pyblueprints

synopsis:	Provides a layer to abstract the Python developer from the graph database system used

Following the set of interfaces provided by tinkerpop for Blueprints, this project aims to give Python developers a similar functionality. A set of abstract classes are defined in order to guide the design of implementations for the different graph database engines.

Features

This is an experimental version only permitting partial functionality to:

• Rexster infrastructure, supporting every database supported by Rexster (https://github.com/tinkerpop/rexster/)
• Neo4j database providing abstraction over the neo4j-rest-client API.

Please keep in mind to backup your data before trying this library.

Installation

The easiest way to get pyblueprints installed in your virtualenv is by:

pip install pyblueprints

Usage

This version of pybluerprints allows you to connect to graph databases by a Rexster Instance or through the neo4j-rest-client API. Therefore a Neo4j database can be accessed with both options, although the Neo4j transactional mode is only available through the later. The Rexster instance also provides connection to the following databases:

• TinkerGraph
• OrientDB
• DEX
• Sail RDF Stores

Rexster

Connecting to a Rexster instance

>>> from pyblueprints import RexsterServer, RexsterGraph, RexsterIndexableGraph
>>> #Connecting to server
>>> HOST = 'http://localhost:8182'
>>> server = RexsterServer(HOST)
>>> #List graphs availbale in server
>>> server.graphs()
[u'tinkergraph', u'gratefulgraph', u'tinkergraph-readonly', u'sailgraph', u'emptygraph']
>>> #Connecting to a given graph
>>> graph = RexsterIndexableGraph(server, 'tinkergraph')

neo4j-rest-client

Creating a graph object through the neo4j-rest-client API

>>> from pyblueprints.neo4j import Neo4jGraph
>>> graph = Neo4jGraph('http://localhost:7474/db/data')

Creating an indexable graph object through the neo4j-rest-client API

>>> from pyblueprints.neo4j import Neo4jIndexableGraph
>>> graph = Neo4jIndexableGraph('http://localhost:7474/db/data')

The available classes are:

• Neo4jGraph
• Neo4jIndexableGraph
• Neo4jTransactionalGraph
• Neo4jTransactionalIndexableGraph

code examples

Add/Remove Vertex

>>> vertex = graph.addVertex()
>>> graph.removeVertex(vertex)

Add/Remove Edge

>>> v1 = graph.addVertex()
>>> v2 = graph.addVertex()
>>> newEdge = graph.addEdge(v1, v2, 'myLabel')
>>> graph.removeEdge(newEdge)

Vertex Methods

>>> graph= Neo4jGraph(HOST)
>>> v1 = graph.addVertex()
>>> v2 = graph.addVertex()
>>> newEdge = graph.addEdge(v1, v2, 'myLabel')
>>> vertex = graph.getVertex(_id)
>>> # get methods return a generator function
>>> edge = list(vertex.getBothEdges())[0]
>>> edge = list(vertex.getOutEdges())[0]
>>> edges = list(vertex.getInEdges())

Vertex/Edges properties

>>> vertex_id = vertex.getId()
>>> vertex.setProperty('name', 'paquito')
>>> print vertex.getPropertyKeys()
>>> print vertex.getProperty('name')
>>> vertex.removeProperty('name')

Edge Methods

>>> outVertex = edge.getOutVertex()
>>> inVertex = edge.getInVertex()
>>> print getLabel()

Add/Remove Manual Index

>>> index = graph.createManualIndex('myManualIndex', 'vertex')
>>> graph.dropIndex('myManualIndex', 'vertex')

Index Methods

>>> index = graph.getIndex('myManualIndex', 'vertex')
>>> vertex = graph.addVertex()
>>> index.put('key1', 'value1', vertex)
>>> print index.count('key1', 'value1')
>>> print index.getIndexName()
>>> print index.getIndexClass()
>>> print index.getIndexType()
>>> # get returns a generator function
>>> vertex2 = list(index.get('key1', 'value1'))[0]
>>> index.remove('key1', 'value1', vertex)

Transactional Methods

>>> graph= Neo4jTransactionalGraph(HOST)
>>> graph.startTransaction()
>>> v = graph.addVertex()
# Stoping calls the commit
>>> graph.stopTransaction()
>>> vertexId = v.getId()
>>> v = graph.getVertex(vertexId)
>>> graph.startTransaction()
>>> v.setProperty('p1', 'v1')
>>> graph.stopTransaction()