Automated cross-browser compatibility testing

Abstract

With the advent of Web 2.0 applications and new browsers, the cross-browser compatibility issue is becoming increasingly important. Although the problem is widely recognized among web developers, no systematic approach to tackle it exists today. None of the current tools, which provide screenshots or emulation environments, specifies any notion of cross-browser compatibility, much less check it automatically. In this paper, we pose the problem of cross-browser compatibility testing of modern web applications as a 'functional consistency' check of web application behavior across different web browsers and present an automated solution for it. Our approach consists of (1) automatically analyzing the given web application under different browser environments and capturing the behavior as a finite-state machine; (2) formally comparing the generated models for equivalence on a pairwise-basis and exposing any observed discrepancies. We validate our approach on several open-source and industrial case studies to demonstrate its effectiveness and real-world relevance.

Invariant-Based Automatic Testing of Modern Web Applications

“…16 [14] detects JavaScript and DOM invariants dynamically ProxyServer, OnNewState, PostCrawling CrossBrowser Tester [26] checks each state in three different browsers looking for cross-browser incompatibilities PostCrawling RegressionTester [33] conducts regression tests PostCrawlng RQ2 What is the fault revealing capability and effectiveness of our testing approach? RQ3 What is the performance of the proposed approach, and how well does it scale?…”

“…http://tudu.sourceforge.net server-side is based on J2EE and consists of around 12K lines of Java/JSP code, of which around 3K forms the presentation layer we are interested in. The client-side extends on a number of AJAX libraries such as DWR 25 and Scriptaculous 26 , and consists of around 11k LOC of external JAVASCRIPT libraries and 580 internal LOC.…”

“…The approach is also being adopted for web modelchecking using the inferred state machine. Recently, we have used the approach to automate cross-browser compatibility testing of modern web applications [26].…”

Recognition of Cross Browser Inconsistency by crawling and comparing extracted attributes

“…Patidar.et.al The recent work on recognizing XBIs has recommended techniques that focus only on some aspects of a web application's execution, and are well suitable for specific types of XBIs. For instance, a tool that uses computer vision to detect XBIs is WebDiff [6], whereas a creature that uses graph isomorphism along with text comparison to find XBIs is Cross T [7].These tools provide just halfway and loose answers for those XBI identification issue. An additional approach tests the trouble for cross-browser similarity of web applications concerning illustration of a practical consistency check from web application conducts a technique over different web browsers for a robotized result.…”

Cross-Browser Testing in Browserbite

“…This procedure is time-consuming, monotonous and non-scalable given the growing number of configurations that need to be supported by Web applications. Existing automated methods for cross-browser testing are generally based on an analysis of the Document Object Model (DOM) [1][2] [3]. However, the fact that a Web page has very similar DOM structure and parameters across different configurations does not guarantee absence of incompatibilities, as rendering engines may display similar DOMs in rather different ways.…”