html-proofer 3.19.4 → 4.0.0.rc1

data/lib/html-proofer/check/scripts.rb

@@ -1,42 +0,0 @@
-# frozen_string_literal: true
-
-class ScriptCheck < ::HTMLProofer::Check
-  attr_reader :src
-
-  def missing_src?
-    !@script.src
-  end
-
-  def run
-    @html.css('script').each do |node|
-      @script = create_element(node)
-      line = node.line
-      content = node.content
-
-      next if @script.ignore?
-      next unless node.text.strip.empty?
-
-      # does the script exist?
-      if missing_src?
-        add_issue('script is empty and has no src attribute', line: line, content: content)
-      elsif @script.remote?
-        add_to_external_urls(@script.src)
-        check_sri(line, content) if @script.check_sri?
-      elsif !@script.exists?
-        add_issue("internal script #{@script.src} does not exist", line: line, content: content)
-      end
-    end
-
-    external_urls
-  end
-
-  def check_sri(line, content)
-    if !defined?(@script.integrity) && !defined?(@script.crossorigin)
-      add_issue("SRI and CORS not provided in: #{@script.src}", line: line, content: content)
-    elsif !defined?(@script.integrity)
-      add_issue("Integrity is missing in: #{@script.src}", line: line, content: content)
-    elsif !defined?(@script.crossorigin)
-      add_issue("CORS not provided for external resource in: #{@script.src}", line: line, content: content)
-    end
-  end
-end

data/lib/html-proofer/check.rb

@@ -1,75 +0,0 @@
-# frozen_string_literal: true
-
-module HTMLProofer
-  # Mostly handles issue management and collecting of external URLs.
-  class Check
-    attr_reader :node, :html, :element, :src, :path, :options, :issues, :internal_urls, :external_urls
-
-    def initialize(src, path, html, logger, cache, options)
-      @src    = src
-      @path   = path
-      @html   = remove_ignored(html)
-      @logger = logger
-      @cache = cache
-      @options = options
-      @issues = []
-      @internal_urls = {}
-      @external_urls = {}
-    end
-
-    def create_element(node)
-      @node = node
-      Element.new(node, self, @logger)
-    end
-
-    def run
-      raise NotImplementedError, 'HTMLProofer::Check subclasses must implement #run'
-    end
-
-    def add_issue(desc, line: nil, path: nil, status: -1, content: nil)
-      @issues << Issue.new(path || @path, desc, line: line, status: status, content: content)
-      false
-    end
-
-    def add_to_internal_urls(url, internal_url)
-      if @internal_urls[url]
-        @internal_urls[url] << internal_url
-      else
-        @internal_urls[url] = [internal_url]
-      end
-    end
-
-    def add_to_external_urls(url)
-      return if @external_urls[url]
-
-      if @external_urls[url]
-        @external_urls[url] << @path
-      else
-        @external_urls[url] = [@path]
-      end
-    end
-
-    def self.subchecks
-      classes = []
-
-      ObjectSpace.each_object(Class) do |c|
-        next unless c.superclass == self
-
-        classes << c
-      end
-
-      classes
-    end
-
-    def blank?(attr)
-      attr.nil? || attr.empty?
-    end
-
-    private
-
-    def remove_ignored(html)
-      html.css('code, pre, tt').each(&:unlink)
-      html
-    end
-  end
-end

