percy-capybara 3.2.0 → 4.0.0.pre.beta1

data/lib/percy/capybara/loaders/base_loader.rb

@@ -1,161 +0,0 @@
-require 'pathname'
-require 'percy/capybara'
-
-module Percy
-  module Capybara
-    module Loaders
-      class BaseLoader
-        # Modified version of Diego Perini's URL regex. https://gist.github.com/dperini/729294
-        URL_REGEX = Regexp.new(
-          # protocol identifier
-          '((?:https?:)?//)' \
-          '(' +
-            # IP address exclusion
-            # private & local networks
-            '(?!(?:10|127)(?:\\.\\d{1,3}){3})' \
-            '(?!(?:169\\.254|192\\.168)(?:\\.\\d{1,3}){2})' \
-            '(?!172\\.(?:1[6-9]|2\\d|3[0-1])(?:\\.\\d{1,3}){2})' +
-            # IP address dotted notation octets
-            # excludes loopback network 0.0.0.0
-            # excludes reserved space >= 224.0.0.0
-            # excludes network & broacast addresses
-            # (first & last IP address of each class)
-            '(?:[1-9]\\d?|1\\d\\d|2[01]\\d|22[0-3])' \
-            '(?:\\.(?:1?\\d{1,2}|2[0-4]\\d|25[0-5])){2}' \
-            '(?:\\.(?:[1-9]\\d?|1\\d\\d|2[0-4]\\d|25[0-4]))' \
-          '|' +
-            # host name
-            '(?:(?:[a-z\\u00a1-\\uffff0-9]-*)*[a-z\\u00a1-\\uffff0-9]+)' +
-            # domain name
-            '(?:\\.(?:[a-z\\u00a1-\\uffff0-9]-*)*[a-z\\u00a1-\\uffff0-9]+)*' \
-          ')' +
-          # port number
-          '(:\\d{2,5})?' +
-          # resource path
-          "(/[^\\s\"']*)?",
-        )
-
-        SKIP_RESOURCE_EXTENSIONS = [
-          '.map', # Ignore source maps.
-          '.gz', # Ignore gzipped files.
-        ].freeze
-
-        MAX_FILESIZE_BYTES = 15 * 1024**2 # 15 MB.
-
-        attr_reader :page
-
-        # @param [Capybara::Session] page The Capybara page.
-        # @param [bool] include_iframes Include iframes in the snapshot
-        def initialize(options = {})
-          @page = options[:page]
-          @include_iframes = options[:include_iframes] || false
-        end
-
-        def build_resources
-          raise NotImplementedError, 'subclass must implement abstract method'
-        end
-
-        def snapshot_resources
-          raise NotImplementedError, 'subclass must implement abstract method'
-        end
-
-        # @private
-        def root_html_resource
-          Percy::Client::Resource.new(
-            current_path, is_root: true, mimetype: 'text/html', content: page.html,
-          )
-        end
-
-        # Transformed version of the current URL to be a relative path.
-        # This important because Rack::Test uses "www.example.com" as the actual current URL,
-        # which would force Percy to actually render example.com instead of the page. By always
-        # using a URL path as the resource URL, we guarantee that Percy will render what it's given.
-        #
-        # @private
-        def current_path
-          current_url = page.current_url
-          url_match = URL_REGEX.match(current_url)
-          return url_match[4] if url_match
-
-          # Special case: prepend a slash to the path to force a valid URL for things like
-          # "about:srcdoc" iframe srcdoc pages.
-          current_url = "/#{current_url}" if current_url[0] != '/'
-
-          current_url
-        end
-
-        # NOTES:
-        # - Doesn't handle multiple iframes with the same URL (`src` attribute)
-        # @private
-        def iframes_resources
-          return [] unless @include_iframes
-
-          resources = []
-
-          page.all(:css, 'iframe').each do |iframe_element|
-            iframe_url = iframe_element[:src]
-            root_page_host = page.current_host
-            begin
-              page.within_frame(iframe_element) do
-                next unless page.current_host == root_page_host
-                path = URI.parse(iframe_url).path
-                content = page.html
-                sha = Digest::SHA256.hexdigest(content)
-                resources <<
-                  Percy::Client::Resource.new(
-                    path,
-                    content: content,
-                    sha: sha,
-                    mimetype: 'text/html',
-                  )
-              end
-            rescue StandardError => e
-              # Skip frame not found errors. This library doesn't explicitly depend on Poltergeist,
-              # so we check the string class name.
-              raise e unless e.class.to_s == 'Capybara::Poltergeist::FrameNotFound' ||
-                  e.class.to_s == 'Capybara::Poltergeist::TimeoutError'
-            end
-          end
-
-          resources
-        rescue ::Capybara::NotSupportedByDriverError
-          []
-        end
-
-        def _resources_from_dir(root_dir, base_url: '/')
-          resources = []
-
-          _find_files(root_dir).each do |path|
-            # Skip certain extensions.
-            next if SKIP_RESOURCE_EXTENSIONS.include?(File.extname(path))
-            # Skip large files, these are hopefully downloads and not used in page rendering.
-            next if File.size(path) > MAX_FILESIZE_BYTES
-
-            # Replace the assets_dir with the base_url to generate the resource_url
-            resource_url = _uri_join(base_url, path.sub(root_dir.to_s, ''))
-
-            sha = Digest::SHA256.hexdigest(File.read(path))
-
-            resources << Percy::Client::Resource.new(resource_url, sha: sha, path: path)
-          end
-
-          resources
-        end
-
-        # A simplified version of Find.find that only returns files and follows symlinks.
-        def _find_files(*paths)
-          paths.flatten.map do |path|
-            path = Pathname.new(path)
-            path.file? ? [path.to_s] : _find_files(path.children)
-          end.flatten
-        end
-
-        def _uri_join(*paths)
-          # We must swap File::SEPARATOR for '/' here because on Windows File.join
-          # will use backslashes and this is a URL.
-          File.join(paths).gsub(File::SEPARATOR, '/')
-        end
-      end
-    end
-  end
-end

