Usage

Installation

pip install streamlit-flow-node-graph

Create a graph element

# define items the user can use
item_types = [
        {"title":'Test 1',"color":"rgb(255,0, 192)", "port_selection" : 'out'},
        {"title":'Test 2',"color":"rgb(255, 0, 192)", "port_selection" : 'in'},
        {"title":'Test 3',"color":"rgb(0,255, 192)", "port_selection" : 'both'},
    ]

diagram = node_graph.node_graph(model=model, item_types=item_types, key='test' )
st.write(diagram.selected) # access currently selected node id
st.write(diagram.model) # access the underlysing model

Features

• specify a node icon with the icon property e.g. {"title":'Purchase',"color":"rgb(255,0, 192)", "port_selection" : 'out', "icon" : 'shopping-cart'} (icons in the font-awesome collection can be used)
• use diagram utils

diagram = node_graph.node_graph(model=model, item_types=item_types, key='`test`' )
digram.get_selected_node()
diagram.get_all_nodes_node_inputs()
formated_nodes, formated_links = diagram.process_diagram_output()

Contribute

Start dev

cd node_graph/frontend && npm install && npm run start

uncomment # _RELEASE = True in node_graph/__init__.py and in another shell streamlit run node_graph/__init__.py

Build

cd node_graph/frontend && npm build cd../.. rm -rf dist/* python setup.py sdist bdist_wheel

Upload

1. Change version in setup.py
2. twine upload dist/*

Developer

Deployment

https://docs.streamlit.io/library/components/publish

python setup.py sdist bdist_wheel

python -m twine upload dist/*