TestRoutes: A Manually Curated Method Level Dataset for Test-to-Code Traceability

Abstract

High test-to-code traceability can be an important aspect of quality assurance and can contribute to bug localization and code maintenance. Several existing techniques and a considerable effort from the scientific community already made significant advances in the field. Despite this, readily accessible data on traceability links is very scarce. To contribute to related research, we present a manually curated test-to-code traceability dataset containing the traceability information on 220 test cases. This method-level data was gathered from 4 open-source software systems written in the Java language, distinguishing not only focal information on test cases but also highlighting the utilized helper methods on both the test and production aspects of code. The data includes more than 2000 of such method classifications.