Graph API, serializer, and loader that conforms to the JSON Graph Spec.

pip install graphlib

The object API makes it simple to create nodes and relationships between them.

from graphlib import Node

city = Node({'location': 'Philadelphia'})
jane = Node({'name': 'Jane'})
john = Node({'name': 'John'})
bob = Node({'name': 'Bob'})

city.relate([jane, john, bob], 'LIVES_IN')
jane.relate(john, 'MARRIED_TO')
john.relate(bob, 'FRIENDS_WITH')

Serializes the object reprsentations into dicts and lists which can be encoded as JSON or loaded into a graph database.

from graphlib import serialize

data = serialize(city)

Takes the serialized data and loads it into a database. This example loads it into Neo4j.

from graphlib import neo4j

neo4j.load(data)

The Neo4j module can be used directly via the command line:

python -m 'graphlib.neo4j' [path/to/file.json] [--load] [uri]

By default, stdin will be read which should be valid JSON that will be parsed and converted into Cypher statements and printed to stdout. If a path supplied, the file will be read instead of stdin. If the --load flag is present, the statements will be executed on the Neo4j server at the default URI unless a custom URI is provided.