html-proofer 4.0.0.rc3 → 4.1.0

data/lib/html_proofer/check/links.rb

@@ -1,118 +1,129 @@
 # frozen_string_literal: true
 
-class HTMLProofer::Check::Links < HTMLProofer::Check
-  def run
-    @html.css('a, link, source').each do |node|
-      @link = create_element(node)
-
-      next if @link.ignore?
-
-      if !allow_hash_href? && @link.node['href'] == '#'
-        add_failure('linking to internal hash #, which points to nowhere', line: @link.line, content: @link.content)
-        next
+module HTMLProofer
+  class Check
+    class Links < HTMLProofer::Check
+      def run
+        @html.css("a, link, source").each do |node|
+          @link = create_element(node)
+
+          next if @link.ignore?
+
+          if !allow_hash_href? && @link.node["href"] == "#"
+            add_failure("linking to internal hash #, which points to nowhere", line: @link.line, content: @link.content)
+            next
+          end
+
+          # is there even an href?
+          if blank?(@link.url.raw_attribute)
+            next if allow_missing_href?
+
+            add_failure("'#{@link.node.name}' tag is missing a reference", line: @link.line, content: @link.content)
+            next
+          end
+
+          # is it even a valid URL?
+          unless @link.url.valid?
+            add_failure("#{@link.href} is an invalid URL", line: @link.line, content: @link.content)
+            next
+          end
+
+          check_schemes
+
+          # intentionally down here because we still want valid? & missing_href? to execute
+          next if @link.url.non_http_remote?
+
+          if !@link.url.internal? && @link.url.remote?
+            check_sri if @runner.check_sri? && @link.link_tag?
+
+            # we need to skip these for now; although the domain main be valid,
+            # curl/Typheous inaccurately return 404s for some links. cc https://git.io/vyCFx
+            next if @link.node["rel"] == "dns-prefetch"
+
+            unless @link.url.path?
+              add_failure("#{@link.url.raw_attribute} is an invalid URL", line: @link.line, content: @link.content)
+              next
+            end
+
+            add_to_external_urls(@link.url, @link.line)
+          elsif @link.url.internal?
+            # does the local directory have a trailing slash?
+            if @link.url.unslashed_directory?(@link.url.absolute_path)
+              add_failure("internally linking to a directory #{@link.url.raw_attribute} without trailing slash",
+                line: @link.line, content: @link.content)
+              next
+            end
+
+            add_to_internal_urls(@link.url, @link.line)
+          end
+        end
       end
 
-      # is there even an href?
-      if blank?(@link.url.raw_attribute)
-        next if allow_missing_href?
-
-        add_failure("'#{@link.node.name}' tag is missing a reference", line: @link.line, content: @link.content)
-        next
+      def allow_missing_href?
+        @runner.options[:allow_missing_href]
       end
 
-      # is it even a valid URL?
-      unless @link.url.valid?
-        add_failure("#{@link.href} is an invalid URL", line: @link.line, content: @link.content)
-        next
+      def allow_hash_href?
+        @runner.options[:allow_hash_href]
       end
 
-      check_schemes
-
-      # intentionally down here because we still want valid? & missing_href? to execute
-      next if @link.url.non_http_remote?
+      def check_schemes
+        case @link.url.scheme
+        when "mailto"
+          handle_mailto
+        when "tel"
+          handle_tel
+        when "http"
+          return unless @runner.options[:enforce_https]
 
-      if !@link.url.internal? && @link.url.remote?
-        check_sri if @runner.check_sri? && @link.link_tag?
-
-        # we need to skip these for now; although the domain main be valid,
-        # curl/Typheous inaccurately return 404s for some links. cc https://git.io/vyCFx
-        next if @link.node['rel'] == 'dns-prefetch'
-
-        unless @link.url.path?
-          add_failure("#{@link.url.raw_attribute} is an invalid URL", line: @link.line, content: @link.content)
-          next
+          add_failure("#{@link.url.raw_attribute} is not an HTTPS link", line: @link.line, content: @link.content)
         end
-
-        add_to_external_urls(@link.url, @link.line)
-      elsif @link.url.internal?
-        # does the local directory have a trailing slash?
-        add_failure("internally linking to a directory #{@link.url.raw_attribute} without trailing slash", line: @link.line, content: @link.content) if @link.url.unslashed_directory?(@link.url.absolute_path)
-
-        add_to_internal_urls(@link.url, @link.line)
       end
-    end
-
-    external_urls
-  end
-
-  def allow_missing_href?
-    @runner.options[:allow_missing_href]
-  end
-
-  def allow_hash_href?
-    @runner.options[:allow_hash_href]
-  end
-
-  def check_schemes
-    case @link.url.scheme
-    when 'mailto'
-      handle_mailto
-    when 'tel'
-      handle_tel
-    when 'http'
-      return unless @runner.options[:enforce_https]
-
-      add_failure("#{@link.url.raw_attribute} is not an HTTPS link", line: @link.line, content: @link.content)
-    end
-  end
 
