006. June 4 to June 8

Planned tasks for this week

• task 1: mine more repos -- java and python ✔️

Progress

task 1

• it just occurred to me that I missed an obvious case: searching for a keyword ex. "bug" will miss commits with different capitalization ex "Bug". Therefore, changed mining script to convert all commit msgs to lowercase first.

• tried out another method of picking repos to mine: went to github and searched for commits containing "memory leak", "performance", etc., to find candidate repos, then mined those with PyDriller

  • pros: we know the repo has at least one true positive, and potentially more when examined fully with PyDriller
  • cons: github doesn't let you filter commits by language, so have to look for .java or .py files manually

Other

• separated python data from java data on github
• also cleaned up the repo a little by removing old irrelevant data files
• added a list of all repos that have been examined on Github 📋
• added a working list of commits that are interesting but not related to this project here
• current tally of TOTAL problems documented = 78 (as of June 8)

Open Problems

• interesting situation that keeps coming up is that some repos have lots of stars but very few commits (<30); these repos don't give many hits

Things we discussed/agreed on

• continue looking for data for this week; then try methods suggested in paper
• document patterns and recurring themes

Next steps

• use GumTree API to look for patterns in code revisions