rebecca.sqla 0.1

INSTALL

Install using pip or easy_install:

$ pip install rebecca.sqla
$ easy_install rebeca.sqla

USAGE

rebecca.sql needs sqlalchemy.orm.scoped_session.

specify session object and dsn string in settings:

sqlalchemy.url = 'sqlite:///'
rebecca.sqla.session = your.app.models.DBSession

rebecca.sqla has include hook:

config.include("rebecca.sqla")

this hook sets up the bind configuration of DBSession to sqlalchemy.url.

ModelLoader and SAContext

rebecca.sqla creates attribute named “sa” to request.context. context.sa is SAContext to load models from ModelLoader registered by config.add_model_loader.

here is example models:

class Person(Base):
    __tablename__ = 'person'
    id = Column(Integer, primary_key=True)
    name = Column(Unicode(255))

and example route:

config.add_route('person', '/person/{person1_id}/{person2_id}')

register model loaders:

config.add_model_loader('person1', Person, [('person1_id', 'id')])
config.add_model_loader('person2', Person, [('person2_id', 'id')])

add_model_loader takes 3 arguments. first is attr name of SAContext, second is a model loaded from ModelLoader and last is mappings that matchdict param to model attribute.

second argument can be dotted name.

in view, context has SAContext named “sa” attribute, that has properties of registered models. ModelLoader s load model with query using matchdict parameters.

def person(context, request):
    person1 = context.sa.person1
    person2 = context.sa.person2