Web Applications today are capable of fairly complex interactions using rich UI interfaces and lots of asynchronous AJAX requests. Many of these applications are built over other frameworks and libraries like ext-js, gwt, dojo etc. Traditionally, open-source tools like Selenium have taken a developer centric approach to web automation. They handle element identification using id, XPath or DOM. Applications may need to be modified to make some applications more “testable”. In this talk, we introduce you to Sahi, another open source web testing tool.
Sahi goes beyond XPaths and ids, and uses various other attrtibutes and fuzzy APIs like “near”, “in” and “under” to identify elements. Sahi also implicitly waits for AJAX requests and page loads, freeing the tester from knowing the internals of an application. As a demo, we will automate a fairly complex application and walk you through the typical problems encountered during web testing and how Sahi solves them.
1) Limitations of XPath
2) Handling dynamic ids
3) Handling run time delays
4) Pros and Cons of a proxy based architecture as used by Sahi
5) Handling multiple domains, popups, SSL, file uploads and other tricky issues
6) “Testability” is NOT adding ids to everything
7) A brief comparison of approaches of Selenium, Selenium 2.0 and Sahi
Working knowledge of Sahi and information to evaluate technologies better.
http://www.slideshare.net/narayanraman/gtac-2010
http://www.youtube.com/watch?v=EYJjXO4ax94
| © Tyto Software Pvt. Ltd. |
