Skip to content

Latest commit

 

History

History
112 lines (79 loc) · 3.16 KB

README.rdoc

File metadata and controls

112 lines (79 loc) · 3.16 KB
Raw

Capybara-PageObject ↑

capybara-pageobject helps you easily define page objects for capybara based tests. Page objects allow you to abstract your tests from low-level UI bindings. This allows you to easily adapt your tests to UI changes.

Setup ↑

Install as a gem:

sudo gem install capybara-pageobject

Usage ↑

Add the following section to your env.rb or acceptance_helper:

Capybara::PageObject.configure do |config|
  config.page_file = 'support/pages/pages.yml'
end

The gem makes a method ‘website’ available to your tests.

Page definitions ↑

Page definitions are specified in a YAML file. for eg:

home_page:
  id: "#header_banner"
  url: "/"
login_form:
  id: "#login-form"
  url: "/sign_in"
  attributes:
    email_address: "#email"
    password: "#password.red"
    error_message: "#alert_message"
  actions:
    login_button: "#login_submit"

page ↑

Each first-level definition specifies a page. The above YAML declares that you have two pages home_page and login_form.

url ↑

The url attribute specifies the url of the page. You can navigate to these pages as below:

website.home_page.visit
website.login_form.visit

Dynamic urls ↑

For RESTful dynamci urls, use the following definition:

comment_page:
  url: "/users/:user_id/comments/:comment_id"

You can then visit the comments page as below:

website.comments_page.visit user_id: 3, comment_id: 1

id ↑

The id attribute specifies the CSS3 selector which uniquely identifies that the page is loaded. You can use this as below:

website.login_form.visit
website.login_form.should be_visible

attributes ↑

Each attribute on the page is specified alongwith the CSS3 selector to uniquely identify it. Each attribute definition can be used as below:

website.login_form.email_address "foo bar"              #Set email address to 'foo bar'
website.login_form.some_checkbox true                   #Check checkbox
website.login_form.email_address.should == "foo bar"    #Assert email address is equal to 'foo bar'
website.login_form.email_address.should be_visible      #Assert email address is visible

Attributes also work with text content. for eg:

website.login_form.error_message.should == "Some error occurred"

actions ↑

Any clickable (button or link) element on the page is defined as an action. Each action definition can be used as below:

website.login_form.login_button.click
website.login_form.login_button.should be_disabled
website.login_form.login_button.should be_visible

Page block mode ↑

To make your tests more readable, you can specify page interactions within a block. for eg:

website.login_form do
   email_address '[email protected]'
   password 'testing'
   login_button.click
   error_message.should be_visible
end

Custom Page Class ↑

If you need to define custom actions on the page, add a class which extends Capybara::PageObject::Page You can them specify it in the YAML file as below:

login_form:
  id: "#login-form"
  url: "/sign_in"
  class: CustomLoginForm

TODO ↑

  • XPath selectors

  • Standard setters ie website.login_form.email_address = “foo bar”