Un algorithme incrémental dirigé par les flots et basé sur les contraintes pour l'aide à la localisation d'erreurs
In this article, we present our improved algorithm for error localization from counterexamples, LocFaults, flow-driven and constraint-based. This algorithm analyzes the paths of CFG (Control Flow Graph) of the erroneous program to calculate the subsets of suspicious instructions to correct the program. Indeed, we generate a system of constraints for paths of control flow graph for which at most k conditional statements can be wrong. Then we compute the MCSs (Minimal Correction Set) of bounded size on each of these paths. Removal of one of these sets of constraints gives maximal satisfiable subset, in other words, a maximal satisfiable subset satisfying the postcondition. To calculate the MCSs, we extend the generic algorithm proposed by Liffiton and Sakallah in order to deal programs with numerical instructions more effectively. We are interested to present the incremental aspect of this new algorithm that is not yet presented.