Back to Repositories

Testing XPath Selector Implementation in Capybara

This test suite validates Capybara’s XPath selector functionality, focusing on the has_xpath? and has_no_xpath? methods. It thoroughly tests element detection, attribute matching, and content verification across different scenarios.

Test Coverage Overview

The test suite provides comprehensive coverage of XPath selector functionality in Capybara.

Key areas tested include:

Basic element presence verification
Attribute-based selector matching (id, class)
Text content matching
Element count validation
Scoped element searching
Asynchronous content loading

Implementation Analysis

The implementation follows RSpec’s behavior-driven development pattern, utilizing Capybara’s session object for DOM interaction. Tests demonstrate both positive and negative assertions using have_xpath and have_no_xpath matchers, with particular attention to selector specificity and wait time handling.

Technical Details

Testing tools and configuration:

RSpec as the testing framework
Capybara for DOM interaction
XPath selectors for element location
JavaScript capability testing
Custom wait time configurations
Session-based testing approach

Best Practices Demonstrated

The test suite exemplifies several testing best practices:

Comprehensive edge case coverage
Clear test organization and grouping
Proper setup and teardown
Explicit wait time handling
Robust selector testing
Scoped testing contexts

teamcapybara/capybara

lib/capybara/spec/session/has_xpath_spec.rb

            
# frozen_string_literal: true

Capybara::SpecHelper.spec '#has_xpath?' do
  before do
    @session.visit('/with_html')
  end

  it 'should be true if the given selector is on the page' do
    expect(@session).to have_xpath('//p')
    expect(@session).to have_xpath("//p//a[@id='foo']")
    expect(@session).to have_xpath("//p[contains(.,'est')]")
  end

  it 'should support :id option' do
    expect(@session).to have_xpath('//h2', id: 'h2one')
    expect(@session).to have_xpath('//h2')
    expect(@session).to have_xpath('//h2', id: /h2o/)
  end

  it 'should support :class option' do
    expect(@session).to have_xpath('//li', class: 'guitarist')
    expect(@session).to have_xpath('//li', class: /guitar/)
    expect(@session).to have_xpath('//li', class: /guitar|drummer/)
    expect(@session).to have_xpath('//li', class: %w[beatle guitarist])
    expect(@session).to have_xpath('//li', class: /.*/)
  end

  it 'should be false if the given selector is not on the page' do
    expect(@session).not_to have_xpath('//abbr')
    expect(@session).not_to have_xpath("//p//a[@id='doesnotexist']")
    expect(@session).not_to have_xpath("//p[contains(.,'thisstringisnotonpage')]")
  end

  it 'should use xpath even if default selector is CSS' do
    Capybara.default_selector = :css
    expect(@session).not_to have_xpath("//p//a[@id='doesnotexist']")
  end

  it 'should respect scopes' do
    @session.within "//p[@id='first']" do
      expect(@session).to have_xpath(".//a[@id='foo']")
      expect(@session).not_to have_xpath(".//a[@id='red']")
    end
  end

  it 'should wait for content to appear', requires: [:js] do
    Capybara.using_wait_time(3) do
      @session.visit('/with_js')
      @session.click_link('Click me') # updates page after 500ms
      expect(@session).to have_xpath("//input[@type='submit' and @value='New Here']")
    end
  end

  context 'with count' do
    it 'should be true if the content occurs the given number of times' do
      expect(@session).to have_xpath('//p', count: 3)
      expect(@session).to have_xpath("//p//a[@id='foo']", count: 1)
      expect(@session).to have_xpath("//p[contains(.,'est')]", count: 1)
      expect(@session).to have_xpath("//p//a[@id='doesnotexist']", count: 0)
      expect(@session).to have_xpath('//li', class: /guitar|drummer/, count: 4)
      expect(@session).to have_xpath('//li', id: /john|paul/, class: /guitar|drummer/, count: 2)
      expect(@session).to have_xpath('//li', class: %w[beatle guitarist], count: 2)
    end

    it 'should be false if the content occurs a different number of times than the given' do
      expect(@session).not_to have_xpath('//p', count: 6)
      expect(@session).not_to have_xpath("//p//a[@id='foo']", count: 2)
      expect(@session).not_to have_xpath("//p[contains(.,'est')]", count: 5)
      expect(@session).not_to have_xpath("//p//a[@id='doesnotexist']", count: 1)
    end
  end

  context 'with text' do
    it 'should discard all matches where the given string is not contained' do
      expect(@session).to have_xpath('//p//a', text: 'Redirect', count: 1)
      expect(@session).not_to have_xpath('//p', text: 'Doesnotexist')
    end

    it 'should discard all matches where the given regexp is not matched' do
      expect(@session).to have_xpath('//p//a', text: /re[dab]i/i, count: 1)
      expect(@session).not_to have_xpath('//p//a', text: /Red$/)
    end
  end
end

Capybara::SpecHelper.spec '#has_no_xpath?' do
  before do
    @session.visit('/with_html')
  end

  it 'should be false if the given selector is on the page' do
    expect(@session).not_to have_no_xpath('//p')
    expect(@session).not_to have_no_xpath("//p//a[@id='foo']")
    expect(@session).not_to have_no_xpath("//p[contains(.,'est')]")
  end

  it 'should be true if the given selector is not on the page' do
    expect(@session).to have_no_xpath('//abbr')
    expect(@session).to have_no_xpath("//p//a[@id='doesnotexist']")
    expect(@session).to have_no_xpath("//p[contains(.,'thisstringisnotonpage')]")
  end

  it 'should use xpath even if default selector is CSS' do
    Capybara.default_selector = :css
    expect(@session).to have_no_xpath("//p//a[@id='doesnotexist']")
  end

  it 'should respect scopes' do
    @session.within "//p[@id='first']" do
      expect(@session).not_to have_no_xpath(".//a[@id='foo']")
      expect(@session).to have_no_xpath(".//a[@id='red']")
    end
  end

  it 'should wait for content to disappear', requires: [:js] do
    Capybara.default_max_wait_time = 2
    @session.visit('/with_js')
    @session.click_link('Click me')
    expect(@session).to have_no_xpath("//p[@id='change']")
  end

  context 'with count' do
    it 'should be false if the content occurs the given number of times' do
      expect(@session).not_to have_no_xpath('//p', count: 3)
      expect(@session).not_to have_no_xpath("//p//a[@id='foo']", count: 1)
      expect(@session).not_to have_no_xpath("//p[contains(.,'est')]", count: 1)
      expect(@session).not_to have_no_xpath("//p//a[@id='doesnotexist']", count: 0)
    end

    it 'should be true if the content occurs a different number of times than the given' do
      expect(@session).to have_no_xpath('//p', count: 6)
      expect(@session).to have_no_xpath("//p//a[@id='foo']", count: 2)
      expect(@session).to have_no_xpath("//p[contains(.,'est')]", count: 5)
      expect(@session).to have_no_xpath("//p//a[@id='doesnotexist']", count: 1)
    end
  end

  context 'with text' do
    it 'should discard all matches where the given string is contained' do
      expect(@session).not_to have_no_xpath('//p//a', text: 'Redirect', count: 1)
      expect(@session).to have_no_xpath('//p', text: 'Doesnotexist')
    end

    it 'should discard all matches where the given regexp is matched' do
      expect(@session).not_to have_no_xpath('//p//a', text: /re[dab]i/i, count: 1)
      expect(@session).to have_no_xpath('//p//a', text: /Red$/)
    end
  end
end