NetworkXGraphs.jl

NetworkXGraphs.jl is a Julia wrapper around Python's networkx built on PythonCall.jl. The current milestone is intentionally narrow: constructors plus the basic Graphs.jl API needed to pass GraphsInterfaceChecker.

networkx is declared as a package dependency and is automatically installed via CondaPkg.jl — no manual Python setup required.

Features

• Wrap networkx.Graph/networkx.DiGraph as Graphs.AbstractGraph using the NetworkXGraph / NetworkXDiGraph constructors
• Convert between Graphs.jl graph types and Python networkx objects via networkx_graph
• Access the raw Python networkx module through NetworkXGraphs.PythonNetworkX
• Validate interface conformance with GraphsInterfaceChecker.jl
• Stress-test multi-threaded use of independent graphs to catch Python/GIL integration regressions

Quick start

using Graphs
using NetworkXGraphs

# Access the raw Python networkx module
nx = NetworkXGraphs.PythonNetworkX.networkx

# Create a Python networkx graph and wrap it as a Graphs.jl-compatible graph
pyg = nx.path_graph(5)
gw = NetworkXGraph(pyg)         # undirected
nv(gw) == 5                     # true

pydg = nx.DiGraph()
pydg.add_edges_from([(1, 2), (2, 3)])
dgw = NetworkXDiGraph(pydg)     # directed

# Convert a Graphs.jl graph to a Python networkx object
g = path_graph(5)
pyg2 = networkx_graph(g)
gw2 = NetworkXGraph(pyg2)
nv(gw2) == nv(g)                # true

Notes

• No graph algorithms are implemented in this package at this stage.