Skip to main content

Abstract Factory design pattern classes for scalable data in dynamic environments.

Project description

Abstract Factories

A collection of classes to support the Abstract Factory design pattern, providing a clear abstraction layer for scalable data in dynamic environments.

abstract_factories will auto-register viable items from any given python module and/or path.

  • Tested on Python 3.8 - 3.12
  • Functional on Python 2.7

    Wait - what? Python 2.7? What year is this?
    I often work professionally on legacy systems that are too fragile or large to update. Providing there's no functional or notable impact supporting 2.7, I have no reason to ignore its existence yet.


Installation

pip install abstract-factories

Usage

Initialize AbstractTypeFactory or AbstractInstanceFactory with the abstract type.
Optionally, provide the attribute/method name to identify items by name (and optionally version).

Registering items can be done directly.

from abstract_factories import AbstractTypeFactory

class AbstractVehicle(object):
    def start(self):
        raise NotImplementedError()

class Car(AbstractVehicle):
    def start(self):
        print('Vrooom...')

# Type Factory
type_factory = AbstractTypeFactory(AbstractVehicle)
type_factory.register_item(Car)
assert type_factory.get('Car') is Car

Or automatically by providing python modules or paths in which the factory can search.

from abstract_factories import AbstractTypeFactory
from . import my_vehicle_package

# Type Factory
type_factory = AbstractTypeFactory(my_vehicle_package.AbstractVehicle)

# Find any AbstractVehicle subclasses in `my_vehicle_package` and register them.
type_factory.register_module(my_vehicle_package)
assert type_factory.get('Car') is my_vehicle_package.Car

# Can also find any AbstractVehicle subclasses in a directory and register those too.
type_factory.register_path('c:/Users/user/downloads/other_vehicles')

In some use-cases, instances are a much better fit for the type of data you want to use in your factory.
In that case, use AbstractInstanceFactory.

from abstract_factories import AbstractInstanceFactory

class AbstractVehicle(object):
    def __init__(self, make=None):
        self.make = make

    def start(self):
        raise NotImplementedError()

class Car(AbstractVehicle):
    def start(self):
        print('Vrooom...')

# Instance Factory
honda = Car('Honda')
instance_factory = AbstractInstanceFactory(AbstractVehicle, name_key='make')
instance_factory.register_item(honda)
assert instance_factory.get('Honda') is honda

Registration:

Register viable items directly.

  • type_factory.register_item(AbstractSubclass)
  • instance_factory.register_item(AbstractSubclass())

Find and register any viable items found in the module's locals.

  • type_factory/instance_factory.register_module(module)

Find and register any viable items found in any nested python file from a dynamic import. Some limitation using relative imports.

  • type_factory/instance_factory.register_path(r'c:/tools/tool_plugins')
  • type_factory/instance_factory.register_path(r'c:/tools/tool_plugins/plugin.py')

Practical Applications

Content Creation

Useful for managing production needs in Film, TV, and Games, allowing easy modifications and versioning of components.

Rigging

Easily support and modify rig component behaviours during production.

import sys
from abstract_factories import AbstractTypeFactory

class AbstractRigComponent:
    Name = 'AbstractRigComponent'
    Version = 0

    def build(self, **kwargs):
        raise NotImplementedError()

class IKChainComponent(AbstractRigComponent):
    Name = 'IKChainComponent'
    Version = 1

    def build(self, **kwargs):
        pass

# --------------------------------------------------------------------------
class RigComponentBuilder:
    def __init__(self):
        self.factory = AbstractTypeFactory(
            abstract=AbstractRigComponent,
            modules=[sys.modules[__name__]],
            name_key='Name',
            version_key='Version'
        )

    def build(self, component_data):
        results = []
        for data in component_data:
            component = self.factory.get(
                data.pop('type'), 
                version=data.pop('version', None),
            )
            instance = component()
            instance.build(**data)
            results.append(instance)
        return results

builder = RigComponentBuilder()
builder.build([
    {'type': 'IKChainComponent', 'name': 'arm'},
    {'type': 'IKChainComponent', 'name': 'leg', 'version': 2},
])

Testing

.tests/ directory contains examples for;

  • Adding, removing & comparing items directly.
  • Adding, removing & comparing items found in modules and/or paths.

Further Information

Abstract factories is influenced by https://github.com/mikemalinowski/factories.


License

This project is licensed under the MIT License.

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

abstract_factories-0.4.0.tar.gz (11.2 kB view hashes)

Uploaded Source

Built Distribution

abstract_factories-0.4.0-py2.py3-none-any.whl (9.7 kB view hashes)

Uploaded Python 2 Python 3

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