data/lib/html-proofer/element.rb

@@ -1,265 +0,0 @@
-# frozen_string_literal: true
-
-require 'addressable/uri'
-require_relative './utils'
-
-module HTMLProofer
-  # Represents the element currently being processed
-  class Element
-    include HTMLProofer::Utils
-
-    attr_reader :id, :name, :alt, :href, :link, :src, :line, :data_proofer_ignore
-
-    def initialize(obj, check, logger)
-      @logger = logger
-      # Construct readable ivars for every element
-      begin
-        obj.attributes.each_pair do |attribute, value|
-          name = attribute.tr('-:.;@', '_').to_s.to_sym
-          (class << self; self; end).send(:attr_reader, name)
-          instance_variable_set("@#{name}", value.value)
-        end
-      rescue NameError => e
-        @logger.log :error, "Attribute set `#{obj}` contains an error!"
-        raise e
-      end
-
-      @aria_hidden = defined?(@aria_hidden) && @aria_hidden == 'true'
-
-      @data_proofer_ignore = defined?(@data_proofer_ignore)
-
-      @text = obj.content
-      @check = check
-      @checked_paths = {}
-      @type = check.class.name
-      @line = obj.line
-
-      @html = check.html
-
-      parent_attributes = obj.ancestors.map { |a| a.respond_to?(:attributes) && a.attributes }
-      parent_attributes.pop # remove document at the end
-      @parent_ignorable = parent_attributes.any? { |a| !a['data-proofer-ignore'].nil? }
-
-      # fix up missing protocols
-      if defined?(@href)
-        @href.insert(0, 'http:') if %r{^//}.match?(@href)
-      else
-        @href = nil
-      end
-
-      if defined?(@src)
-        @src.insert(0, 'http:') if %r{^//}.match?(@src)
-      else
-        @src = nil
-      end
-
-      if defined?(@srcset)
-        @srcset.insert(0, 'http:') if %r{^//}.match?(@srcset)
-      else
-        @srcset = nil
-      end
-    end
-
-    def url
-      return @url if defined?(@url)
-
-      @url = (@src || @srcset || @href || '').delete("\u200b").strip
-      @url = Addressable::URI.join(base.attr('href') || '', url).to_s if base
-      return @url if @check.options[:url_swap].empty?
-
-      @url = swap(@url, @check.options[:url_swap])
-    end
-
-    def valid?
-      !parts.nil?
-    end
-
-    def path?
-      !parts.host.nil? && !parts.path.nil?
-    end
-
-    def parts
-      @parts ||= Addressable::URI.parse url
-    rescue URI::Error, Addressable::URI::InvalidURIError
-      @parts = nil
-    end
-
-    def path
-      Addressable::URI.unencode parts.path unless parts.nil?
-    end
-
-    def hash
-      parts&.fragment
-    end
-
-    def scheme
-      parts&.scheme
-    end
-
-    # path is to an external server
-    def remote?
-      %w[http https].include? scheme
-    end
-
-    def non_http_remote?
-      !scheme.nil? && !remote?
-    end
-
-    def ignore?
-      return true if @data_proofer_ignore
-      return true if @parent_ignorable
-
-      return true if /^javascript:/.match?(url)
-
-      # ignore base64 encoded images
-      return true if %w[ImageCheck FaviconCheck].include?(@type) && /^data:image/.match?(url)
-
-      # ignore user defined URLs
-      return true if ignores_pattern_check(@check.options[:url_ignore])
-    end
-
-    def ignore_alt?
-      return true if ignores_pattern_check(@check.options[:alt_ignore]) || @aria_hidden
-    end
-
-    def ignore_empty_alt?
-      @check.options[:empty_alt_ignore]
-    end
-
-    def allow_missing_href?
-      @check.options[:allow_missing_href]
-    end
-
-    def allow_hash_href?
-      @check.options[:allow_hash_href]
-    end
-
-    def check_img_http?
-      @check.options[:check_img_http]
-    end
-
-    def check_sri?
-      @check.options[:check_sri]
-    end
-
-    def ignore_empty_mailto?
-      @check.options[:ignore_empty_mailto]
-    end
-
-    # path is external to the file
-    def external?
-      !internal?
-    end
-
-    def internal?
-      relative_link? || internal_absolute_link?
-    end
-
-    def internal_absolute_link?
-      url.start_with?('/')
-    end
-
-    def relative_link?
-      return false if remote?
-
-      hash_link || param_link || url.start_with?('.') || url =~ /^\S/
-    end
-
-    def link_points_to_same_page?
-      hash_link || param_link
-    end
-
-    def hash_link
-      url.start_with?('#')
-    end
-
-    def param_link
-      url.start_with?('?')
-    end
-
-    def absolute_path?(path)
-      path.start_with?('/')
-    end
-
-    def file_path
-      return if path.nil? || path.empty?
-
-      path_dot_ext = ''
-
-      path_dot_ext = path + @check.options[:extension] if @check.options[:assume_extension]
-
-      base = if absolute_path?(path) # path relative to root
-               # either overwrite with root_dir; or, if source is directory, use that; or, just get the current file's dirname
-               @check.options[:root_dir] || (File.directory?(@check.src) ? @check.src : File.dirname(@check.src))
-             elsif File.exist?(File.expand_path(path, @check.src)) || File.exist?(File.expand_path(path_dot_ext, @check.src)) # relative links, path is a file
-               File.dirname(@check.path)
-             elsif File.exist?(File.join(File.dirname(@check.path), path)) || File.exist?(File.join(File.dirname(@check.path), path_dot_ext)) # rubocop:disable Lint/DuplicateBranch; relative links in nested dir, path is a file
-               File.dirname(@check.path)
-             else # relative link, path is a directory
-               @check.path
-             end
-
-      file = File.join(base, path)
-
-      if @check.options[:assume_extension] && File.file?("#{file}#{@check.options[:extension]}")
-        file = "#{file}#{@check.options[:extension]}"
-      elsif File.directory?(file) && !unslashed_directory?(file) # implicit index support
-        file = File.join file, @check.options[:directory_index_file]
-      end
-
-      file
-    end
-
-    # checks if a file exists relative to the current pwd
-    def exists?
-      return @checked_paths[absolute_path] if @checked_paths.key?(absolute_path)
-
-      @checked_paths[absolute_path] = File.exist?(absolute_path)
-    end
-
-    def absolute_path
-      path = file_path || @check.path
-
-      File.expand_path(path, Dir.pwd)
-    end
-
-    def ignores_pattern_check(links)
-      return false unless links.is_a?(Array)
-
-      links.each do |ignore|
-        case ignore
-        when String
-          return true if ignore == url
-        when Regexp
-          return true if ignore&.match?(url)
-        end
-      end
-
-      false
-    end
-
-    def unslashed_directory?(file)
-      File.directory?(file) && !file.end_with?(File::SEPARATOR) && !follow_location?
-    end
-
-    def follow_location?
-      @check.options[:typhoeus] && @check.options[:typhoeus][:followlocation]
-    end
-
-    def base
-      @base ||= @html.at_css('base')
-    end
-
-    def html
-      # If link is on the same page, then URL is on the current page. use the same HTML as for current page
-      if link_points_to_same_page?
-        @html
-      elsif internal?
-        # link on another page, e.g. /about#Team - need to get HTML from the other page
-        create_nokogiri(absolute_path)
-      else
-        raise NotImplementedError, 'HTMLProofer should not have gotten here. Please report this as a bug.'
-      end
-    end
-  end
-end

