Proving the Correctness of Regula Deterministic Programs: a Unifying S Using Dynamic Logic

The simple set WL of deterministic while programs is defined iand a number of known methods for proving the correctness of these programs are surveyed. Emphasis is placed on the tradeoff existing between data-directed and syntax-directed methods, and on providing, especially for the latter, a uniform description enabling comparison and assessment. Among the works considered are the Floyd/Hoare invariant assertion method for partial correctness, Floyd’s well-founded sets method for termination, Dijkstra’s notion of weakest precondition, the Burstall/Manna and Waldinger intermittent assertion method and more. Also, a brief comparison is carried out between three logics of programs: dynamic logic, algorithmic logic and programming logic.