Where Should I Look at? Recommending Lines that Reviewers Should Pay Attention To

Code review is an effective quality assurance practice, yet can be time-consuming since reviewers have to carefully review all new added lines in a patch. Our motivating analysis shows that at the median, patch author waited 15-64 hours to receive initial feedback from reviewers. Moreover, we found that large patches tend to receive initial feedback from reviewers slower than the small patches. Hence, it would be beneficial to reviewers to reduce their effort by pinpointing the lines that they should pay attention to.

In this paper, we proposed REVSPOT—a line-level recommendation approach to predict lines that will receive a comment and lines that will be revised. Through a case study of three open-source projects (i.e., Openstack Nova, Openstack Ironic, and Qt Base), the experimental results show that REVSPOT can accurately predict files that will receive comments with an AUC of 0.70-0.78 and files that will be revised with an AUC of 0.68-0.77. REVSPOT also can predict and prioritize lines that will receive comments and will be revised with a median recall of 0.5 and 0.45, a false alarm rate of 0.42 and 0.4 and top-10 accuracy of 0.81 and 0.93, which outperforms the baseline. These findings suggest that REVSPOT could help reviewers to reduce their reviewing effort by reviewing a smaller set of lines, increasing code review speed and reviewer productivity.