data/lib/percy/capybara/loaders/ember_cli_rails_loader.rb

@@ -1,52 +0,0 @@
-require 'percy/capybara/loaders/sprockets_loader'
-require 'set'
-
-module Percy
-  module Capybara
-    module Loaders
-      class EmberCliRailsLoader < SprocketsLoader
-        attr_reader :mounted_apps
-
-        def initialize(mounted_apps, options = {})
-          super(options)
-
-          raise 'mounted_apps is required' unless mounted_apps
-          @mounted_apps = mounted_apps
-        end
-
-        def build_resources
-          resources = super # adds sprockets resources first
-
-          sprockets_resource_urls = resources.collect(&:resource_url)
-          loaded_resource_urls = Set.new(sprockets_resource_urls)
-
-          @mounted_apps.map do |app_name, mount_path|
-            # full path on disk to this ember app
-            # e.g. /Users/djones/Code/rails-ember-app/tmp/ember-cli/apps/frontend
-            dist_path = _dist_path_for_app(app_name)
-
-            resources_from_dir = _resources_from_dir(dist_path, base_url: mount_path)
-
-            resources_from_dir.each do |resource|
-              # avoid loading in duplicate resource_urls
-              next if loaded_resource_urls.include? resource.resource_url
-
-              resources << resource
-              loaded_resource_urls << resource.resource_url
-            end
-          end
-
-          resources
-        end
-
-        def _dist_path_for_app(app_name)
-          _ember_cli.apps[app_name].dist_path
-        end
-
-        def _ember_cli
-          EmberCli if defined?(EmberCli)
-        end
-      end
-    end
-  end
-end

data/lib/percy/capybara/loaders/filesystem_loader.rb

@@ -1,37 +0,0 @@
-require 'percy/capybara/loaders/base_loader'
-require 'digest'
-require 'find'
-require 'pathname'
-
-module Percy
-  module Capybara
-    module Loaders
-      # Resource loader that looks for resources in the specified folder.
-      class FilesystemLoader < BaseLoader
-        def initialize(options = {})
-          # @assets_dir should point to a _compiled_ static assets directory, not source assets.
-          @assets_dir = options[:assets_dir]
-          @base_url = options[:base_url] || '/'
-
-          raise ArgumentError, 'assets_dir is required' if @assets_dir.nil? || @assets_dir == ''
-          unless Pathname.new(@assets_dir).absolute?
-            raise ArgumentError, "assets_dir needs to be an absolute path. Received: #{@assets_dir}"
-          end
-          unless Dir.exist?(@assets_dir)
-            raise ArgumentError, "assets_dir provided was not found. Received: #{@assets_dir}"
-          end
-
-          super
-        end
-
-        def snapshot_resources
-          [root_html_resource]
-        end
-
-        def build_resources
-          _resources_from_dir(@assets_dir, base_url: @base_url)
-        end
-      end
-    end
-  end
-end

