On 10 Dec 2009, at 00:53, Rodrigo Rosenfeld Rosas wrote:

Matt Wynne escreveu:

On 8 Dec 2009, at 22:04, Rodrigo Rosenfeld Rosas wrote:

I would like to be able to write a custom matcher so that I could call this test, for instance:

@user.should be_allowed_to_visit(url)
@non_welcome.should_not be_allowed_to_visit(url)

The matcher would call Webrat methods such as 'visit'.

The problem is that it would try to call 'visit' from User class instead of the binding in which @user access is verified...

I have manually set the binding to the Matcher in a before(:all), to avoid calling 'be_allowed_to_visit(url, binding)', but I would like to know if there would be a better way to "get" this binding...

Any thoughts on that?

How have you defined the matcher? If you use the old-school technique where you create a matcher class and a helper method to construct the class (instead of using the shiny new matcher DSL) then the method should execute in the context that can see the binding, so you can pick it up and pass it into the matcher class at that point.

Make sense?
Thank you, Matt, I have finally got it working. The trick is passing self to the initialize of the matcher class like:

def be_allowed_to_visit(url)
BeAllowedToVisit.new(url, self) # def initialize(url, scope); scope.visit url...;end
end

One problem less :) Next one to be solved: making Selenium work faster with Webrat :)

You might want to look at replacing webrat with Capybara then. I've not tried it myself, but I understand Selenium2 (which Capybara supports) is much faster.


Thanks,

Rodrigo.

__________________________________________________
Faça ligações para outros computadores com o novo Yahoo! Messenger 
http://br.beta.messenger.yahoo.com/
_______________________________________________
rspec-users mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/rspec-users

cheers,
Matt

+447974 430184
[email protected]
http://mattwynne.net

_______________________________________________
rspec-users mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/rspec-users

Reply via email to