Camelot makes it easy to create views for any type of Python objects.
SQLAlchemy is a very powerful Object Relational Mapper (ORM) with lots of possibilities for handling simple or sophisticated datastructures. The SQLAlchemy website has extensive documentation on all these features. An important part of Camelot is providing an easy way to create views for objects mapped through SQLAlchemy.
SQLAlchemy comes with the Declarative extension to make it easy to define an ORM mapping using the Active Record Pattern. This is used through the documentation and in the example code.
An alternative to Declarative is Elixir, which was used in previous Camelot versions, and is still supported.
The code above defines the model for a GeographicBoundary class, a base class that will be used later on to subclass into Countries and Cities, and is part of the default Persons and Organizations data model of Camelot. This code has some things to notice :
- GeographicBoundary is a subclass of Entity, Entity is the base class for all classes that are mapped to the database
- the using_options statement allows us to fine tune the ORM, in this case the GeographicBoundary class will be mapped to the geographic_boundary table
- The Field statement add fields of a certain type, in this case Unicode, to the GeographicBoundary class as well as to the geographic_boundary table
- The ColumnProperty full_name is a more advanced feature of the ORM, when an GeographicBoundary object is read from the database, the ORM will ask the database to concatenate the fields code and name, and map the result to the full_name attribute of the object
- The __unicode__ method is implemented, this method will be called within Camelot whenever a textual representation of the object is needed, eg in a window title or a many to one widget
When a new Camelot project is created, the camelot-admin tool creates an empty models.py file that can be used as a place to start the model definition.