One major effort of Software Engineering has been to provide a way of optimally separating the logical parts that constitute a software. This issue becomes a question of outstanding importance in middle/large interactive software, where an agent like an human user interacts with the software, likely through a Graphic User Interface (GUI). To separate the logic (data) of an application from the representation of those data, several Architectural Patterns [1] have been studied; one is the Model–View–Controller (MVC) pattern that splits the application into three separate parts, the Model, the View and the Controller. The level of mutual knowledge among these three entities is kept as minimal as possible, and this results in:
To improve re-usability and robustness of software, the Observer pattern is used as well. This pattern identifies two entities: the Observable data and the Observer over those data.
The implementation of the Observer pattern is intended to make the Observer take some action when the Observable data change. This triggering mechanism is a further abstraction layer that helps to design and implement robust software.
A possible use of the Observer pattern is in combination with the MVC pattern, where the model communicates indirectly with the presentation side through the observer pattern.
gtkmvc is a framework that implements both the MVC and Observer patterns to be used to produce middle/large applications in Python and PyGTK.
The main goal is to provide a minimal (but not trivial) implementation, where practical aspects are taken into serious consideration, and complexity is kept behind the scene. This makes the users able to mainly focus their attention on the application they need to produce, instead of dealing with the underlying framework.
Pointless to say that goals are clear and likely easy to share by anyone. It is quite a different thing to prove that those goals are reached by the proposed framework. Frankly, this is up to the reader to decide.
Footnotes
| [1] | http://en.wikipedia.org/wiki/Architectural_pattern_(computer_science%29 |