论文信息 - Identification and Characterization of Crosscutting Concerns in MATLAB Systems 1

Identification and Characterization of Crosscutting Concerns in MATLAB Systems 1

The current state-of-the-art in aspect mining is well advanced for object-oriented programming languages but until now neglected the MATLAB language. This paper contributes to fill that gap by proposing a novel notion of crosscutting concern, tailored for the specific characteristics of MATLAB code bases. We present an exploratory, token-based, approach to aspect mining for MATLAB. An analysis of data obtained from a tool using this approach over 209 publicly available MATLAB files indicate the approach is valid for detecting several kinds of crosscutting concerns in MATLAB systems.

Miguel P. Monteiro | João M. P. Cardoso | Nova de Lisboa | Simona Posea

[1] Arie van Deursen,et al. An evaluation of clone detection techniques for crosscutting concerns , 2004, 20th IEEE International Conference on Software Maintenance, 2004. Proceedings..

[2] William G. Griswold,et al. An Overview of AspectJ , 2001, ECOOP.

[3] Laurie J. Hendren,et al. AspectMatlab: an aspect-oriented scientific programming language , 2010, AOSD.

[4] João M. Fernandes,et al. Towards a Catalogue of Refactorings and Code Smells for AspectJ , 2006, LNCS Trans. Aspect Oriented Softw. Dev..

[5] Miguel P. Monteiro,et al. Adding aspect-oriented features to MATLAB , 2006 .

[6] Arie van Deursen,et al. Identifying aspects using fan-in analysis , 2004, 11th Working Conference on Reverse Engineering.

[7] Jens Krinke,et al. Aspect mining using event traces , 2004 .

[8] Paolo Tonella,et al. A Survey of Automated Code-Level Aspect Mining Techniques , 2007, LNCS Trans. Aspect Oriented Softw. Dev..

[9] Emily Hill,et al. Using natural language program analysis to locate and understand action-oriented concerns , 2007, AOSD.

[10] João Saraiva,et al. A domain-specific aspect language for transforming MATLAB programs , 2010 .