Automated Fault Localization Using Potential Invariants

TL;DR

This paper introduces a method for fault localization in programs using potential invariants, aiming to improve debugging by analyzing program traces, though initial results indicate more research is needed.

Contribution

The paper proposes a novel approach to fault localization leveraging potential invariants and implements a tool based on Ernst's concept, advancing the use of invariants in debugging.

Findings

Experiments so far have been unsatisfactory.

Further research is needed to improve invariant-based fault localization.

The method provides a general framework for analyzing program traces.

Abstract

We present a general method for fault localization based on abstracting over program traces, and a tool that implements the method using Ernst's notion of potential invariants. Our experiments so far have been unsatisfactory, suggesting that further research is needed before invariants can be used to locate faults.