-  def handle_mailto
-    if @link.url.path.empty?
-      add_failure("#{@link.url.raw_attribute} contains no email address", line: @link.line, content: @link.content) unless ignore_empty_mailto?
-    elsif !/#{URI::MailTo::EMAIL_REGEXP}/o.match?(@link.url.path)
-      add_failure("#{@link.url.raw_attribute} contains an invalid email address", line: @link.line, content: @link.content)
-    end
-  end
+      def handle_mailto
+        if @link.url.path.empty?
+          add_failure("#{@link.url.raw_attribute} contains no email address", line: @link.line,
+            content: @link.content) unless ignore_empty_mailto?
+        elsif !/#{URI::MailTo::EMAIL_REGEXP}/o.match?(@link.url.path)
+          add_failure("#{@link.url.raw_attribute} contains an invalid email address", line: @link.line,
+            content: @link.content)
+        end
+      end
 
-  def handle_tel
-    add_failure("#{@link.url.raw_attribute} contains no phone number", line: @link.line, content: @link.content) if @link.url.path.empty?
-  end
+      def handle_tel
+        add_failure("#{@link.url.raw_attribute} contains no phone number", line: @link.line,
+          content: @link.content) if @link.url.path.empty?
+      end
 
-  def ignore_empty_mailto?
-    @runner.options[:ignore_empty_mailto]
-  end
+      def ignore_empty_mailto?
+        @runner.options[:ignore_empty_mailto]
+      end
 
-  # Whitelist for affected elements from Subresource Integrity specification
-  # https://w3c.github.io/webappsec-subresource-integrity/#link-element-for-stylesheets
-  SRI_REL_TYPES = %(stylesheet)
+      # Allowed elements from Subresource Integrity specification
+      # https://w3c.github.io/webappsec-subresource-integrity/#link-element-for-stylesheets
+      SRI_REL_TYPES = %(stylesheet)
+
+      def check_sri
+        return unless SRI_REL_TYPES.include?(@link.node["rel"])
+
+        if blank?(@link.node["integrity"]) && blank?(@link.node["crossorigin"])
+          add_failure("SRI and CORS not provided in: #{@link.url.raw_attribute}", line: @link.line,
+            content: @link.content)
+        elsif blank?(@link.node["integrity"])
+          add_failure("Integrity is missing in: #{@link.url.raw_attribute}", line: @link.line, content: @link.content)
+        elsif blank?(@link.node["crossorigin"])
+          add_failure("CORS not provided for external resource in: #{@link.link.url.raw_attribute}", line: @link.line,
+            content: @link.content)
+        end
+      end
 
-  def check_sri
-    return unless SRI_REL_TYPES.include?(@link.node['rel'])
+      private def source_tag?
+        @link.node.name == "source"
+      end
 
-    if blank?(@link.node['integrity']) && blank?(@link.node['crossorigin'])
-      add_failure("SRI and CORS not provided in: #{@link.url.raw_attribute}", line: @link.line, content: @link.content)
-    elsif blank?(@link.node['integrity'])
-      add_failure("Integrity is missing in: #{@link.url.raw_attribute}", line: @link.line, content: @link.content)
-    elsif blank?(@link.node['crossorigin'])
-      add_failure("CORS not provided for external resource in: #{@link.link.url.raw_attribute}", line: @link.line, content: @link.content)
+      private def anchor_tag?
+        @link.node.name == "a"
+      end
     end
   end
-
-  private def source_tag?
-    @link.node.name == 'source'
-  end
-
-  private def anchor_tag?
-    @link.node.name == 'a'
-  end
 end

data/lib/html_proofer/check/open_graph.rb

@@ -1,34 +1,39 @@
 # frozen_string_literal: true
 
-class HTMLProofer::Check::OpenGraph < HTMLProofer::Check
-  def run
-    @html.css('meta[property="og:url"], meta[property="og:image"]').each do |node|
-      @open_graph = create_element(node)
+module HTMLProofer
+  class Check
+    class OpenGraph < HTMLProofer::Check
+      def run
+        @html.css('meta[property="og:url"], meta[property="og:image"]').each do |node|
+          @open_graph = create_element(node)
 
-      next if @open_graph.ignore?
+          next if @open_graph.ignore?
 
