Abusing DotMotif as a Graph Database

Using DotMotif only for its database

UPDATE Sep 2020: If the previous update warning didn't scare you off, then here's some new info! Grand is a new Python library that emulates the NetworkX API (among others), but stores and manages graph information in a variety of backends, including SQL and DynamoDB. If you are still looking for an easy Neo4j provisioning system, you may still want to consider Tamarind.

The rest of this page is left as a legacy documentation, but it sure is a roundabout way of doing things now :)

UPDATE v0.4.2: If you are just using DotMotif as a graph database container-management system, you may want to consider using Tamarind instead. DotMotif uses Tamarind under the hood.

---

DotMotif does a lot of heavy-lifting so that you can ignore the fact that you're using a graph database. This can come in handy when you want to use a graph database but don't want to do work to set it up.

Note that this uses private library APIs and this is not guaranteed to always work.

DotMotif does the following for you:

• Attaches to a graph database if available
• Provisions a docker container if no graph database is available
• Imports your data from a variety of formats
• Authenticates with the database
• Sets a new password (mandatory for new neo4j databases)
• Provides an avenue for you to query the database

Using DotMotif to create a database

We don't need most of the dotmotif capabilities if we're not trying to run a subgraph isomorphism query, so we can ignore the dotmotif class altogether. Instead, run the following:

>>> # Imagine you have a graph of type networkx.DiGraph called G:
>>> G = nx.DiGraph(...)

>>> from dotmotif.executors import Neo4jExecutor
>>> E = Neo4jExecutor(graph=G)

You will need to wait a few minutes now (may be even longer depending on if you have already pulled down the neo4j docker image previously, and, thither, your internet speed).

Once that completes, you will have access to the neo4j shell to execute commands. You can either use this for good (i.e. dotmotif queries) or evil (i.e. anything else you want to run on a graph database). Note that if you are exposing this graph database to the world, this procedure is INSECURE and gives others admin access to the database.

>>> E.run(cypher_query)

...where cypher_query is a query....that you wrote...in cypher.

Take it from the top

An example, given a simple graph:

import networkx as nx
G = nx.DiGraph()
G.add_path([1, 2, 3])

from dotmotif.executors import Neo4jExecutor
E = Neo4jExecutor(graph=G)

E.run("SELECT (A)-[B]->(C) RETURN DISTINCT A, B, C")
# returns [[1, 2], [2, 3]]