html-proofer 3.19.4 → 4.0.0

data/lib/html-proofer/cache.rb

@@ -1,194 +0,0 @@
-# frozen_string_literal: true
-
-require_relative 'utils'
-require 'date'
-require 'json'
-require 'uri'
-
-module HTMLProofer
-  class Cache
-    include HTMLProofer::Utils
-
-    DEFAULT_STORAGE_DIR = File.join('tmp', '.htmlproofer')
-    DEFAULT_CACHE_FILE_NAME = 'cache.log'
-
-    URI_REGEXP = URI::DEFAULT_PARSER.make_regexp
-
-    attr_reader :exists, :cache_log, :storage_dir, :cache_file
-
-    def initialize(logger, options)
-      @logger = logger
-      @cache_log = {}
-
-      @cache_datetime = DateTime.now
-      @cache_time = @cache_datetime.to_time
-
-      if options.nil? || options.empty?
-        define_singleton_method('use_cache?') { false }
-      else
-        define_singleton_method('use_cache?') { true }
-        setup_cache!(options)
-        @parsed_timeframe = parsed_timeframe(options[:timeframe])
-      end
-    end
-
-    def within_timeframe?(time)
-      return false if time.nil?
-
-      (@parsed_timeframe..@cache_time).cover?(Time.parse(time))
-    end
-
-    def urls
-      @cache_log['urls'] || []
-    end
-
-    def size
-      @cache_log.length
-    end
-
-    def parsed_timeframe(timeframe)
-      time, date = timeframe.match(/(\d+)(\D)/).captures
-      time = time.to_i
-      case date
-      when 'M'
-        time_ago(time, :months)
-      when 'w'
-        time_ago(time, :weeks)
-      when 'd'
-        time_ago(time, :days)
-      when 'h'
-        time_ago(time, :hours)
-      else
-        raise ArgumentError, "#{date} is not a valid timeframe!"
-      end
-    end
-
-    def add(url, filenames, status, msg = '')
-      return unless use_cache?
-
-      data = {
-        time: @cache_time,
-        filenames: filenames,
-        status: status,
-        message: msg
-      }
-
-      @cache_log[clean_url(url)] = data
-    end
-
-    def detect_url_changes(found, type)
-      found_urls = found.keys.map { |url| clean_url(url) }
-
-      # if there were no urls, bail
-      return {} if found_urls.empty?
-
-      existing_urls = @cache_log.keys.map { |url| clean_url(url) }
-
-      # prepare to add new URLs detected
-      additions = found.reject do |url, _|
-        url = clean_url(url)
-        if existing_urls.include?(url)
-          true
-        else
-          @logger.log :debug, "Adding #{url} to cache check"
-          false
-        end
-      end
-
-      new_link_count = additions.length
-      new_link_text = pluralize(new_link_count, 'link', 'links')
-      @logger.log :info, "Adding #{new_link_text} to the cache..."
-
-      # remove from cache URLs that no longer exist
-      deletions = 0
-      @cache_log.delete_if do |url, _|
-        url = clean_url(url)
-
-        if found_urls.include?(url)
-          false
-        elsif url_matches_type?(url, type)
-          @logger.log :debug, "Removing #{url} from cache check"
-          deletions += 1
-          true
-        end
-      end
-
-      del_link_text = pluralize(deletions, 'link', 'links')
-      @logger.log :info, "Removing #{del_link_text} from the cache..."
-
-      additions
-    end
-
-    # TODO: Garbage performance--both the external and internal
-    # caches need access to this file. Write a proper versioned
-    # schema in the future
-    def write
-      File.write(cache_file, @cache_log.to_json)
-    end
-
-    def load?
-      @load.nil?
-    end
-
-    def retrieve_urls(urls, type)
-      urls_to_check = detect_url_changes(urls, type)
-
-      @cache_log.each_pair do |url, cache|
-        next if within_timeframe?(cache['time']) && cache['message'].empty? # these were successes to skip
-
-        if url_matches_type?(url, type)
-          urls_to_check[url] = cache['filenames'] # recheck expired links
-        end
-      end
-      urls_to_check
-    end
-
-    # FIXME: it seems that Typhoeus actually acts on escaped URLs,
-    # but there's no way to get at that information, and the cache
-    # stores unescaped URLs. Because of this, some links, such as
-    # github.com/search/issues?q=is:open+is:issue+fig are not matched
-    # as github.com/search/issues?q=is%3Aopen+is%3Aissue+fig
-    def unescape_url(url)
-      Addressable::URI.unescape(url)
-    end
-
-    def clean_url(url)
-      unescape_url(url)
-    end
-
-    def setup_cache!(options)
-      @storage_dir = options[:storage_dir] || DEFAULT_STORAGE_DIR
-
-      FileUtils.mkdir_p(storage_dir) unless Dir.exist?(storage_dir)
-
-      cache_file_name = options[:cache_file] || DEFAULT_CACHE_FILE_NAME
-
-      @cache_file = File.join(storage_dir, cache_file_name)
-
-      return unless File.exist?(@cache_file)
-
-      contents = File.read(@cache_file)
-      @cache_log = contents.empty? ? {} : JSON.parse(contents)
-    end
-
-    private
-
-    def time_ago(measurement, unit)
-      case unit
-      when :months
-        @cache_datetime >> -measurement
-      when :weeks
-        @cache_datetime - (measurement * 7)
-      when :days
-        @cache_datetime - measurement
-      when :hours
-        @cache_datetime - Rational(measurement / 24.0)
-      end.to_time
-    end
-
-    def url_matches_type?(url, type)
-      return true if type == :internal && url !~ URI_REGEXP
-      return true if type == :external && url =~ URI_REGEXP
-    end
-  end
-end

