Interface with SQLite database in python

on
sqlite_python

Interface with SQLite database in python

What is SQLite database

  • SQLite database is a file based database
  • Just like a CSV, an SQLite database will be a single file (like app.db)
  • It does not need a separate database server. Applications can interface with the database file directly. Hence it is easily portable
  • SQLite databases are generally used in use cases where concurrency, high data exchange speed is not required.

SQL to create a database table

CREATE TABLE persons (
	id INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT,
	name TEXT(300) NOT NULL,
	dob TIMESTAMP NOT NULL,
	phone TEXT,
    CONSTRAINT persons_name_IDX UNIQUE (name),
    CONSTRAINT persons_phone_IDX UNIQUE (phone)
);

Create Table example

  • The SQL command to create the table is run using the execute method on the connection object.
import sqlite3
conn = sqlite3.connect("app.db")
conn.execute("""
CREATE TABLE if not exists persons (
	id INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT,
	name TEXT(300) NOT NULL,
	dob TIMESTAMP NOT NULL,
	phone TEXT,
    CONSTRAINT persons_name_IDX UNIQUE (name),
    CONSTRAINT persons_phone_IDX UNIQUE (phone)
);
""")
conn.close()

VS Code extension to view SQLite database

image.png

  • Use CTRL+SHIFT+P in VS code to run the command SQLite: Open Database and open an SQLite database file as shown below. Also SQL commands can be executed easily with the commands SQLite: New Query and SQLite: Run Query

image.png

Insert single row example

  • An SQL INSERT command can executed as shown below to insert rows into a table
  • The INSERT command would be like
INSERT INTO table_name (column1, column2, column3) VALUES (12, "TEST", 36);
  • The data for named parameters :name, :dob, :phone in the SQL is provided as a python dictionary {"name":"John","dob": dt.datetime(2000,1,24),"phone":"9876543210"}
import sqlite3
import datetime as dt

conn = sqlite3.connect("app.db")

employee = {
    "name": "John",
    "dob": dt.datetime(2000,1,24),
    "phone": "9876543210"
}

res = conn.execute("""
INSERT INTO persons (name,dob,phone) VALUES (:name,:dob,:phone);
""",employee)

print(f"number of inserted rows = {res.rowcount}")

conn.commit()
conn.close()

Insert multiple rows example

  • executemany can be used to run a single SQL INSERT command multiple times for each parameters dictionary as shown below. Instead of using for loop to insert each row, this approach is more efficient and quick
import sqlite3
import datetime as dt

conn = sqlite3.connect("app.db")

employees = [
    {"name": "Bob", "dob": dt.datetime(1992, 1, 24), "phone": "9765432942"},
    {"name": "Peter", "dob": dt.datetime(1994, 2, 10), "phone": "9876543087"},
    {"name": "Mark", "dob": dt.datetime(1998, 6, 15), "phone": "9876321049"}
]

res = conn.executemany("""
INSERT INTO persons (name,dob,phone) VALUES (:name,:dob,:phone);
""", employees)

print(f"number of inserted rows = {res.rowcount}")

conn.commit()
conn.close()
  • This approach can be used for update and delete database operations also

Read table rows example

  • The following example gets the columns name, dob and phone from the table “persons” with a condition that dob is less than 01-Jan-2000
  • The SELECT command would be like
SELECT column1, column2, column3 from table_name WHERE column1>10 AND column2="TEST";


import sqlite3
import datetime as dt

conn = sqlite3.connect("app.db")
cur = conn.cursor()

sqlParams = {"dob": dt.datetime(2000, 1, 1)}
rowsCursor = cur.execute("""
SELECT id, name, dob, phone from persons where dob<:dob;
""", sqlParams)

for r in rowsCursor:
    print(f"ID = {r[0]}, Name = {r[1]}, DoB = {r[2]}, Phone = {r[3]}")

# get all rows into a list
# print(rowsCursor.fetchall())

# get the column names of the returned rows
# fetchedColumns = [x[0] for x in rowsCursor.description]
# print(fetchedColumns)

cur.close()
conn.close()
  • The cursor object contains the rows returned by the SQL SELECT statement
  • Each row is a python tuple like (6, "Bob", dt.datetime(1992,1,24), "9765432942”)
  • The cursor rows can be iterated with a for…in loop as shown below
  • All rows can be retrieved at once as a list of tuples by calling the fetchall() method on the cursor

Update table rows example

  • The following example updates the name column of the rows with id = 1
  • The UPDATE command would be like
UPDATE table_name SET column1=value1, column2=value2 where column3="TEST";


import sqlite3
import datetime as dt

conn = sqlite3.connect("app.db")

sqlParams = {"name": "John Doe", "id": 1}
res = conn.execute("""
UPDATE persons SET name=:name where id=:id;
""", sqlParams)

print(f"number of updated rows = {res.rowcount}")

conn.commit()
conn.close()

Delete table rows example

  • The following example deletes the rows with dob greater than Jan 1st 2000
  • The DELETE command would be like
DELETE from table_name WHERE column1>10 AND column2<50;


import sqlite3
import datetime as dt

conn = sqlite3.connect("app.db")

sqlParams = {"dob": dt.datetime(2000, 1, 1)}
res = conn.execute("""
DELETE from persons where dob>:dob;
""", sqlParams)

print(f"number of deleted rows = {res.rowcount}")

conn.commit()
conn.close()

Transactions example

  • A transaction can be started by executing the “BEGIN” statement
  • A transaction can be rolled-back / aborted by executing the “ROLLBACK” statement
  • A transaction can be committed to the database by executing the “COMMIT” statement
  • The following example creates a transaction deletes a row but rolls back the transaction. Hence no rows are deleted in the table after the transaction
import sqlite3
import datetime as dt

conn = sqlite3.connect("app.db")

conn.execute("BEGIN")

sqlParams = {"name": "John"}
res = conn.execute("""
DELETE from persons where name=:name;
""", sqlParams)

print(f"number of deleted rows = {res.rowcount}")

rowsCursor = conn.execute("SELECT * from persons")
print(rowsCursor.fetchall())

conn.rollback()

rowsCursor = conn.execute("SELECT * from persons")
print(rowsCursor.fetchall())

conn.close()

References

Location: Mumbai, Maharashtra, India

Comments

Post a Comment