Uma análise empírica da ocorrência dos testes quebradiços em aplicativos híbridos

Abstract

Considering Software Engineering (ES) activities, Software Testing is considered one of the most important activities because of it one can find inconsistencies, allowing fix problems before delivery. A feasible way to perform software testing is through automation, where testing data can be executed several times at various occasions. The automation process has been growing every day in many types of projects. In this context of automation, a problem that has emerged is known as Flaky Test, which can be broadly defined as software testing whose outcome is uncertain, which means in some executions it is given as pass with success, but in other executions it is rejected and interrupted by fail. For this, in this study, we aimed to present the occurrence of Flaky Tests in hybrid apps, by performing a case study. The case study we performed analyzed automated tests in different scenarios of various hybrid apps projects. The results empirically obtained through the case study, showed the occurrence of Flaky tests in hybrid apps, also it helped to identify the causes for Flaky tests and, finally, which type of software testing technique is more error-prone to this flaky circumstances. The contributions obtained showed that the functional software testing technique is more error-prone to Flaky tests, also made it possible to identify that the cause of asynchronous waiting occurs more easily in hybrid apps projects. The work we conducted has promoted contributions to academia and industry. For the academy, information gathered and synthesized helps on future studies and promotes new studies on Flaky tests. For the industry, the information obtained assists in software projects that may suffer from Flaky testing, providing data on how flaky testing occurs for possible predictions.