README.md

django-bulk-update

Simple bulk update over Django ORM or with helper function.

This project aims to bulk update given objects using one query over Django ORM.

Installation

pip install django-bulk-update

Usage

With manager:

import random
from bulk_update.manager import BulkUpdateManager
from tests.models import Person

class Person(models.Model):
    ...
    objects = BulkUpdateManager()

random_names = ['Walter', 'The Dude', 'Donny', 'Jesus']
people = Person.objects.all()
for person in people:
  person.name = random.choice(random_names)

Person.objects.bulk_update(people, update_fields=['name'])  # updates only name column
Person.objects.bulk_update(people, exclude_fields=['username'])  # updates all columns except username
Person.objects.bulk_update(people)  # updates all columns
Person.objects.bulk_update(people, batch_size=50000)  # updates all columns by 50000 sized chunks

With helper:

import random
from bulk_update.helper import bulk_update
from tests.models import Person

random_names = ['Walter', 'The Dude', 'Donny', 'Jesus']
people = Person.objects.all()
for person in people:
  person.name = random.choice(random_names)

bulk_update(people, update_fields=['name'])  # updates only name column
bulk_update(people, exclude_fields=['username'])  # updates all columns except username
bulk_update(people, using='someotherdb')  # updates all columns using the given db
bulk_update(people)  # updates all columns using the default db
bulk_update(people, batch_size=50000)  # updates all columns by 50000 sized chunks using the default db

Performance Tests:

Here we test the performance of the bulk_update function vs. simply calling .save() on every object update (dmmy_update). The interesting metric is the speedup using the bulk_update function more than the actual raw times.

# Note: SQlite is unable to run the `timeit` tests
# due to the max number of sql variables
In [1]: import os
In [2]: import timeit
In [3]: import django

In [4]: os.environ['DJANGO_SETTINGS_MODULE'] = 'tests.test_settings'
In [5]: django.setup()

In [6]: from tests.fixtures import create_fixtures

In [7]: django.db.connection.creation.create_test_db()
In [8]: create_fixtures(1000)

In [9]: setup='''
import random
from bulk_update import helper
from tests.models import Person
random_names = ['Walter', 'The Dude', 'Donny', 'Jesus']
ids = list(Person.objects.values_list('id', flat=True)[:1000])
people = Person.objects.filter(id__in=ids)
for p in people:
    name = random.choice(random_names)
    p.name = name
    p.email = '%[email protected]' % name
bu_update = lambda: helper.bulk_update(people, update_fields=['name', 'email'])
'''

In [10]: bu_perf = min(timeit.Timer('bu_update()', setup=setup).repeat(7, 100))

In [11]: setup='''
import random
from tests.models import Person
from django.db.models import F
random_names = ['Walter', 'The Dude', 'Donny', 'Jesus']
ids = list(Person.objects.values_list('id', flat=True)[:1000])
people = Person.objects.filter(id__in=ids)
def dmmy_update():
    for p in people:
        name = random.choice(random_names)
        p.name = name
        p.email = '%[email protected]' % name
        p.save(update_fields=['name', 'email'])
'''

In [12]: dmmy_perf = min(timeit.Timer('dmmy_update()', setup=setup).repeat(7, 100))
In [13]: print 'Bulk update performance: %.2f. Dummy update performance: %.2f. Speedup: %.2f.' % (bu_perf, dmmy_perf, dmmy_perf / bu_perf)
Bulk update performance: 7.05. Dummy update performance: 373.12. Speedup: 52.90.

Requirements

• Django 1.2+

Contributors

• aykut
• daleobrien
• sruon
• HowerHell
• c-nichols
• towr
• joshblum

TODO

• Geometry Fields support