Exceptions for Dependability. Sekerinski, E. In Petre, L., Sere, K., & Troubitsyna, E., editors, Dependability and Computer Engineering: Concepts for Software-Intensive Systems—a Handbook on Dependability Research, pages 11–35. IGI Global, July, 2011.
Exceptions for Dependability [pdf]Paper  doi  abstract   bibtex   3 downloads  
Exception handling allows a program to be structured such that the original design can be preserved in presence of possibly failing components, it allows for an unobtrusive treatment of rare or undesired cases, and can be used to address imperfections in programs. This chapter develops a theory of exception handling with try-catch statements and shows its use in the design of dependable systems by giving a formal account of the patterns of masking, propagating, flagging, rollback, degraded service, recovery block, repeated attempts, and conditional retry. The theory is based on weakest exceptional preconditions, which are used for both defining statements and for proofs. Proof outlines are introduced and used to establish the correctness of the patterns.

Downloads: 3