Conversions between NetworkX, nx-arangodb, and nx-cugraph

This page describes how to convert graphs across ecosystems using nx_arangodb.convert for interoperability and performance.

Overview

_to_nx_graph: nxadb.Graph | nx.Graph -> nx.Graph
_to_nxadb_graph: nx.Graph | nxadb.Graph -> nxadb.Graph
_to_nxcg_graph: nxadb.Graph | nxcg.Graph -> nxcg.Graph (requires nx-cugraph and CuPy)

Behavior notes

Database-backed nxadb graphs: converting to NetworkX pulls node and adjacency dictionaries from ArangoDB and constructs a plain NetworkX graph.
Local nxadb graphs (graph_exists_in_db is False) are already NetworkX-compatible and may be returned as-is.
Directed/multi settings are preserved; as_directed forces directed variants without changing multiplicity.
If nx-cugraph or CuPy is unavailable, GPU conversions are disabled and _to_nxcg_graph raises NotImplementedError.

Performance and caching

Database pulls are timed and logged.
nxadb_to_nx returns a standard NetworkX graph and does not re-wrap the custom dict implementations in nx_arangodb.classes.dict.
nxadb_to_nxcg can cache the constructed GPU graph when G.use_nxcg_cache is True on the nxadb.Graph instance.

Examples

import networkx as nx
import nx_arangodb as nxadb

# Start with a NetworkX graph
G_nx = nx.karate_club_graph()

# Convert to nx-arangodb (optionally force directed)
G_adb = nxadb.convert._to_nxadb_graph(G_nx, as_directed=False)

# Convert back to NetworkX
G_nx2 = nxadb.convert._to_nx_graph(G_adb)

# Convert to nx-cugraph if available
try:
    G_cg = nxadb.convert._to_nxcg_graph(G_adb)
except NotImplementedError:
    # nx-cugraph/CuPy not available
    pass

API Reference

nx_arangodb.convert._to_nx_graph(G: Any, *args: Any, **kwargs: Any) → Graph[source]

Convert a graph to a NetworkX graph.

Parameters

GAny

The graph to convert.

Currently supported types: - nx.Graph - nxadb.Graph

Returns

nx.Graph: The converted graph.

nx_arangodb.convert._to_nxadb_graph(G: Any, *args: Any, as_directed: bool = False, **kwargs: Any) → Graph[source]

Convert a graph to a NetworkX-ArangoDB graph.

Parameters

GAny

The graph to convert.

Currently supported types: - nx.Graph - nxadb.Graph

as_directedbool, optional

Whether to convert the graph to a directed graph. Default is False.

Returns

nxadb.Graph: The converted graph.

nx_arangodb.convert._to_nxcg_graph(G: Any, as_directed: bool = False) → nxcg.Graph[source]

nx_arangodb.convert.nx_to_nxadb(graph: Graph, *args: Any, as_directed: bool = False, **kwargs: Any) → Graph[source]

Convert a NetworkX graph to a NetworkX-ArangoDB graph.

Parameters

graphnx.Graph: The NetworkX graph to convert.
as_directedbool, optional: Whether to convert the graph to a directed graph. Default is False.

Returns

nxadb.Graph: The converted graph.

nx_arangodb.convert.nxadb_to_nx(G: Graph) → Graph[source]

Convert a NetworkX-ArangoDB graph to a NetworkX graph.

This function will pull the graph from the database if it does not exist in the cache. A new NetworkX graph will be created using the node and adjacency dictionaries that are fetched.

NOTE: The current downside of this approach is that we are not able to take advantage of the custom Dictionary classes that we have implemented in nx_arangodb.classes.dict. This is because the node and adjacency dictionaries are fetched as regular Python dictionaries. Furthermore, we don’t cache the dictionaries themselves, so we have to fetch them every time we convert the graph, which is currently being invoked on every algorithm call. See the note below for a potential solution. As a temporary workaround, users can do the following:

```

import networkx as nx import nx_arangodb as nxadb

G_ADB = nxadb.Graph(name=”MyGraph”) # Connect to the graph G_NX = nxadb.convert._to_nx_graph(G_ADB) # Pull the graph

nx.pagerank(G_NX) nx.betweenness_centrality(G_NX) …

```

Parameters

Gnxadb.Graph: The NetworkX-ArangoDB graph to convert.

Returns

nx.Graph: The converted graph.