Skip to main content
Help us improve PyPI by participating in user testing. All experience levels needed!

Esper is a lightweight Entity System for Python, with a focus on performance.

Project description


Esper is a lightweight Entity System for Python, with a focus on performance.

Esper is an MIT licensed Entity System, or, Entity Component System (ECS). The design is based on the Entity System concepts outlined by Adam Martin in his blog at, and others. Efforts were made to keep it as lightweight and performant as possible.

There is a fairly accurate writeup of what Entity Systems are in this Wikipedia article:

Inspired by Sean Fisk’s ecs, and Marcus von Appen’s ebs

What’s New

0.9.9 - The big change in this release is that esper has been condensed into a single
file: This will make it simple to just drop into your project folder, without cluttering your project with additional folders that didn’t really need to exist. You can still install it from PyPi via pip if you wish, but it’s easy enough to just ship with your project (and of course the license allows for this).
0.9.8 - This release contains a new timer that can be enabled to profile Processor execution
time. Simply pass the “timed=True” parameter to the World on instantiation, and a new World.process_times dictionary will be available. This contains the total execution time of each Processor in milliseconds, and can be logged, printed, or displayed on screen as is useful. It’s useful to see a quick profile of which processors are using the most cpu time, without fully profiling your game. This release also contains some consolidations and cleanups for the benchmarks.
0.9.7 - By default, entities are now lazily deleted. When calling World.delete_entity(entity_id),
Entities are now placed into a queue to be deleted at the beginning of the next call to World.process(). This means it is now safe to delete entities even while iterating over components in your processors. This should allow for cleaner Processor classes, by removing the need to manually track and delete “dead” Entities after iteration. If you do wish to delete an Entity immediately, simply pass the new optional immediate=True argument. Ie:, immediate=True).

1) Compatibility

Esper is developed for Python 3. It will also work on Pypy3. Being written in pure Python, it should work on any compliant interpreter. Continuous Integration (automated testing) is done for both CPython and PyPy. Python 2 is not supported, but Christopher Arndt is currently maintaining a branch here:

2) Installation

No installation is necessary. Esper is a tiny library with no dependencies. Simply copy the esper directory into the top level of your project folder, and import esper.

If you prefer, Esper is also available on PyPI for easy installation via pip.

3) Project Structure

  • World

A World is the main point of interaction in Esper. After creating a World object, you will use that object to create Entities and assigning Components to them. A World is also assigned all of your Processor instances, and handles smoothly running everything with a single call per frame. Of course, Entities, Components and Processors can be created and assigned, or deleted while your application is running.

  • Entities

Entities are simple integer IDs (1, 2, 3, 4, etc.). Entities are “created”, but they are generally not used directly. Instead, they are simply used as IDs in the internal Component database, to track collections of Components. Creating an Entity is done with the World.create_entity() method.

  • Components

Components are defined as simple Python classes. In keeping with a pure Entity System design philosophy, they should not contain any logic. They might have initialization code, but no processing logic whatsoever. A simple Component might look like:

class Position:
    def __init__(self, x=0.0, y=0.0):
        self.x = x
        self.y = y
  • Processors

Processors, also commonly known as “Systems”, are where all processing logic is defined and executed. All Processors must inherit from the esper.Processor class, and have a method called process. Other than that, there are no restrictions. All Processors will have access to the World instance, to allow easy querying of Components. A simple Processor might look like:

class MovementProcessor(esper.Processor):
    def __init__(self):

    def process(self):
        for ent, (vel, pos) in, Position):
            pos.x += vel.x
            pos.y += vel.y

In the above code, you can see the standard usage of the World.get_components() method. This method allows efficient iteration over all Entities that contain the specified Component types. You also get a reference to the Entity ID for the current pair of Velocity/Position Components, in case you should need it. For example, you may have a Processor that will delete certain Entites. You could add these Entity IDs to a list, and call the method on them after you’re done iterating over the Components.

4) Usage

The first step after importing Esper is to create a World instance. You can have a single World instance for your entire game, or you can have a separate instance for each of your game scenes. Whatever makes sense for your design. Create a World instance like this:

world = esper.World()

Create some Processor instances, and assign them to the World. You can specify an optional processing priority (higher numbers are processed first). All Processors are priority “0” by default:

movement_processor = MovementProcessor()
collision_processor = CollisionProcessor()
rendering_processor = RenderingProcessor()
world.add_processor(movement_processor, priority=2)
world.add_processor(collision_processor, priority=3)

Create an Entity, and assign some Component instances to it:

player = world.create_entity()
world.add_component(player, Velocity(x=0.9, y=1.2))
world.add_component(player, Position(x=5, y=5))

Optionally, Component instances can be assigned directly to the Entity on creation:

player = world.create_entity(
    Velocity(x=0.9, y=1.2),
    Position(x=5, y=5)

Running all Processors is done with a single call to world.process(). This will call the process method on all assigned Processors, in order of their priority:


Note: You can pass any args you need to world.process(), but you must also make sure to recieve them properly in the process() methods of your Processors. For example, if you pass a delta time argument as world.process(dt), your Processor’s process() methods should all receive it as: def process(self, dt):

  • Additional Methods

Have a look through esper/ for an idea of what additional functionality is available. All methods have docstrings with details on usage, which will be put into a real API document at some point. Here is a quick list of the methods, whose names should be semi-explanitory:

World.add_component(entity, component_instance)
World.remove_component(entity, ComponentType)
World.get_components(ComponentTypeA, ComponentTypeB, Etc)
World.component_for_entity(entity, ComponentType)
World.has_component(entity, ComponentType)

5) Examples

See the /examples folder to get an idea of how the basic structure of a game looks.

Project details

Release history Release notifications

This version
History Node

History Node


History Node


Download files

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

Filename, size & hash SHA256 hash help File type Python version Upload date
esper- (24.9 kB) Copy SHA256 hash SHA256 Source None Oct 21, 2017

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging CloudAMQP CloudAMQP RabbitMQ AWS AWS Cloud computing Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page