JNose: Java Test Smell Detector

• Tássio Virgínio UFBA
• Luana Martins UFBA
• Larissa Rocha UFBA
• Railana Santana UFBA
• Adriana Cruz UFLA
• Heitor Costa UFLA
• Ivan Machado UFBA

Resumo

Several strategies have been proposed for test quality measurement and analysis. Code coverage is likely the most widely used one. It enables to verify the ability of a test case to cover as many source code branches as possible. Although code coverage has been widely used, novel strategies have been recently employed. It is the case of test smells analysis, which has been introduced as an affordable strategy to evaluate the quality of test code. Test smells are poor design choices in implementation, and their occurrence in test code might reduce the quality of test suites. Test smells identification is clearly dependent on tool support, otherwise it could become a cost-ineffective strategy. However, as far as we know, there is no tool that combines code coverage and test smells to address test quality measurement. In this work, we present the JNose Test, a tool aimed to analyze test suite quality in the perspective of test smells. JNose Test detects code coverage and software evolution metrics and a set of test smells throughout software versions.

Palavras-chave: Test Smells, Quality of Tests, Code Coverage, Test Suite Evolution