http://blog.ardekantur.com/2008/06/tddreverse/ A blog about computer science, Ruby, .NET, and other fun things on the edge of software engineering. Thu, 20 Nov 2008 10:30:51 +0000 http://wordpress.org/?v=2.7-beta1 hourly 1 http://blog.ardekantur.com/2008/06/tddreverse/comment-page-1/#comment-827 JoeNotCharles Wed, 02 Jul 2008 03:19:41 +0000 http://blog.ardekantur.com/archives/95#comment-827 <p>It would be better to get the XML responses through a different library, so that bugs in your code don't affect them. What if your XML parser adds some garbage data, and then your tests pass because your client works around it (because when you write the it you assume the XML looked like that coming from NewsGator). Then you fix that bug but don't recapture all the XML you use in your client - now it passes all the tests, but fails on the actual NewsGator site because the XML parser's no longer mangling things.</p> It would be better to get the XML responses through a different library, so that bugs in your code don’t affect them. What if your XML parser adds some garbage data, and then your tests pass because your client works around it (because when you write the it you assume the XML looked like that coming from NewsGator). Then you fix that bug but don’t recapture all the XML you use in your client - now it passes all the tests, but fails on the actual NewsGator site because the XML parser’s no longer mangling things.

]]>