PyFlow is a general purpose visual scripting framework for python.
Base structure of program implemented, such things as packages discovery, input customization, exporters, base classes for entities and tools.
You can completely remove nodes and tools that ships by default and start your own library from scratch.
You can evaluate programs without running user interface.
We have simple CLI interface and setup.py script.
Expectedly, we faced problems with our runtime execution model, it is good for certain types of tasks despite it is slow. Also, complicated flow control requires cycles handling.
For example current for loop node implementation doesn't work well with any kind of
delay stuff in it's body. Of course it can be solved by adding more logic to loop processing, but we don't know exactly how to make it efficient yet.
We will add new types of graphs (undirected graphs, trees) and algorithms to work with those structures.
Code generation is probably the future of PyFlow.
We are going to implement more building blocks. Functions - which unlike subgraphs will have single definition and multiple entry points across the program. Objects - very important concept of programming in general, object will have own methods, access levels and variables. Events and delegates - similar to ue4.
Application presets and api for writing own presets. Templates like in any IDE (shader, command line tool, software installer, etc.).
User interface improvements and new tools to enhance user experience.
We want to have most python standard library wrapped into a nodes!
Clone or download repository.
Install requirements for your use case.
requirement lists can be found under
To run the program in standalone mode, run
pyflow.py. It can be found in the root folder of repository.
Or using pip
Latest development version. Contains latest features
pip install git+https://github.com/wonderworks-software/PyFlow.git@master
Latest release version
pip install git+https://github.com/wonderworks-software/PyFlow.git@release
After package is installed, you can invoke
pyflow.exe on windows or
pyflow on unix OS. Program is located inside
Add own node sets, pins, shelf tools, dock tools, preferences widgets, data exporters/importers
PyFlow is a normal python program. It can run under any python environment
Python3 and 2 support
Logic and UI are separated
Main logic elements (Node and Pin) has non gui representation, so programs can be evaluated without GUI
Collapse parts of graphs into subgraphs
Watch what is going on
Expose properties from subgraphs
Any node can share it's properties with owning sub graph node
Share subgraphs with others
If compound node was exported to some package, it will appear in node box
Write whatever code directly into a node
Python node import/export
Share python nodes with others
If python node was exported to some package, it will appear in node box
Fast node generation from functions
Single decorator to expose node from function
Package loading from arbitrary locations
Packages can be loaded from provided list of directories. It can be your development folder or folder in network drive inside your studio
Custom input system
If you got used to specific keymap of your software of choice, you can easily remap PyFlow input actions
Drop on wire
Automatically сonnect the node by dragging it onto the wire
Easy to use powerful undo system
By a single line call
Modify node and pins attributes using property panels
Save arbitrary data into a variables
Editable shareable themes
Create and share themes with others
Editor data can be exported/imported to/from arbitrary format
Command line interface
Exported programs can be edited and evaluated from command line without gui
Convenient wires interaction
Node deprecation flag
Tell users to update their programs
Plugin wizards framework
Generate package template to start development. Or create own wizard for your module
Easy configuration files framework
While adding own modules you probably want to have some configuration data. We provide easy to use api for this
We have documentation
We use a semantic versioning policy.
See also the list of contributors who participated in this project.