A Brief Survey on Oracle-based Test Adequacy Metrics

Even though code coverage is a widespread and popular test adequacy metric, it has several limitations. One of the major limitations is that code coverage does not satisfy the necessary conditions for effective fault detection, as it only cares about executing different parts of a program. Studies showed that code coverage as a test adequacy metric is a poor indicator of the quality of a test suite as it does not consider the quality of test oracles. To address this limitation, researchers proposed extensions to traditional code coverage metrics that explicitly consider test oracle quality. We name these metrics as \textit{oracle-based code coverage}. This survey paper has discussed all oracle-based techniques published so far, starting from 2007.