This is an in-memory backend implementation for the django models. With dmqs it's possible to load Models, patch them, and a call to model_instance.save() will never hit the database. You can still perform Model.objects.all() and the result will work like if django was running with sqlite in memory, but even faster.

The initial motivation is to speed up fixture loading in Django.

dmqs could be used for many testing and learning purposes, focusing on Testing.

Instead of TestCase use dmqs.integration.MemoryTestCase

from dmqs.integration.testcase import MemoryTestCase

class FastTest(MemoryTestCase):
    fixtures = ['dmqs_app/tests/fixtures/fixture.json']

    def test_friend(self):
        friend = Friend.objects.get(id=1)
        self.assertTrue(friend != None)

First time:

pip install -r requirements.txt

To actually run the tests:

make test

First time:

pip install -r coverage_requirements.txt

And then:

make coverage

In few weeks I expect to release a beta version. The integration with django test mechanism isn't ready yet.

Felipe Cruz felipecruz@loogica.net

Thiago Garcia thiagogds14@gmail.com