摘要: Intuitively we know, some software errors are more complex than others. If the error can be fixed by changing one faulty statement, it is a simple error. The substantial fix must be, consider In this work, formally define and quantify complexity of an w.r.t. error's least complex, correct fix. As concrete measure for such fixes, introduce Cyclomatic Change Complexity which inspired existing program metrics. Moreover, CoREBench, collection 70 regression systematically extracted from several open-source C-projects compare their with that seeded in two most popular benchmarks, SIR Siemens Suite. We find significantly less i.e., require compared to actual errors. For example, among 42% 8% ones. This concern external validity studies based on propose CoREBench controlled study testing, debugging, repair techniques.