The adoption of refactoring techniques for continuous integration received much less attention from the research community comparing to root-canal refactoring to fix the quality issues in the whole system. Several recent empirical studies show that developers, in practice, are applying refactoring incrementally when they are fixing bugs or adding new features. There is an urgent need for refactoring tools that can support continuous integration and some recent development processes such DevOps that are based on rapid releases.

In this paper, we propose for the first time an intelligent software refactoring bot, called RefBot. Integrated into the version control system (e.g. GitHub), our bot continuously monitors the software repository and it is triggered by any open or merge actions on pull-requests. The bot analyzes the files changed during that pull-request to identify refactoring opportunities using a set of quality attributes then it will find the best sequence of refactorings to fix the quality issues, if any. The bot recommends all these refactorings through an automatically generated pull-request. The developer is able to review the recommendations and their impacts in a detailed report and select the code changes that he wants to keep or ignore. After this review, the developer can close and approve the merge of the bot’s pull request. We quantitatively and qualitatively evaluated the performance and effectiveness of RefBot by a survey conducted with experienced developers who used the bot.