reactivepy

A reactive Python kernel. Whenever a variable value is changed, the kernel automatically executes its dependencies (any cells which use that variable) with the updated value. As of now, reactivepy can also support asynchronous functions.

Each notebook cell for a reactive kernel contains a single definition of a variable, function, or class.

When a cell is run, the kernel conducts static analysis to automatically extract and run dependencies -- saving users the trouble of remembering dependencies and re-running individual cells.

Dependencies

ipython>=4.0.0
jupyter_client
tornado>=4.0
ipykernel>=4.8

Install

git clone https://github.com/jupytercalpoly/reactivepy.git

# install dependencies
`Python 3.6` or above.

pip install .

Develop

git clone https://github.com/jupytercalpoly/reactivepy.git

# install dependencies
`Python 3.6` or above.

pip install -e .

TODO

Move execution out of a queue, this will allow executions to interleave
Support deleting unnamed nodes (nodes which do not export any variables). This will reduce space usage and reduce unnecessary computation
Fix error with being unable to define a node after initially erroring out while creating it

Testing

Requirements

Install pytest and pytest-asyncio using conda or pip:

conda install pytest pytest-asyncio

Then run

pytest .

gnestor / reactivepy