Enhancing Trace Debugging with Algorithmic and Omniscient Debugging

During many years, Print Debugging has been the most used method for debugging. Nowadays, however, industrial languages come with a trace debugger that allows programmers to trace computations step by step using breakpoints and state viewers. Almost all modern programming environments include a trace debugger that allows us to inspect the state of a computation in any given point. Nevertheless, this debugging method has been criticized for being completely manual and time-consuming. Other debugging techniques have appeared to solve some of the problems of Trace Debugging, but they suffer from other problems such as scalability. In this work we present a new hybrid debugging technique. It is based on a combination of Trace Debugging, Algorithmic Debugging and Omniscient Debugging to produce a synergy that exploits the best properties and strong points of each technique. We describe the architecture of our hybrid debugger and our implementation that has been integrated into Eclipse as a plugin.This work has been partially supported by the Spanish Ministerio de Economía y Competitividad (Secretaría de Estado de Investigación, Desarrollo e Innovación) under grant TIN2008-06622-C03-02 and by the Generalitat Valenciana under grant PROMETEO/2011/052. David Insa was partially supported by the Spanish Ministerio de Educación under FPU grant AP2010-4415.González, J.; Insa Cabrera, D.; Silva Galiana, JF. (2013). Enhancing Trace Debugging with Algorithmic and Omniscient Debugging. En Logic-Based Program Synthesis and Transformation. Springer. 183-201. http://hdl.handle.net/10251/7282718320