-      # does the open_graph exist?
-      if missing_content?
-        add_failure('open graph has no content attribute', line: @open_graph.line, content: @open_graph.content)
-      elsif empty_content?
-        add_failure('open graph content attribute is empty', line: @open_graph.line, content: @open_graph.content)
-      elsif !@open_graph.url.valid?
-        add_failure("#{@open_graph.src} is an invalid URL", line: @open_graph.line)
-      elsif @open_graph.url.remote?
-        add_to_external_urls(@open_graph.url, @open_graph.line)
-      else
-        add_failure("internal open graph #{@open_graph.url.raw_attribute} does not exist", line: @open_graph.line, content: @open_graph.content) unless @open_graph.url.exists?
-      end
-    end
+          # does the open_graph exist?
+          if missing_content?
+            add_failure("open graph has no content attribute", line: @open_graph.line, content: @open_graph.content)
+          elsif empty_content?
+            add_failure("open graph content attribute is empty", line: @open_graph.line, content: @open_graph.content)
+          elsif !@open_graph.url.valid?
+            add_failure("#{@open_graph.src} is an invalid URL", line: @open_graph.line)
+          elsif @open_graph.url.remote?
+            add_to_external_urls(@open_graph.url, @open_graph.line)
+          else
+            add_failure("internal open graph #{@open_graph.url.raw_attribute} does not exist", line: @open_graph.line,
+              content: @open_graph.content) unless @open_graph.url.exists?
+          end
+        end
 
-    external_urls
-  end
+        external_urls
+      end
 
-  private def missing_content?
-    @open_graph.node['content'].nil?
-  end
+      private def missing_content?
+        @open_graph.node["content"].nil?
+      end
 
-  private def empty_content?
-    @open_graph.node['content'].empty?
+      private def empty_content?
+        @open_graph.node["content"].empty?
+      end
+    end
   end
 end

data/lib/html_proofer/check/scripts.rb

@@ -1,38 +1,46 @@
 # frozen_string_literal: true
 
-class HTMLProofer::Check::Scripts < HTMLProofer::Check
-  def run
-    @html.css('script').each do |node|
-      @script = create_element(node)
+module HTMLProofer
+  class Check
+    class Scripts < HTMLProofer::Check
+      def run
+        @html.css("script").each do |node|
+          @script = create_element(node)
 
-      next if @script.ignore?
-      next unless @script.content.strip.empty?
+          next if @script.ignore?
+          next unless @script.content.strip.empty?
 
-      # does the script exist?
-      if missing_src?
-        add_failure('script is empty and has no src attribute', line: @script.line, content: @script.content)
-      elsif @script.url.remote?
-        add_to_external_urls(@script.src, @script.line)
-        check_sri if @runner.check_sri?
-      elsif !@script.url.exists?
-        add_failure("internal script reference #{@script.src} does not exist", line: @script.line, content: @script.content)
-      end
-    end
+          # does the script exist?
+          if missing_src?
+            add_failure("script is empty and has no src attribute", line: @script.line, content: @script.content)
+          elsif @script.url.remote?
+            add_to_external_urls(@script.src, @script.line)
+            check_sri if @runner.check_sri?
+          elsif !@script.url.exists?
+            add_failure("internal script reference #{@script.src} does not exist", line: @script.line,
+              content: @script.content)
+          end
+        end
 
-    external_urls
-  end
+        external_urls
+      end
 
-  def missing_src?
-    @script.node['src'].nil?
-  end
+      def missing_src?
+        @script.node["src"].nil?
+      end
 
-  def check_sri
-    if blank?(@script.node['integrity']) && blank?(@script.node['crossorigin'])
-      add_failure("SRI and CORS not provided in: #{@script.url.raw_attribute}", line: @script.line, content: @script.content)
-    elsif blank?(@script.node['integrity'])
-      add_failure("Integrity is missing in: #{@script.url.raw_attribute}", line: @script.line, content: @script.content)
-    elsif blank?(@script.node['crossorigin'])
-      add_failure("CORS not provided for external resource in: #{@script.url.raw_attribute}", line: @script.line, content: @script.content)
+      def check_sri
+        if blank?(@script.node["integrity"]) && blank?(@script.node["crossorigin"])
+          add_failure("SRI and CORS not provided in: #{@script.url.raw_attribute}", line: @script.line,
+            content: @script.content)
+        elsif blank?(@script.node["integrity"])
+          add_failure("Integrity is missing in: #{@script.url.raw_attribute}", line: @script.line,
+            content: @script.content)
+        elsif blank?(@script.node["crossorigin"])
+          add_failure("CORS not provided for external resource in: #{@script.url.raw_attribute}", line: @script.line,
+            content: @script.content)
+        end
+      end
     end
   end
 end

