Behavioral Similarity Matching using Concrete Source Code Templates in Logic Queries

Printer-friendly version

Publication Type:

Conference Paper

Authors:

Coen De Roover; Johan Brichau; Carlos Noguera; Theo D'Hondt; Duchien, Laurence

Source:

Proceedings of the ACM-SIGPLAN Symposium on Partial Evaluation and semantics-based Program Manipulation (PEPM07) (2007)

Keywords:

logic meta programming, pattern detection, program analysis, program querying, program validation, source code templates

Abstract:

<p>Program query languages and pattern-detection techniques are an essential part of program analysis and manipulation systems. Queries and patterns permit the identification of the parts of interest in a program's implementation through a representation dedicated to the intent of the system (e.g. call-graphs to detect behavioral flaws, abstract syntax trees for transformations, concrete source code to verify programming conventions, etc). This requires that developers understand and manage all the different representations and techniques in order to detect various patterns of interest. To alleviate this overhead, we present a logic-based language that allows the program's implementation to be queried using concrete source code templates. The queries are matched against a combination of structural and behavioral program representations, including call-graphs, points-to analysis results and abstract syntax trees. The result of our approach is that developers can detect patterns in the queried program using source code excerpts (embedded in logic queries) which act as prototypical samples of the structure and behavior they intend to match.</p>