Can't find element to click on using Capybara

问题:

Background: I'm using Capybara with Rspec to test a Rails 3 app. Driver used: Selenium

Problem: I can't find the "Sign in" button in order to click on from within my test.

HTML code:

<form accept-charset="UTF-8" action="/" class="filter_form" id="login" method="post">
        <fieldset>
          <div class="modal-body">
            <div class="clearfix login-fields">
              <label for="user_email">Email</label>
              <div class="input login-inputs">
                <input class="input-text" id="user_email" name="user[email]" placeholder="email" size="30" type="email" value="">
              </div>
            </div>
            <div class="clearfix login-fields">
              <label for="user_password">Password</label>
              <div class="input login-inputs">
                <input class="input-text" id="user_password" name="user[password]" placeholder="password" size="30" type="password">
              </div>
            </div>
          </div>
          <div class="modal-footer">
            <input class="btn btn-primary login_btn" id="btn_login" name="commit" type="submit" value="Sign in">
            <a href="/lms/forgot_password" class="btn">Forgot password...</a>
            <a href="#" class="btn close cancel" data-dismiss="modal">Cancel</a>
          </div>
        </fieldset>
</form>

Failing test

it "should login correctly if the right credentials are given", :js => true do

    Capybara.default_wait_time = 5
    Capybara.reset_sessions!
    visit '/'
    click_link('login_link') #this will bring a modal window with the code posted above using js
    within("#login") do
      fill_in 'user_email', :with => "[e-mail]"
      fill_in 'user_password', :with => "mypwd"
    end

    response.should have_selector('input#btn_login') #This passes
    click_on("input#btn_login") #Here it fails, saying it can't find an element with that selector
    response.should have_selector(:xpath, '//div[@class="alert-message block-message info"]')
  end

My test file is inside spec/requests.

Any ideas? Thanks!

回答1:

Please try this:

page.find("#btn_login").click

回答2:

This: https://stackoverflow.com/a/11348065/1504796

Is the right answer.

click_on does not take a CSS selector, but instead the text or id of a link. You want click_on("btn_login"). No hash sign or anything.

回答3:

try

find('#id',:visible => true).click

回答4:

Try adding "gem 'launchy'" to your Gemfile and put "save_and_open_page" line before failed line in step file.

REFERENCE: http://techiferous.com/2010/04/using-capybara-in-rails-3/

回答5:

It looks like you're trying to click on a submit button inside some modal css. You'll need to invoke whatever displays that modal element first.

回答6:

I don't think click_on will take a locator like that--I think it may want just an id, name or value. As an experiment, try replacing click_on("input#btn_login") with:

page.find('#btn_login').click

回答7:

• click_button("Sign in")
• find(:xpath, "//*[@id='btn_login']").click (or .trigger('click'))

If else fails, go through the console and see if you can check the presence of the button: document.getElementById("btn_login")

回答8:

If you know the link text, you can use page.find_link(text).click. (source)

回答9:

Try adding a save_and_open_page before the line in question. This should allow you to see the page and any errors that may prevent the clicking action.

回答10:

I use capybara with chrome

Capybara.register_driver :chrome do |app|   Capybara::Selenium::Driver.new(app, 
                                                           :browser => :chrome) 
end

Capybara.javascript_driver = :chrome

and install chrome driver:

brew install chromedriver

http://collectiveidea.com/blog/archives/2011/09/27/use-chrome-with-cucumber-capybara/