data/lib/percy/capybara/loaders/native_loader.rb

@@ -1,287 +0,0 @@
-require 'percy/capybara/loaders/base_loader'
-require 'digest'
-require 'uri'
-
-module Percy
-  module Capybara
-    module Loaders
-      # Resource loader that uses the native Capybara browser interface to discover resources.
-      # This loader uses JavaScript to discover page resources, so specs must be tagged with
-      # "js: true" because the default Rack::Test driver does not support executing JavaScript.
-      class NativeLoader < BaseLoader
-        PATH_REGEX = /\A\/[^\\s\"']*/
-        DATA_URL_REGEX = /\Adata:/
-        LOCAL_HOSTNAMES = [
-          'localhost',
-          '127.0.0.1',
-          '0.0.0.0',
-        ].freeze
-
-        def initialize(options = {})
-          super(options)
-
-          @asset_hostnames = options[:asset_hostnames] || []
-          @assets_from_stylesheets = options[:include_assets_from_stylesheets] || :all
-          @assets_from_stylesheets = ->(_) { true } if @assets_from_stylesheets == :all
-        end
-
-        def snapshot_resources
-          resources = []
-          resources << root_html_resource
-          resources += _get_css_resources
-          resources += _get_image_resources
-          resources += iframes_resources
-          resources
-        end
-
-        def build_resources
-          []
-        end
-
-        # @private
-        def _get_css_resources
-          resources = []
-          # Find all CSS resources.
-          # http://www.quirksmode.org/dom/w3c_css.html#access
-          script = <<-JS
-            function findStylesRecursively(stylesheet, css_urls) {
-              if (stylesheet.href) {  // Skip stylesheet without hrefs (inline stylesheets).
-                css_urls.push(stylesheet.href);
-
-                // Remote stylesheet rules cannot be accessed because of the same-origin policy.
-                // Unfortunately, if you touch .cssRules in Selenium, it throws a JavascriptError
-                // with 'The operation is insecure'. To work around this, skip reading rules of
-                // remote stylesheets but still include them for fetching.
-                //
-                // TODO: If a remote stylesheet has an @import, it will be missing because we don't
-                // notice it here. We could potentially recursively fetch remote imports in
-                // ruby-land below.
-                var parser = document.createElement('a');
-                parser.href = stylesheet.href;
-                if (parser.host != window.location.host) {
-                  return;
-                }
-              }
-              for (var i = 0; i < stylesheet.cssRules.length; i++) {
-                var rule = stylesheet.cssRules[i];
-                // Depth-first search, handle recursive @imports.
-                if (rule.styleSheet) {
-                  findStylesRecursively(rule.styleSheet, css_urls);
-                }
-              }
-            }
-
-            var css_urls = [];
-            for (var i = 0; i < document.styleSheets.length; i++) {
-              findStylesRecursively(document.styleSheets[i], css_urls);
-            }
-            return css_urls;
-          JS
-          resource_urls = _evaluate_script(page, script)
-          urls_referred_by_css = []
-
-          resource_urls.each do |url|
-            next unless _should_include_url?(url)
-            response = _fetch_resource_url(url)
-            urls_referred_by_css.concat(_parse_urls_from_css(response.body))
-            _absolute_url_to_relative!(url, _current_host_port)
-            next unless response
-            resources << Percy::Client::Resource.new(
-              url, mimetype: 'text/css', content: response.body,
-            )
-          end
-          @urls_referred_by_css = urls_referred_by_css
-          resources
-        end
-        private :_get_css_resources
-
-        # @private
-        def _get_image_resources
-          resources = []
-          image_urls = Set.new
-
-          # Find all image tags on the page.
-          page.all('img').each do |image_element|
-            srcs = []
-            srcs << image_element[:src] unless image_element[:src].nil?
-
-            srcset_raw_urls = image_element[:srcset] || ''
-            temp_urls = srcset_raw_urls.split(',')
-            temp_urls.each do |temp_url|
-              srcs << temp_url.split(' ').first
-            end
-
-            srcs.each do |url|
-              image_urls << url
-            end
-          end
-
-          raw_image_urls = _evaluate_script(page, _find_all_css_loaded_background_image_js)
-          raw_image_urls.each do |raw_image_url|
-            temp_urls = raw_image_url.scan(/url\(["']?(.*?)["']?\)/)
-            # background-image can accept multiple url()s, so temp_urls is an array of URLs.
-            temp_urls.each do |temp_url|
-              url = temp_url[0]
-              image_urls << url
-            end
-          end
-
-          if @assets_from_stylesheets && @assets_from_stylesheets != :none
-            image_urls.merge(@urls_referred_by_css.select { |path| @assets_from_stylesheets[path] })
-          end
-
-          image_urls.each do |image_url|