data/lib/html_proofer/check.rb

@@ -21,11 +21,12 @@ module HTMLProofer
     end
 
     def run
-      raise NotImplementedError, 'HTMLProofer::Check subclasses must implement #run'
+      raise NotImplementedError, "HTMLProofer::Check subclasses must implement #run"
     end
 
     def add_failure(description, line: nil, status: nil, content: nil)
-      @failures << Failure.new(@runner.current_path, short_name, description, line: line, status: status, content: content)
+      @failures << Failure.new(@runner.current_filename, short_name, description, line: line, status: status,
+        content: content)
     end
 
     def self.subchecks(runner_options)
@@ -43,11 +44,11 @@ module HTMLProofer
     end
 
     def short_name
-      self.class.name.split('::').last
+      self.class.name.split("::").last
     end
 
     def self.short_name
-      name.split('::').last
+      name.split("::").last
     end
 
     def add_to_internal_urls(url, line)
@@ -57,10 +58,10 @@ module HTMLProofer
 
       metadata = {
         source: @runner.current_source,
-        current_path: @runner.current_path,
+        filename: @runner.current_filename,
         line: line,
         base_url: base_url,
-        found: nil
+        found: false,
       }
       @internal_urls[url_string] << metadata
     end
@@ -70,21 +71,21 @@ module HTMLProofer
 
       @external_urls[url_string] = [] if @external_urls[url_string].nil?
 
-      @external_urls[url_string] << { filename: @runner.current_path, line: line }
+      @external_urls[url_string] << { filename: @runner.current_filename, line: line }
     end
 
     private def base_url
       return @base_url if defined?(@base_url)
 
-      return (@base_url = '') if (base = @html.at_css('base')).nil?
+      return (@base_url = "") if (base = @html.at_css("base")).nil?
 
-      @base_url = base['href']
+      @base_url = base["href"]
     end
 
     private def remove_ignored(html)
       return if html.nil?
 
-      html.css('code, pre, tt').each(&:unlink)
+      html.css("code, pre, tt").each(&:unlink)
       html
     end
   end

data/lib/html_proofer/configuration.rb

@@ -2,45 +2,46 @@
 
 module HTMLProofer
   module Configuration
-    DEFAULT_TESTS = %w[Links Images Scripts].freeze
+    DEFAULT_TESTS = ["Links", "Images", "Scripts"].freeze
 
     PROOFER_DEFAULTS = {
       allow_hash_href: true,
       allow_missing_href: false,
-      assume_extension: '.html',
+      assume_extension: ".html",
       check_external_hash: true,
       checks: DEFAULT_TESTS,
-      directory_index_file: 'index.html',
+      directory_index_file: "index.html",
       disable_external: false,
+      ignore_empty_alt: true,
       ignore_empty_mailto: false,
       ignore_files: [],
       ignore_missing_alt: false,
       ignore_status_codes: [],
       ignore_urls: [],
       enforce_https: true,
-      extensions: ['.html'],
+      extensions: [".html"],
       log_level: :info,
       only_4xx: false,
       swap_attributes: {},
-      swap_urls: {}
+      swap_urls: {},
     }.freeze
 
     TYPHOEUS_DEFAULTS = {
       followlocation: true,
       headers: {
-        'User-Agent' => "Mozilla/5.0 (compatible; HTML Proofer/#{HTMLProofer::VERSION}; +https://github.com/gjtorikian/html-proofer)",
-        'Accept' => 'application/xml,application/xhtml+xml,text/html;q=0.9, text/plain;q=0.8,image/png,*/*;q=0.5'
+        "User-Agent" => "Mozilla/5.0 (compatible; HTML Proofer/#{HTMLProofer::VERSION}; +https://github.com/gjtorikian/html-proofer)",
+        "Accept" => "application/xml,application/xhtml+xml,text/html;q=0.9, text/plain;q=0.8,image/png,*/*;q=0.5",
       },
       connecttimeout: 10,
-      timeout: 30
+      timeout: 30,
     }.freeze
 
     HYDRA_DEFAULTS = {
-      max_concurrency: 50
+      max_concurrency: 50,
     }.freeze
 
