crvidya / python-datafork

Forkable global state

datafork

datafork is a Python utility library for creating data slots whose content can be temporarily "forked", possibly mutated, and then merged back into what was already present.

This allows for code to evaluate "what if" scenarios, and then conditionally merge the result depending on whether the outcome was acceptable.

This model could be thought of as a sort of "transactional memory", though it also evokes the idea of dynamic scope.

The simplest use-case is a transaction block whose side-effects apply only if the block runs to completion without raising an exception:

import datafork

with datafork.root() as root:
    example_slot = root.slot(initial_value=1)
    with root.transaction():
        example_slot.value = 2
        might_raise_exception()
    # this will print 1 if an exception was raised, or 2 if not.
    print example_slot.value

For other examples and reference documentation, see the manual.

Installation

Stable versions of this module can be installed from pypi as datafork.

Development Environment Setup

If you'd like to contribute to this library you can clone this repository and run the following steps to get set up. This assumes you already have Python installed, and virtualenv available.

• virtualenv env
• source env/bin/activate
• pip install -r requirements-dev.txt

You should then be able to run the tests by running nosetests.

Contributing

Contributions are welcome but please ensure that any changes are tested and that code style follows what's already present. The baseline code style for this library is PEP8.