Skip to main content

Python package for eventsourcing with Django.

Project description

Event Sourcing in Python with Django

This package is a Django app that supports using the Python eventsourcing library with the Django ORM.

Installation

Use pip to install the stable distribution from the Python Package Index. Please note, it is recommended to install Python packages into a Python virtual environment.

$ pip install eventsourcing_django

Synopsis

To use Django with your Python eventsourcing application, use the topic eventsourcing_django.factory:Factory as the INFRASTRUCTURE_FACTORY environment variable.

First define a domain model and application, in the usual way. You may set the INFRASTRUCTURE_FACTORY environment variable on the application class, so it can always use the Django ORM for storing events.

from eventsourcing.application import Application
from eventsourcing.domain import Aggregate, event


class World(Aggregate):
    def __init__(self):
        self.history = []

    @event("SomethingHappened")
    def make_it_so(self, what):
        self.history.append(what)


class Worlds(Application):
    env = {
        "INFRASTRUCTURE_FACTORY": "eventsourcing_django.factory:Factory",
        "IS_SNAPSHOTTING_ENABLED": "yes",
    }
    snapshotting_intervals = {
        World: 5,
    }

    def create_world(self):
        world = World()
        self.save(world)
        return world.id

    def make_it_so(self, world_id, what):
        world = self.repository.get(world_id)
        world.make_it_so(what)
        self.save(world)

    def get_world_history(self, world_id):
        world = self.repository.get(world_id)
        return world.history

Setup Django, in the usual way.

import os

import django
from django.core.management import call_command


# Set DJANGO_SETTINGS_MODULE.
os.environ.update({
    "DJANGO_SETTINGS_MODULE": "tests.djangoproject.settings",
})

# Setup Django.
django.setup()

# Setup the database.
call_command('migrate', 'eventsourcing_django')

The application's environment can use other environment variables supported by the library, for example to enable application-level compression and encryption of stored events, set COMPRESSOR_TOPIC and CIPHER_KEY.

from eventsourcing.cipher import AESCipher


# Generate a cipher key (keep this safe).
cipher_key = AESCipher.create_key(num_bytes=32)

# Set environment variables.
os.environ.update({
    "COMPRESSOR_TOPIC": "zlib",
    "CIPHER_KEY": cipher_key,
})

Construct and use the application. You may wish to do this within your Django project. The application can be created on a signal when the project is ready (use the ready() method of the AppConfig class in your Django app's apps.py module). The application command and query methods may be called from Django view and form classes.

# Construct the application.
app = Worlds()

# Call application command methods.
world_id = app.create_world()
app.make_it_so(world_id, "dinosaurs")
app.make_it_so(world_id, "trucks")
app.make_it_so(world_id, "internet")
app.make_it_so(world_id, "covid")

# Call application query methods.
history = app.get_world_history(world_id)
assert history == ["dinosaurs", "trucks", "internet", "covid"]

We can see the automatic snapshotting is working, by looking in the snapshots store.

snapshots = list(app.snapshots.get(world_id))
assert len(snapshots) == 1

We can see the application is using the Django infrastructure, and that compression and encryption are enabled, by checking the attributes of the application object.

from eventsourcing_django.factory import Factory
from eventsourcing_django.recorders import DjangoAggregateRecorder
from eventsourcing_django.recorders import DjangoApplicationRecorder
from eventsourcing_django.models import StoredEventRecord
from eventsourcing_django.models import SnapshotRecord
import zlib

assert isinstance(app.factory, Factory)
assert isinstance(app.events.recorder, DjangoApplicationRecorder)
assert isinstance(app.snapshots.recorder, DjangoAggregateRecorder)
assert issubclass(app.events.recorder.model, StoredEventRecord)
assert issubclass(app.snapshots.recorder.model, SnapshotRecord)
assert isinstance(app.mapper.cipher, AESCipher)
assert app.mapper.compressor == zlib

For more information, please refer to the Python eventsourcing library and the Django project.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

eventsourcing-django-0.1.0.tar.gz (9.5 kB view details)

Uploaded Source

Built Distribution

eventsourcing_django-0.1.0-py3-none-any.whl (8.9 kB view details)

Uploaded Python 3

File details

Details for the file eventsourcing-django-0.1.0.tar.gz.

File metadata

  • Download URL: eventsourcing-django-0.1.0.tar.gz
  • Upload date:
  • Size: 9.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.1.8 CPython/3.10.0 Darwin/19.6.0

File hashes

Hashes for eventsourcing-django-0.1.0.tar.gz
Algorithm Hash digest
SHA256 b549e93a61d4f6ed3c659aa80727316e71c56ef2ff282fa056c4ce43a8ec7923
MD5 8133cf467bd51b108183abc79368ad4d
BLAKE2b-256 6099ecdf4f433c1958cd2878f9425b5452aaa9b44d8ba45d34f138b169e02180

See more details on using hashes here.

File details

Details for the file eventsourcing_django-0.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for eventsourcing_django-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 276b818ec033f314bcfcf547b4cd38728e4b86e6271a8346c57f6e5a12a55511
MD5 2a2b39584cb93431bf222f9456ba7986
BLAKE2b-256 b7742302245a228933eaba239aeea228c323b080fc2b0a5a0d4675786908f3d2

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page