Meaningful method names are crucial for the readability and maintainability of software. Existing naming conventions focus on syntactic details, leaving programmers with little or no support in assuring meaningful names. In this paper, we show that naming conventions can go much further: we can mechanically check whether or not a method name and implementation are likely to be good matches for each other. The vast amount of software written in Java defines an implicit convention for pairing names and implementations. We exploit this to extract rules for method names, which are used to identify "naming bugs" in well-known Java applications. We also present an approach for automatic suggestion of more suitable names in the presence of mismatch between name and implementation. © 2009 Springer Berlin Heidelberg.
CITATION STYLE
Høst, E. W., & Østvold, B. M. (2009). Debugging method names. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5653 LNCS, pp. 294–317). https://doi.org/10.1007/978-3-642-03013-0_14
Mendeley helps you to discover research relevant for your work.