Satisfying the graphical requirements of visual languages in the DV-Centro Framework

The graphical requirements for implementing a visual language include defining the visual elements of the language, the rules for connecting them together, and the graphical relationships that must be maintained once they are connected. The solutions for these graphical requirements are intricate and inherently difficult to implement. The resulting implementations tend to be so specialized as to be applicable only to a single visual language. The DV-Centro Framework makes it possible to implement sophisticated visual languages without having to develop all of this machinery. We examine the patterns of interaction between the components of the DV-Centro Framework. We introduce the Supervisor-Agent pattern as a means of understanding these interactions. Through this pattern we see how it is possible to assemble sophisticated application specific visual language elements from DV-Centro library objects. We also see how it is possible to provide the low level user interface for interacting with the visual language elements as library components