The UniForM WorkBench - A Higher Order Tool Integration Framework

The UniForM Workbench is an open ended tool integration framework for developing (formal) Software Development Environments (SDE) from the basis of prefabricated oo-the-shelf development tools. The integration framework provides support for data, control and presentation integration as well as utilities for wrapping Haskell interfaces around existing development tools. Entire SDE's are then glued together on the basis of these encapsulations using Concurrent Haskell as the integration language, thus allowing integration to be done at a level of abstraction that is very close to the one ooered by constructive formal speciications. So far, the integration framework has successfully been used to integrate tools for Haskell program development as well as spec-iication and proof tools for Z speciications. During the 80's there were several attempts to provide environments for synthesizing tightly integrated SDE's from the basis of abstract language speciica-tions. The Synthesizer Generator (CSG) 46], Gandalf 13], PSG 2] and IPSEN 28] are prominent examples of this approach. They are not perfect solutions, due either to the development costs involved in requiring tools to be developed from scratch in a homogeneous language framework or due to the inapplicability of the language framework itself (e.g. the attribute grammars of CSG) in handling the problem domain proper (e.g. proof tools) 25]. More recent systems, such as Centaur 5] and the ASF+DSF environment 8], are more open by providing a toolbus for hooking in foreign tools. However, both approaches lack the required features for supporting control and data integration "in the large" and "in the many". In the context of a formal program development environment, no single tool or language is believed to be suuciently powerful enough to support an entire formal development process, rather it is more likely that several loosely coupled and prefabricated tools are put together, each supporting a single dedicated task of the overall development 26]. This may suggest that SDE's for formal methods should be based on a loosely coupled architecture 27, 48] where prefabricated