-            # If url references are blank, browsers will often fill them with the current page's
-            # URL, which makes no sense and will never be renderable. Strip these.
-            next if image_url == current_path \
-              || image_url == page.current_url \
-              || image_url.strip.empty?
-
-            # Make the resource URL absolute to the current page. If it is already absolute, this
-            # will have no effect.
-            resource_url = URI.join(page.current_url, image_url).to_s
-
-            # Skip duplicates.
-            next if resources.find { |r| r.resource_url == resource_url }
-
-            next unless _should_include_url?(resource_url)
-
-            # Fetch the images.
-            # TODO(fotinakis): this can be pretty inefficient for image-heavy pages because the
-            # browser has already loaded them once and this fetch cannot easily leverage the
-            # browser's cache. However, often these images are probably local resources served by a
-            # development server, so it may not be so bad. Re-evaluate if this becomes an issue.
-            response = _fetch_resource_url(resource_url)
-            _absolute_url_to_relative!(resource_url, _current_host_port)
-            next unless response
-
-            resources << Percy::Client::Resource.new(
-              resource_url, mimetype: response.content_type, content: response.body,
-            )
-          end
-          resources
-        end
-        private :_get_image_resources
-
-        # @private
-        def _find_all_css_loaded_background_image_js
-          <<-JS
-            var raw_image_urls = [];
-
-            var tags = document.getElementsByTagName('*');
-            var el;
-            var rawValue;
-
-            for (var i = 0; i < tags.length; i++) {
-              el = tags[i];
-              if (el.currentStyle) {
-                rawValue = el.currentStyle['backgroundImage'];
-              } else if (window.getComputedStyle) {
-                rawValue = window.getComputedStyle(el).getPropertyValue('background-image');
-              }
-              if (!rawValue || rawValue === "none") {
-                continue;
-              } else {
-                raw_image_urls.push(rawValue);
-              }
-            }
-            return raw_image_urls;
-          JS
-        end
-
-        # @private
-        def _parse_urls_from_css(css_content)
-          css_content.scan(/url\(([^\)]+)\)/)
-            .map { |i| _remove_quotes(i.first) }
-            .select { |path| _should_include_url?(path) }
-            .map { |path| _remove_hash_from_url(path) }
-        end
-
-        # @private
-        def _remove_hash_from_url(string)
-          if /^(?<url_base>.+)?\#[^#]+$/ =~ string
-            if url_base.end_with?('?')
-              url_base[0...-1]
-            else
-              url_base
-            end
-          else
-            string
-          end
-        end
-
-        # @private
-        def _remove_quotes(string)
-          if string.length >= 2 && (string[0] == string[-1]) && ['"', "'"].include?(string[0])
-            string[1...-1]
-          else
-            string
-          end
-        end
-
-        # @private
-        def _fetch_resource_url(url)
-          response = Percy::Capybara::HttpFetcher.fetch(url)
-          unless response
-            STDERR.puts '[percy] Warning: failed to fetch page resource, ' \
-              "this might be a bug: #{url}"
-            return nil
-          end
-          response
-        end
-        private :_fetch_resource_url
-
-        # @private
-        def _evaluate_script(page, script)
-          script = <<-JS
-            (function() {
-              #{script}
-            })();
-          JS
-          page.evaluate_script(script)
-        end
-        private :_evaluate_script
-
-        # @private
-        def _should_include_url?(url)
-          # It is a URL or a path, but not a data URI.
-          url_match = URL_REGEX.match(url)
-          data_url_match = DATA_URL_REGEX.match(url)
-          result = (url_match || PATH_REGEX.match(url)) && !data_url_match
-
-          # Is not a remote URL.
-          if url_match && !data_url_match
-            host = url_match[2]
-            result = asset_hostnames.include?(host) || _same_server?(url, _current_host_port)
-          end
-
-          !!result
-        end
-
-        # @private
-        def _current_host_port
-          url_match = URL_REGEX.match(page.current_url)
-          url_match[1] + url_match[2] + (url_match[3] || '')
-        end
-        private :_current_host_port
-
-        # @private
-        def _same_server?(url, host_port)
-          url.start_with?(host_port + '/') || url == host_port
-        end
-        private :_same_server?
-
-        # @private
-        def _absolute_url_to_relative!(url, host_port)
-          url.gsub!(host_port + '/', '/') if url.start_with?(host_port + '/')
-        end
-        private :_absolute_url_to_relative!
-
-        def asset_hostnames
-          LOCAL_HOSTNAMES + @asset_hostnames
-        end
-        private :asset_hostnames
-      end
-    end
-  end
-end