FineRTS

Select regression tests more precisely by reasoning about semantics-modifying changes.

Based on Ekstazi, we implemented FineEkstazi

Based on STARTS, we implemented FineSTARTS

Content of this repository

insights-of-class-level

Contains the manual inspection results for the 5 projects. The file, "insightclass.json" puts them together in one place.

change-levels

Contains the automatically generated classification of change levels in the 23 evaluation projects.

cmp-tools

Contains the execution results (including intermediate data) of Ekstazi, FineEkstazi^F, FineEkstazi, STARTS, FineSTARTS^F, FineSTARTS, and HyRTS.

shas-with-java-files-change

Contains the SHAs used in evaluation on 23 projects.

predictiverts-results

Contains the ML models prediction results.

appendix.pdf

Contains the appendix of the paper.

python

Contains the python scripts used in the evaluation.

Citation

If you have used FineRTS in a research project, please cite the research paper in any related publication:

Title: More Precise Regression Test Selection via Reasoning about Semantics-Modifying Changes

Authors: Yu Liu, Jiyang Zhang, Pengyu Nie, Milos Gligoric, Owolabi Legunsen

@inproceedings{liu2023more,
  title={More precise regression test selection via reasoning about semantics-modifying changes},
  author={Liu, Yu and Zhang, Jiyang and Nie, Pengyu and Gligoric, Milos and Legunsen, Owolabi},
  booktitle={Proceedings of the 32nd ACM SIGSOFT International Symposium on Software Testing and Analysis},
  pages={664--676},
  year={2023}
}