     PARALLEL_DEFAULTS = {
-      enable: true
+      enable: true,
     }.freeze
 
     CACHE_DEFAULTS = {}.freeze
@@ -60,20 +61,20 @@ module HTMLProofer
     end
 
     def self.to_regex?(item)
-      if item.start_with?('/') && item.end_with?('/')
-        Regexp.new item[1...-1]
+      if item.start_with?("/") && item.end_with?("/")
+        Regexp.new(item[1...-1])
       else
         item
       end
     end
 
     def self.parse_json_option(option_name, config, symbolize_names: true)
-      raise ArgumentError, 'Must provide an option name in string format.' unless option_name.is_a?(String)
-      raise ArgumentError, 'Must provide an option name in string format.' if option_name.strip.empty?
+      raise ArgumentError, "Must provide an option name in string format." unless option_name.is_a?(String)
+      raise ArgumentError, "Must provide an option name in string format." if option_name.strip.empty?
 
       return {} if config.nil?
 
-      raise ArgumentError, 'Must provide a JSON configuration in string format.' unless config.is_a?(String)
+      raise ArgumentError, "Must provide a JSON configuration in string format." unless config.is_a?(String)
 
       return {} if config.strip.empty?
 

data/lib/html_proofer/element.rb

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 
-require 'addressable/uri'
+require "addressable/uri"
 
 module HTMLProofer
   # Represents the element currently being processed
@@ -26,66 +26,88 @@ module HTMLProofer
 
     def meta_content
       return nil unless meta_tag?
-      return swap_attributes('content') if attribute_swapped?
+      return swap_attributes("content") if attribute_swapped?
 
-      @node['content']
+      @node["content"]
     end
 
     def meta_tag?
-      @node.name == 'meta'
+      @node.name == "meta"
     end
 
     def src
       return nil if !img_tag? && !script_tag? && !source_tag?
-      return swap_attributes('src') if attribute_swapped?
+      return swap_attributes("src") if attribute_swapped?
 
-      @node['src']
+      @node["src"]
     end
 
     def img_tag?
-      @node.name == 'img'
+      @node.name == "img"
     end
 
     def script_tag?
-      @node.name == 'script'
+      @node.name == "script"
     end
 
     def srcset
       return nil if !img_tag? && !source_tag?
-      return swap_attributes('srcset') if attribute_swapped?
+      return swap_attributes("srcset") if attribute_swapped?
 
-      @node['srcset']
+      @node["srcset"]
     end
 
     def source_tag?
-      @node.name == 'source'
+      @node.name == "source"
     end
 
     def href
       return nil if !a_tag? && !link_tag?
-      return swap_attributes('href') if attribute_swapped?
+      return swap_attributes("href") if attribute_swapped?
 
-      @node['href']
+      @node["href"]
     end
 
     def a_tag?
-      @node.name == 'a'
+      @node.name == "a"
     end
 
     def link_tag?
-      @node.name == 'link'
+      @node.name == "link"
     end
 
     def aria_hidden?
-      @node.attributes['aria-hidden']&.value == 'true'
+      @node.attributes["aria-hidden"]&.value == "true"
     end
 
     def multiple_srcsets?
-      !blank?(srcset) && srcset.split(',').size > 1
+      !blank?(srcset) && srcset.split(",").size > 1
+    end
+
+    def srcsets
+      return nil if blank?(srcset)
+
+      srcset.split(",").map(&:strip)
+    end
+
+    def multiple_sizes?
+      return false if blank?(srcsets)
+
+      srcsets.any? do |srcset|
+        !blank?(srcset) && srcset.split(" ").size > 1
+      end
+    end
+
+    def srcsets_wo_sizes
+      return nil if blank?(srcsets)
+
+      srcsets.map do |srcset|
+        srcset.split(" ").first
+      end
     end
 
     def ignore?
-      return true if @node.attributes['data-proofer-ignore']
+      return true if @node.attributes["data-proofer-ignore"]
       return true if ancestors_ignorable?
 
       return true if url&.ignore?
@@ -116,7 +138,7 @@ module HTMLProofer
     private def ancestors_ignorable?
       ancestors_attributes = @node.ancestors.map { |a| a.respond_to?(:attributes) && a.attributes }
       ancestors_attributes.pop # remove document at the end
-      ancestors_attributes.any? { |a| !a['data-proofer-ignore'].nil? }
+      ancestors_attributes.any? { |a| !a["data-proofer-ignore"].nil? }
     end
   end
 end