Knowledge is a polymorphic vertical database based on SQLAlchemy. It provides a dict-like interface to the database.
Knowledge is comprised of two related objects: Fact()s and Entity()s. Entities can be created and then have arbitrary facts applied to them. Before Knowledge can be used, the DB has to be set up with SQLAlchemy as follows:
from sqlalchemy import create_engine from knowledge.model import init_model, metadata engine = create_engine('sqlite:///knowledge.db') init_model(engine) metadata.create_all(engine)
Using knowledge is easy. Entities are created with a key, then facts about the entity can be applied like values in a dictionary:
from knowledge.model import Entity, DBSession monster = Entity(u'Monster') fairy = Entity(u'Fairy') rjbean = Entity(u'rjbean') monster[u'color'] = u'Green' monster[u'name'] = u'Lotharrr' fairy[u'flies'] = True fairy[u'name'] = u'Bell' rjbean[u'name'] = u'ralph' rjbean[u'flies'] = False rjbean[u'hacks'] = True DBSession.add(monster) DBSession.add(fairy) DBSession.add(rjbean) DBSession.commit()
Retrieving Entities and Facts from the DB works just like any other SQLAlchemy application:
from knowledge.model import Entity, DBSession # Query all the Entities out of knowledge knowledge_query = DBSession.query(Entity).all() for entity in knowledge_query: print entity, entity.facts.values()