Test Data Generation for Programs with Quantified First-Order Logic Specifications

Abstract

AbstractWe present a novel algorithm for test data generation that is based on techniques used in formal software verification. Prominent examples of such formal techniques are symbolic execution, theorem proving, satisfiability solving, and usage of specifications and program annotations such as loop invariants. These techniques are suitable for testing of small programs, such as, e.g., implementations of algorithms, that have to be tested extremely well.In such scenarios test data is generated from test data constraints which are first-order logic formulas. These constraints are constructed from path conditions, specifications, and program annotation describing program paths that are hard to be tested randomly. A challenge is, however, to solve quantified formulas. The presented algorithm is capable of solving quantified formulas that state-of-the-art satisfiability modulo theory (SMT) solvers cannot solve. The algorithm is integrated in the formal verification and test generation tool KeY .KeywordsDynamic LogicProof ObligationSymbolic ExecutionCore AtomTest Data GenerationThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.