NORM

Not an ORM - a CRUD helper for python

NORM is a very lightweight database layer for python, built on top of DBAPI. It aims to make it extremely easy to do the simple stuff with an RDBMS, while still allowing you to drop down to SQL for the complex stuff.

Record objects behave as dictionaries, which can be persisted to the database by calling the store method.

On construction, the record object is given a standard python database connection object, which it uses for storing and retrieving records.

Class methods are provided for performing simple queries on tables.
These return iterators that yield instances of the required class.

Using NORM

Defining a table mapping:

class Person(NORM.DBObject):
	TABLE = 'people'
	FIELDS = ['firstname','surname','age']

Loading an object:

conn = psycopg2.connect('dbname=employees')
person = Person(conn, id = 1)
print "{o[firstname]} {o[surname]} is {o[age]} years old".format(
	o = person
)

Creating an object:

person = Person(conn)
person.update({
	'firstname': 'fred',
	'surname': 'bloggs',
	'age': 27
})
person.store()

Updating an object:

person = Person(conn, id = 7)
person.update({
	'age': 28
})
person.store()

or

person = Person(conn, id = 7)
person['age'] = 28
person.store()

Querying the table:

for person in People.select(conn, surname = 'bloggs'):
	print "{o[firstname]} {o[surname]} is {o[age]} years old".format(
		o = person
	)

To find people over 20:

for person in People.select(conn, age = ('>', 20)):
	print "{o[firstname]} {o[surname]} is {o[age]} years old".format(
		o = person
	)

Sorting (surname descending):

for person in People.select(conn, age = ('>', 20), _orderby = '-surname'):
	print "{o[firstname]} {o[surname]} is {o[age]} years old".format(
		o = person
	)

Supported databases

• PostgreSQL
• MySQL coming soon

License

NORM is licensed under a BSD style open source license.