Model at runtime computation

PAMELA model at runtime is computed dynamically, working on the classpath of launched java application, and starting from a simple java interface (or a collection of java interfaces) which is/are PAMELA-annotated. From a mathematical point of view, internal representation of the underlying model is a graph whose vertex are PAMELA ModelEntities (annotated java interface), and edges are either inheritance links or reference links (a property whose type is another ModelEntity). @Imports and @Import annotations allows to include some other ModelEntities in the model. On the contrary, an annotation attribute @Getter(...ignoreType=true) allows to ignore the link. In that context, PAMELA model computation is a graph closure computation, starting from a collection of vertices.

A PAMELA model at runtime is represented by a ModelContext. PAMELA instances (instances of ModelEntity) are handled though the use of ModelFactory, which is instantiated from a ModelContext.

PAMELA model closure computation on-the-fly provides an interesting approach to deal with model fragmentation.