The program understanding problem: analysis and a heuristic approach

Abstract

Program understanding is the process of making sense of a complex source code. This process has been considered as computationally dificult and conceptually complex. So fal; no formal complexity results have been presented, and conceptual models differ from one researcher to the next.In this paper we formally prove that program understanding is NP-had. Furthermore, we show that even a much simpler subproblem remains NP-hard. Howevel; we do not despair by this result, but rathel; offer an attractive problem-solving model for the program understanding problem. Our model is built on a framework for solving Constraint Satisfaction Problems, or CSPs, which are known to have interesting heuristic solutions. Specifically, we can represent and heuristically address previous and new heuristic approaches to the program understanding problem with both existing and specially designed constraint propagation and search algorithms.

Extracting Constraint Knowledge from Code: A Case-Based Reasoning Approach

“…A case can be thought of as several inferences chunked together, which can be used to solve a new problem in one step rather than many small inference steps. The case holds these inferences together so that the system does not have to search the space of possible inferences, which could lead to combinatorial explosion (code understanding is an NP-complete problem [9]). …”

Improving program comprehension by combining code understanding with comment understanding

“…Research often included comparing data flow and/or control flow graphs derived from analysis of a source program to an a priori library of known constructs [12,13,20]. However, much research in this golden age of code understanding (mid-1980s to mid-1990s) focused on formal non-heuristic program comprehension which was shown by Woods and Yang [27] to be NP-hard. For this reasons, heuristic approaches acquired new importance [27].…”

Program understanding as constraint satisfaction

“…• Woods and Yang (1996a) present an examination of the complexity of the program understanding task and a discussion of how the various problem aspects contribute to problem complexity. The paper suggests the likely success of using a constraint-based control of search as a means of mediating this complexity in practice.…”

“…• Woods et al (1997) is a mature, carefully restructured review of the interpretation of program understanding as a process of constraint-based concept recovery. This book brings together the entire range of material published by the authors between 1995 and 1997 in a single, comprehensive framework.…”

“…The basis for such heuristic approaches has been the assumed intractability of the complete understanding problem in general. In (Woods and Yang, 1996b), not only is program understanding shown to be NP-hard, but also the intuitively easier problem of locating partial local understandings.…”