AppIntent: Intuitive Automation Specification Framework for Mobile AppTesting
This addresses the need for faster and more efficient testing for mobile app developers, though it appears incremental as it builds on existing automation tools by adding a higher-level abstraction layer.
The authors tackled the problem of mobile app testing automation requiring low-level scripting by proposing AppIntent, a high-level intent-based specification language and framework that allows users to capture test automation steps intuitively, resulting in a tool that abstracts away scripting complexities.
The proliferation of mobile apps and reduced time in mobile app releases mandates the need for faster and efficient testing of mobile apps, their GUI and functional capabilities. Though, there are wide variety of open source tools and frameworks that are developed to provide automated test infrastructure for testing mobile apps. Each of these automation tools supports different scripting languages for automating the app testing. These frameworks fundamentally lacks the ability to directly capture the intent of the users who intend to effectively test the mobile app and its cross-app functional capabilities and performance without worrying about the low-level scripting language associated with each tool. Hence, to address this limitation, we propose a high-level intent-based automation specification language and APIs that could effectively address following aspects: (i)capture the test automation steps to be captured as high-level intents using intuitive automation specification language, and (ii) provides framework support to effectively capture the users behavior patterns effectively for testing the apps. We develop, AppIntent a high level automation specification language-based framework that directly captures the test automation intents with in and across multiple apps using high-level and intuitive language without worrying about how to actually develop scripts for automation.