如何从使用 Capybara 的 Cucumber Step 获取整个页面的响应 HTML？

Question

When running my specs i get errors where Capybara couldn't find elements based on their labels and such. Then i thought i wanted the HTML, or somehow debug the errors.

This is the /features/merge.feature

#Just a test!
Scenario: Cannot merge articles when not admin
    Given I am not an admin
    When I am on the edit article page
    Then I should not see the "merge articles" form

And this is: /feature/step_definitions/merge.rb - and it only contains "Given" statements!

Given /am not an admin$/ do
  visit '/accounts/login'
  puts "TEZT"
  puts page.native
  puts page.native.text
  fill_in 'user_login', :with => 'editor_rico'
  fill_in 'user_password', :with => 'a'
  click_button 'Login'
end

Given /^I am an admin$/ do
  visit '/accounts/login'
  fill_in 'user_login', :with => 'admin_rico'
  fill_in 'user_password', :with => 'a'
  click_button 'Login'
end

Then i get this error:

Scenario: Cannot merge articles when not admin    
# features/article_merging.feature:20
Given I am not an admin                 
#features/step_definitions/article_steps.rb:39
cannot fill in, no text field, text area or password field with id, name, or label
'user_login' found (Capybara::ElementNotFound)
(eval):2:in `fill_in'
./features/step_definitions/article_steps.rb:44:in `/am not an admin$/'
features/article_merging.feature:21:in `Given I am not an admin'

Havin a rough time working with Cucumber here ..

1) Why do i get the above

2) How do i debug it, i wanna see the HTML code!

Answer 1

This errors appears when your page HTML does not contain any input or textarea element with name, id or label 'user_login' . If you add the launchy gem to your Gemfile you will be able to see how your page looks:

In your Gemfile :

group :test do
  gem 'launchy'
  gem 'capybara'
  ...
end

Then, inside your feature/step_definitions/merge.rb file:

Given /^I am not an admin$/ do
  visit '/accounts/login'
  save_and_open_page
end

The save_and_open_page will open the current page in your browser, so then you can inspect the HTML and see what it looks like.

Hope it helps! :)

Answer 2

It think the issue with the above code is that 'page' doesn't become available until after you end the section.

Try putting your 'puts' into the 'When'or 'Then' step

You can query page as follows (assuming you're using RSpec)

page.should have_content 'foo'
page.should have_selector 'h1', text: => 'Hello dave'

Hope this helps

Answer 3

You can also see the HTML response by printing page.body to stdout.

Given /^I am not an admin$/ do
  visit '/accounts/login'
  puts page.body
end