Thanks Mike. fyi, I am trying out Cucumber and Culerity for integration testing. Culerity provides for AJAX testing.
Having written my first Cucumber (with Culerity) feature and set of scenarios, I was keen to then use RSpec to begin driving out the views, controllers etc so that I can eventually run the e2e Cucumber scenarios. On May 12, 3:04 pm, Mike Doel <m...@mikedoel.com> wrote: > On May 12, 2009, at 8:36 AM, Lee Longmore wrote: > > > I am new to RSpec and have just started to drive out my first Rails > > view using a spec. I have used the The RSpec Book (beta) to do the > > basic stuff like testing for the presence of a field but I am unsure > > where to start for driving out some AJAX functionality. > > Check out Adam McCrea's fork of webrat: > > http://github.com/adamlogic/webrat/tree/master > > It includes a submit_form_via_ajax method that I use to do this. It > basically just forces Rails into treating the form request as asking > for a .js response. There are limitations on what you can do with the > response (e.g. inspecting the entire page is no longer possible), but > you are able to look at the response body of the call and match it > against regular expressions. So, I use features like: > > When I go to the homepage > And I use a search term of "foo" > And I press "search" > And the edit_search_form is submitted via ajax > Then the search results should have 2 bars > > with the critical step like: > > When /^the (.+) is submitted via ajax$/ do |form| > submit_form_via_ajax form > end > > I've left out a couple of things here (in particular, a step that > fills in form fields to mimic javascript action), but this basic idea > makes it possible to have acceptance tests for ajax based features > that are written in the language of the customer and yet operate > quickly (i.e. you don't need to have selenium or something else like > that involved to run this). > > The drawback of course is that it's not really exercising the full end- > to-end system. While I understand that this is frowned upon (and for > good reason), in my case, I have decided that the speed of the tests > (and thus my willingness to use them) makes up for the drawbacks > involved. To limit my exposure, I'm working on including javascript > unit tests that will ensure that the behavior I stub out above > actually works as expected. > > Mike Doel > > _______________________________________________ > rspec-users mailing list > rspec-us...@rubyforge.orghttp://rubyforge.org/mailman/listinfo/rspec-users _______________________________________________ rspec-users mailing list rspec-users@rubyforge.org http://rubyforge.org/mailman/listinfo/rspec-users
