachDatabaseSQL,dbSpec)
# Drop any existing tables with the same name
cursorObject.execute("drop table billing.cashbill")
# Create billing table in the newly attached database
createTableSQL = "CREATE TABLE billing.cashbill(billid int, itemcode int, rate int, quantity int, price int, taxrate int, tax int, total int)"
cursorObject.execute(createTableSQL)
# Insert a row of data into the billing.cashbill table
billid = 1
itemcode = 256
rate = 300
quantity = 2
price = 600
taxrate = 2
tax = 12
total = 621
insertDataSQL = "INSERT INTO billing.cashbill VALUES (?, ?,?,?,?,?,?,?)"
# Substitution parameters for the insert statement- ? will be replaced by members of this tuple
insertSpec = (billid,itemcode,rate,quantity,price,taxrate,tax,total)
#Execute SQL insert using parameter substitution
cursorObject.execute(insertDataSQL, insertSpec)
# Commit the changes
connectionObject.commit()
# Query the cashbill table
querySpec = (billid,)
cursorObject.execute("SELECT * FROM billing.cashbill WHERE billid=?", querySpec)
# print the record
print("Fetched record from the attached SQLite database table:");
print(cursorObject.fetchone())
# detach the database
detachDatabaseSQL = "DETACH DATABASE billing"
cursorObject.execute(detachDatabaseSQL)
# Close the database connection as the resource is no longer needed
connectionObject.close()Output
Fetched record from the attached SQLite database table:
(1, 256, 300, 2, 600, 2, 12, 621)