Inferring Hints for Defect Fixing Order from Requirements-to-test-case Mappings
Stanisław Jarząbek , Konrad Liszewski , Cezary Bołdak
AbstractIn what order should we debug defects reported in regression testing to minimize the overall effort of repairing a program? Other than prioritizing defects from business and project managerial perspectives, technical considerations such as dependencies among defects have also to do with answering this question. Errors ripple through code - on one hand, we may want to fix defects that have wide impact on other program failures revealed by testing. But some of such root causes of failures that are affected by other defects may not be ready for correction. Without systematic impact analysis among functional program units, programmers determine the order of fixing defects mostly manually, relying on intuition. We propose a semi-automated method with filtering, visualizations and heuristics-based computations to solve the problem. Instead of dependency graphs for impact analysis, the source of information for our method are easier to build and often used mappings between test cases and relevant software requirements. We derived and validated our method over 15 years of teaching a project course in which test cases are tagged with relevant requirements.
|Publication size in sheets||0.5|
|Book||Jarząbek Stanisław, Poniszewska-Marańda Aneta, Madeyski Lech (eds.): Integrating Research and Practice in Software Engineering, Studies in Computational Intelligence, vol. 851, 2020, Springer, ISBN 978-030-26573-1, [978-3-030-26574-8], 258 p., DOI:10.1007/978-3-030-26574-8|
|Keywords in English||Regression testing, Defects Debugging, Test library, Impact analysis|
|Internal identifier||ROC 19-20|
|Score||= 20.0, 13-01-2020, MonographChapterAuthor|
|Publication indicators||: 2018 = 0.447|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.