X-Check: Improving Effectiveness and Efficiency of Cross-Browser Issues Detection for JavaScript-Based Web Applications
| Autor: | Hua Zhong, Wei Chen, Jun Wei, Guoquan Wu, Meimei He |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Information Systems and Management
Computer Networks and Communications Event (computing) business.industry Computer science JavaScript Computer Science Applications Cross-browser Hardware and Architecture Asynchronous communication Application domain Embedded system Server Web application Timer business computer computer.programming_language |
| Zdroj: | IEEE Transactions on Services Computing. 14:1123-1137 |
| ISSN: | 2372-0204 |
| DOI: | 10.1109/tsc.2018.2860983 |
| Popis: | Web 2.0 application based on JavaScript is a wide-spread application domain today as it delivers rich, interactive user experiences. However, with the increasing number of browsers and platforms on which the applications can be executed, cross-browser incompatibilities (XBIs) are becoming a serious problem for organizations to develop modern JavaScript-based Web applications. Although lots of XBIs detection techniques have been proposed, there are still some limitations: 1) existing techniques are prone to generating certain false positives/negatives that result from the fact that they ignore non-deterministic events (e.g., timer, asynchronous request/response) inside the browser; 2) detection process is inefficient, as the same elements located in different pages will be repeatedly checked even if they stay unchanged after an event is triggered. Leveraging existing record/replay technique, we proposed X-Check, a novel cross-browser testing technique, which supports automated XBIs detection effectively. To improve the efficiency of XBIs detection, this paper further designed an incremental detection algorithm by only checking DOM-mutated and layout-changed nodes. Our empirical evaluation shows that X-Check is effective and efficient. For the selected 21 real-world Web applications, it identifies XBIs with a fairly high precision (83 percent) and recall (93 percent), and improves the performance of XBIs detection about 5.79 times compared to its non-optimized version. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|