Incremental Service Composition Based on Partial Matching of Visual Contracts

Services provide access to software components that can be discovered dynamically via the Internet. The increasing number of services a requester may be able to use demand support for finding and selecting services. In particular, it is unrealistic to expect that a single service will satisfy complex requirements, so services will have to be combined to match clients’ requests. In this paper, we propose a visual, incremental approach for the composition of services, in which we describe the requirements of a requester as a goal which is matched against multiple provider offers. After every match with an offer we decompose the goal into satisfied and remainder parts. We iterate the decomposition until the goal is satisfied or we run out of offers, leading to a resolution-like matching strategy. Finally, the individual offers can be composed into a single combined offer and shown to the requester for feedback. Our approach is based on visual specifications of pre- and postconditions by graph transformation systems with loose semantics, where a symbolic approach based on constraints is used to represent attributes and their computation in graphs.