django-anonymous

Simple Django module to anonymize production data for safe usage on non-production environments.

Installation

pip install django-anonymous

Usage

In your app create a file anon.py:

from django_anonymous import Anonymizer, Faker, register
from .model import YourModel


@register(YourModel)
class YourModelAnonymizer(Anonymizer):
    
    # You can give any callable, Faker is a small wrapper around the `faker` library. 
    email = Faker("email", unique=True)
    
    # You can also use any static value
    first_name = "Anon"

Run the anonymizer

python manage.py anonymize

Custom QuerySet

You can set a custom QuerySet to filter out some objects

from django_anonymous import Anonymizer, Faker, register
from .model import YourModel


@register(YourModel)
class YourModelAnonymizer(Anonymizer):
    email = Faker("email", unique=True)

    def get_queryset(self):
        return super().get_queryset().filter(is_staff=True)

Faker seed

Default it will use the object id as seed, to generate the same data for every run. You can disable this by overriding the get_object_seed and return falsy value.

from django_anonymous import Anonymizer, Faker, register
from .model import YourModel


@register(YourModel)
class YourModelAnonymizer(Anonymizer):
    email = Faker("email", unique=True)

    def get_object_seed(self, obj):
        return None

Settings for Anonymizer

Per Anonymizer you can set the select chunk size and update batch size.

from django_anonymous import Anonymizer, Faker, register
from .model import YourModel


@register(YourModel)
class YourModelAnonymizer(Anonymizer):
    SELECT_CHUNK_SIZE = 100
    UPDATE_BATCH_SIZE = 25
    
    email = Faker("email", unique=True)

Inspired by

• https://github.com/Tesorio/django-anon
• https://github.com/FactoryBoy/factory_boy