Provided below are supplements to the research on refactoring documentation.
What patterns do developers use to describe their refactoring activities?
To identify self-affirmed refactoring patterns, we perform manual analysis similar to our previous work [AlOmar et al. 2019a]. Since commit messages are written in natural language and we need to understand how developers express refactoring, we manually analyzed commit messages by reading through each message to identify self-affirmed refactorings.
Click to continue reading...Do Commits containing the label Refactor indicate more refactoring activity than those without the label?
[Murphy-Hill et al. 2008] proposed several hypotheses related to four methods that gather refactoring data and outlined
experiments for testing those hypotheses. One of these methods concerns mining the commit log. [Murphy-Hill et al.
2008] hypothesize that commits labeled with the keyword “refactor” do not indicate more refactoring instances than
unlabeled commits. In an empirical context, we test this hypothesis in two steps. In the first steps, we used the keyword
“refactor”, exactly as dictated by the authors.
Click to continue reading...