In this paper we present a case study of the effect of refac- toring activities in the structure of two large software system, from the complex networks perspective. We analyze two large open source Java software systems, using data previously extracted from Software Con- figuration Management and Bug Tracking systems, and a statistical ap- proach to figure out how many of them could be classified as refactoring activities. We find that software elementary units interested by refac- toring operations are inclined to be connected with each other, forming clusters whose number is systematically below a given threshold. Moreover, this behaviour does not depend on the number of classes involved by the same refactoring.
Two Case Studies on Clusterization of Refactored Classes
MARCHESI, MICHELE;MONNI, CRISTINA;ORRU', MATTEO;TONELLI, ROBERTO
2013-01-01
Abstract
In this paper we present a case study of the effect of refac- toring activities in the structure of two large software system, from the complex networks perspective. We analyze two large open source Java software systems, using data previously extracted from Software Con- figuration Management and Bug Tracking systems, and a statistical ap- proach to figure out how many of them could be classified as refactoring activities. We find that software elementary units interested by refac- toring operations are inclined to be connected with each other, forming clusters whose number is systematically below a given threshold. Moreover, this behaviour does not depend on the number of classes involved by the same refactoring.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.