Comments on: Avoiding Brittle Element Selection with Selenium2 http://www.rapaul.com/2010/07/18/avoiding-brittle-element-selection-with-selenium2/ A technical blog written by Richard Paul Tue, 07 Feb 2012 02:12:57 +0000 http://wordpress.org/?v=2.9 hourly 1 By: A Smattering of Selenium #21 « Official Selenium Blog http://www.rapaul.com/2010/07/18/avoiding-brittle-element-selection-with-selenium2/comment-page-1/#comment-1182 A Smattering of Selenium #21 « Official Selenium Blog Mon, 26 Jul 2010 11:33:56 +0000 http://www.rapaul.com/?p=608#comment-1182 [...] Avoiding Brittle Element Selection with Selenium2 tackles a constant topic using Se2 — but it can be applied to Se1 as well Leave a Comment [...] [...] Avoiding Brittle Element Selection with Selenium2 tackles a constant topic using Se2 — but it can be applied to Se1 as well Leave a Comment [...]

]]> By: Richard Paul http://www.rapaul.com/2010/07/18/avoiding-brittle-element-selection-with-selenium2/comment-page-1/#comment-1162 Richard Paul Wed, 21 Jul 2010 17:49:07 +0000 http://www.rapaul.com/?p=608#comment-1162 With many browsers supporting querySelector and querySelectorAll http://www.w3.org/TR/selectors-api/#examples I think the legibility/maintainability of Selenium scenarios trumps any micro optimisations an XPath may provide (assuming XPath can actually be faster). My anecdotal evidence suggests our Selenium tests aren't slow at selecting elements, its normally navigating between pages or restarting the browser instance between scenarios. How are you getting on with rolling out your acceptance tests? With many browsers supporting querySelector and querySelectorAll http://www.w3.org/TR/selectors-api/#examples I think the legibility/maintainability of Selenium scenarios trumps any micro optimisations an XPath may provide (assuming XPath can actually be faster).

My anecdotal evidence suggests our Selenium tests aren’t slow at selecting elements, its normally navigating between pages or restarting the browser instance between scenarios.

How are you getting on with rolling out your acceptance tests?

]]> By: Martin Anderson http://www.rapaul.com/2010/07/18/avoiding-brittle-element-selection-with-selenium2/comment-page-1/#comment-1161 Martin Anderson Wed, 21 Jul 2010 16:46:06 +0000 http://www.rapaul.com/?p=608#comment-1161 Great to see how you are dealing with the fragility of xpath. I know it was one of the things I was most interested at the Java Unconference :-)I know that there can be a x20 speed increase changing from a bad xpath (//../div[@id='post-1') to a good one (id('post-1')) but I would be interested in seeing the speed difference between css selectors and ids. Both css selectors and ids are far better than full xpaths but I suppose which you use depends on what you are most likely to refactor; code layout or look and feel. Great to see how you are dealing with the fragility of xpath. I know it was one of the things I was most interested at the Java Unconference :-)I know that there can be a x20 speed increase changing from a bad xpath (//../div[@id=’post-1′) to a good one (id(‘post-1′)) but I would be interested in seeing the speed difference between css selectors and ids. Both css selectors and ids are far better than full xpaths but I suppose which you use depends on what you are most likely to refactor; code layout or look and feel.

]]>