data/lib/html-proofer/issue.rb

@@ -1,65 +0,0 @@
-# frozen_string_literal: true
-
-module HTMLProofer
-  class Issue
-    attr_reader :path, :desc, :status, :line, :content
-
-    def initialize(path, desc, line: nil, status: -1, content: nil)
-      @line = line.nil? ? '' : " (line #{line})"
-      @path = path
-      @desc = desc
-      @status = status
-      @content = content
-    end
-
-    def to_s
-      "#{@path}: #{@desc}#{@line}"
-    end
-  end
-
-  class SortedIssues
-    attr_reader :issues
-
-    def initialize(issues, error_sort, logger)
-      @issues = issues
-      @error_sort = error_sort
-      @logger = logger
-    end
-
-    def sort_and_report
-      case @error_sort
-      when :path
-        sorted_issues = sort(:path, :desc)
-        report(sorted_issues, :path, :desc)
-      when :desc
-        sorted_issues = sort(:desc, :path)
-        report(sorted_issues, :desc, :path)
-      when :status
-        sorted_issues = sort(:status, :path)
-        report(sorted_issues, :status, :path)
-      end
-    end
-
-    def sort(first_sort, second_sort)
-      issues.sort_by { |t| [t.send(first_sort), t.send(second_sort)] }
-    end
-
-    def report(sorted_issues, first_report, second_report)
-      matcher = nil
-
-      sorted_issues.each do |issue|
-        if matcher != issue.send(first_report)
-          @logger.log :error, "- #{issue.send(first_report)}"
-          matcher = issue.send(first_report)
-        end
-        if first_report == :status
-          @logger.log :error, "  *  #{issue}"
-        else
-          msg = "  *  #{issue.send(second_report)}#{issue.line}"
-          msg = "#{msg}\n     #{issue.content}" if !issue.content.nil? && !issue.content.empty?
-          @logger.log(:error, msg)
-        end
-      end
-    end
-  end
-end

data/lib/html-proofer/middleware.rb

@@ -1,82 +0,0 @@
-# frozen_string_literal: true
-
-module HTMLProofer
-  class Middleware
-    include HTMLProofer::Utils
-
-    class InvalidHtmlError < StandardError
-      def initialize(failures)
-        super
-        @failures = failures
-      end
-
-      def message
-        "HTML Validation errors (skip by adding `?proofer-ignore` to URL): \n#{@failures.join("\n")}"
-      end
-    end
-
-    def self.options
-      @options ||= {
-        type: :file,
-        allow_missing_href: true, # Permitted in html5
-        allow_hash_href: true,
-        check_external_hash: true,
-        check_html: true,
-        url_ignore: [%r{^/}], # Don't try to check if local files exist
-        validation: { report_eof_tags: true }
-      }
-    end
-
-    def initialize(app)
-      @app = app
-    end
-
-    HTML_SIGNATURE = [
-      '<!DOCTYPE HTML',
-      '<HTML',
-      '<HEAD',
-      '<SCRIPT',
-      '<IFRAME',
-      '<H1',
-      '<DIV',
-      '<FONT',
-      '<TABLE',
-      '<A',
-      '<STYLE',
-      '<TITLE',
-      '<B',
-      '<BODY',
-      '<BR',
-      '<P',
-      '<!--'
-    ].freeze
-
-    def call(env)
-      result = @app.call(env)
-      return result if env['REQUEST_METHOD'] != 'GET'
-      return result if /proofer-ignore/.match?(env['QUERY_STRING'])
-      return result if result.first != 200
-
-      body = []
-      result.last.each { |e| body << e }
-
-      body = body.join
-      begin
-        html = body.lstrip
-      rescue StandardError
-        return result # Invalid encoding; it's not gonna be html.
-      end
-      if HTML_SIGNATURE.any? { |sig| html.upcase.start_with? sig }
-        parsed = HTMLProofer::Runner.new(
-          'response',
-          Middleware.options
-        ).check_parsed(
-          Nokogiri::HTML5(html, max_errors: -1), 'response'
-        )
-
-        raise InvalidHtmlError, parsed[:failures] unless parsed[:failures].empty?
-      end
-      result
-    end
-  end
-end