SHF: Small: LeakDroid: Exposing Leaks and Jank in Android Applications

Android软件广泛用于智能手机、平板电脑和其他各种设备。在资源有限的设备中，消耗太多资源的应用程序可能会导致速度减慢、崩溃和负面的用户体验。与资源泄漏相关的软件缺陷非常普遍且多种多样，但没有针对此类泄漏的全面测试和调试策略。Android软件的另一个常见问题是应用程序对用户操作的响应时间太长，在这种情况下，它被认为速度慢且质量低。目前还不存在用于暴露导致响应不良的底层软件缺陷的测试技术或工具。资源泄漏和响应不良会对软件可靠性、性能和市场成功产生严重影响。该项目开发了LeakDroid，这是一种用于测试和调试此类缺陷的新方法和工具集。静态代码分析用于提取Android应用程序的精确图形用户界面模型；这种分析对于许多用于程序理解、优化、测试和调试的算法是必不可少的。接下来，根据常见的泄漏模式，通过不应导致资源使用增加的一系列图形用户界面事件来生成图形用户界面测试。在选定的代码位置引入延迟的测试暴露出响应能力差。然后使用调试技术来发现泄漏的数据结构和响应能力差的操作，以及相应的缺陷代码。这些贡献提升了Android软件分析、测试和调试这一重要领域的技术水平。当这些进步和工具成为软件开发实践的一部分时，它们将为数百万移动设备用户带来更低的成本和更高质量的应用程序。该项目的教育努力培养了下一代移动软件创建者的技能

Android software is widely used in smartphones, tablets, and variousother devices. In devices with limited resources, an application thatconsumes too many resources can lead to slowdowns, crashes, andnegative user experience. Software defects related to leaking ofresources are widespread and varied, but there does not exist acomprehensive strategy for testing and debugging of such leaks.Another common problem in Android software is when an applicationtakes too long to respond to a user action, in which case it isperceived as sluggish and of low quality. There do not exist testingtechniques or tools for exposing the underlying software defectsleading to poor responsiveness.Resource leaks and poor responsiveness can have severe effects onsoftware reliability, performance, and marketplace success. Thisproject develops LeakDroid, a novel approach and toolset for testingand debugging of such defects. Static code analyses are used toextract precise GUI models of Android applications; such analyses areessential for many algorithms for program understanding, optimization,testing, and debugging. Next, GUI tests are generated based on commonleak patterns, through sequences of GUI events that should not lead toincreases in resource usage. Poor responsiveness is exposed with teststhat introduce delays at chosen code locations. Debugging techniquesare then employed to find leaking data structures and poorresponsiveness operations, as well as the corresponding defectivecode. These contributions advance the state of the art in theimportant area of analysis, testing, and debugging for Androidsoftware. When these advances and tools become part of softwaredevelopment practices, they would result in lower costs and increasedquality applications for millions of users of mobile devices. Theeducational efforts of the project develop the skills of the nextgeneration of creators of mobile software