Awesome Open Source
Awesome Open Source

Event Sourcing in Python

Latest Release Build Status Coverage Status

A library for event sourcing in Python.


Use pip to install the stable distribution from the Python Package Index.

$ pip install eventsourcing

Please refer to the documentation for installation and usage guides.

Register questions, requests and issues on GitHub.

There is a Slack channel for this project, which you are welcome to join.


Event store — appends and retrieves domain events. Uses a sequenced item mapper with a record manager to map domain events to database records in ways that can be easily extended and replaced.

Layer base classes — suggest how to structure an event sourced application. The library has base classes for application objects, domain entities, entity repositories, domain events of various types, mapping strategies, snapshotting strategies, cipher strategies, etc. They are well factored, relatively simple, and can be easily extended for your own purposes. If you wanted to create a domain model that is entirely stand-alone (recommended by purists for maximum longevity), you might start by replicating the library classes.

Notifications and projections — reliable propagation of application events with pull-based notifications allows the application state to be projected accurately into replicas, indexes, view models, and other applications.

Process and systems — scalable event processing with application pipelines. Runnable with single thread, multiprocessing on a single machine, and in a cluster of machines using the actor model. Parallel pipelines are synchronised with causal dependencies.

Additional features

Versioning - allows model changes to be introduced after an application has been deployed. Both domain events and domain entity classes can be versioned. The recorded state of an older version can be upcast to be compatible with a new version. Stored events and snapshots are upcast from older versions to new versions before the event or entity object is reconstructed.

Snapshotting — avoids replaying an entire event stream to obtain the state of an entity. A snapshot strategy is included which reuses the capabilities of this library by implementing snapshots as events.

Hash chaining — Sequences of events can be hash-chained, and the entire sequence of events checked for data integrity. Information lost in transit or on the disk from database corruption can be detected. If the last hash can be independently validated, then so can the entire sequence.

Correlation and causation IDs - Domain events can easily be given correlation and causation IDs, which allows a story to be traced through a system of applications.

Compression - reduces the size of stored domain events and snapshots, usually by around 25% to 50% of the original size. Compression reduces the size of data in the database and decreases transit time across a network.

Application-level encryption — encrypts and decrypts stored events and snapshots, using a cipher strategy passed as an option to the sequenced item mapper. Can be used to encrypt some events, or all events, or not applied at all (the default). This means data will be encrypted in transit across a network ("on the wire") and at disk level including backups ("at rest"), which is a legal requirement in some jurisdictions when dealing with personally identifiable information (PII) for example the EU's GDPR.

Optimistic concurrency control — ensures a distributed or horizontally scaled application doesn't become inconsistent due to concurrent method execution. Leverages optimistic concurrency controls in adapted database management systems.

Worked examples — simple example application and systems, with an example entity class, example domain events, and an example database table. Plus lots of examples in the documentation.


Define a domain model. Here we define a model with one aggregate root class called "World", which has a command method called "make_it_so" that triggers a domain event called "SomethingHappened", which mutates the aggregate by appending the "what" of the domain event to the "history" of the world.

from eventsourcing.domain.model.aggregate import AggregateRoot

class World(AggregateRoot):

    def __init__(self, **kwargs):
        super(World, self).__init__(**kwargs)
        self.history = []

    def make_it_so(self, something):
        self.__trigger_event__(World.SomethingHappened, what=something)

    class SomethingHappened(AggregateRoot.Event):
        def mutate(self, aggregate):

Generate a cipher key (optional).

from eventsourcing.utils.random import encoded_random_bytes

# Keep this safe.
cipher_key = encoded_random_bytes(num_bytes=32)

Configure environment variables (optional).

import os

# Cipher key (random bytes encoded with Base64).
os.environ['CIPHER_KEY'] = cipher_key

# SQLAlchemy-style database connection string.
os.environ['DB_URI'] = 'sqlite:///:memory:'

Run the code. Construct application and use as context manager.

from eventsourcing.application.notificationlog import NotificationLogReader
from eventsourcing.application.sqlalchemy import SQLAlchemyApplication
from eventsourcing.exceptions import ConcurrencyError

with SQLAlchemyApplication(persist_event_type=World.Event) as app:

    # Create new aggregate.
    world = World.__create__()

    # Aggregate not yet in repository.
    assert not in app.repository

    # Execute commands.

    # View current state of aggregate object.
    assert world.history[0] == 'dinosaurs'
    assert world.history[1] == 'trucks'

    # Note version of object at this stage.
    version = world.__version__

    # Execute another command.

    # Store pending domain events.

    # Aggregate now exists in repository.
    assert in app.repository

    # Show the notification log has four items.
    assert len(app.notification_log['current'].items) == 4

    # Replay stored events for aggregate.
    copy = app.repository[]

    # View retrieved aggregate.
    assert isinstance(copy, World)
    assert copy.history[0] == 'dinosaurs'
    assert copy.history[1] == 'trucks'
    assert copy.history[2] == 'internet'

    # Verify retrieved state (cryptographically).
    assert copy.__head__ == world.__head__

    # Delete aggregate.

    # Discarded aggregate not found.
    assert not in app.repository
        # Repository raises key error.
    except KeyError:
        raise Exception("Shouldn't get here")

    # Get historical state (at version from above).
    old = app.repository.get_entity(, at=version)
    assert old.history[-1] == 'trucks' # internet not happened
    assert len(old.history) == 2

    # Optimistic concurrency control (no branches).
    except ConcurrencyError:
        raise Exception("Shouldn't get here")

    # Check domain event data integrity (happens also during replay).
    events = app.event_store.list_events(
    last_hash = ''
    for event in events:
        assert event.__previous_hash__ == last_hash
        last_hash = event.__event_hash__

    # Verify stored sequence of events against known value.
    assert last_hash == world.__head__

    # Check records are encrypted (values not visible in database).
    record_manager = app.event_store.record_manager
    items = record_manager.get_items(
    for item in items:
        for what in [b'dinosaurs', b'trucks', b'internet']:
            assert what not in item.state
        assert == item.originator_id

    # Project application event notifications.
    reader = NotificationLogReader(app.notification_log)
    notification_ids = [n['id'] for n in]
    assert notification_ids == [1, 2, 3, 4, 5]

    # - create two more aggregates
    world = World.__create__()

    world = World.__create__()

    # - get the new event notifications from the reader
    notification_ids = [n['id'] for n in]
    assert notification_ids == [6, 7]

The double leading and trailing underscore naming style, seen above, is used consistently in the library's domain entity and event base classes for attribute and method names, so that developers can begin with a clean namespace. The intention is that the library functionality is included in the application by aliasing these library names with names that work within the project's ubiquitous language.

This style breaks PEP8, but it seems worthwhile in order to keep the "normal" Python object namespace free for domain modelling. It is a style used by other libraries (such as SQLAlchemy and Django) for similar reasons.

The exception is the id attribute of the domain entity base class, which is assumed to be required by all domain entities (or aggregates) in all domains.


This project is hosted on GitHub.

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
python (44,740
python3 (1,212
django (859
distributed-systems (251
cqrs (136
ddd (135
event-sourcing (120
sqlalchemy (87
domain-driven-design (72
eventsourcing (32

Find Open Source By Browsing 7,000 Topics Across 59 Categories