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

@@ -1,29 +0,0 @@
-# frozen_string_literal: true
-
-class FaviconCheck < ::HTMLProofer::Check
-  def run
-    found = false
-    @html.xpath('//link[not(ancestor::pre or ancestor::code)]').each do |node|
-      favicon = create_element(node)
-      next if favicon.ignore?
-
-      found = true if favicon.rel.split.last.eql? 'icon'
-      break if found
-    end
-
-    return if found
-
-    return if immediate_redirect?
-
-    add_issue('no favicon specified')
-  end
-
-  private
-
-  # allow any instant-redirect meta tag
-  def immediate_redirect?
-    @html.xpath("//meta[@http-equiv='refresh']").attribute('content').value.start_with? '0;'
-  rescue StandardError
-    false
-  end
-end

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

@@ -1,37 +0,0 @@
-# frozen_string_literal: true
-
-class HtmlCheck < ::HTMLProofer::Check
-  # tags embedded in scripts are used in templating languages: http://git.io/vOovv
-  SCRIPT_EMBEDS_MSG = /Element script embeds close tag/.freeze
-  INVALID_TAG_MSG = /Tag ([\w\-:]+) invalid/.freeze
-  INVALID_PREFIX = /Namespace prefix/.freeze
-  PARSE_ENTITY_REF = /htmlParseEntityRef: no name/.freeze
-  DOCTYPE_MSG = /Expected a doctype token/.freeze
-  EOF_IN_TAG = /End of input in tag/.freeze
-  MISMATCHED_TAGS = /That tag isn't allowed here/.freeze
-
-  def run
-    @html.errors.each do |error|
-      add_issue(error.message, line: error.line) if report?(error.message)
-    end
-  end
-
-  def report?(message)
-    case message
-    when SCRIPT_EMBEDS_MSG
-      options[:validation][:report_script_embeds]
-    when INVALID_TAG_MSG, INVALID_PREFIX
-      options[:validation][:report_invalid_tags]
-    when PARSE_ENTITY_REF
-      options[:validation][:report_missing_names]
-    when DOCTYPE_MSG
-      options[:validation][:report_missing_doctype]
-    when EOF_IN_TAG
-      options[:validation][:report_eof_tags]
-    when MISMATCHED_TAGS
-      options[:validation][:report_mismatched_tags]
-    else
-      true
-    end
-  end
-end

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

@@ -1,48 +0,0 @@
-# frozen_string_literal: true
-
-class ImageCheck < ::HTMLProofer::Check
-  SCREEN_SHOT_REGEX = /Screen(?: |%20)Shot(?: |%20)\d+-\d+-\d+(?: |%20)at(?: |%20)\d+.\d+.\d+/.freeze
-
-  def empty_alt_tag?
-    @img.alt.nil? || @img.alt.strip.empty?
-  end
-
-  def terrible_filename?
-    @img.url =~ SCREEN_SHOT_REGEX
-  end
-
-  def missing_src?
-    blank?(@img.url)
-  end
-
-  def run
-    @html.css('img').each do |node|
-      @img = create_element(node)
-      line = node.line
-      content = node.content
-
-      next if @img.ignore?
-
-      # screenshot filenames should return because of terrible names
-      if terrible_filename?
-        add_issue("image has a terrible filename (#{@img.url})", line: line, content: content)
-        next
-      end
-
-      # does the image exist?
-      if missing_src?
-        add_issue('image has no src or srcset attribute', line: line, content: content)
-      elsif @img.remote?
-        add_to_external_urls(@img.url)
-      elsif !@img.exists?
-        add_issue("internal image #{@img.url} does not exist", line: line, content: content)
-      end
-
-      add_issue("image #{@img.url} does not have an alt attribute", line: line, content: content) if empty_alt_tag? && !@img.ignore_empty_alt? && !@img.ignore_alt?
-
-      add_issue("image #{@img.url} uses the http scheme", line: line, content: content) if @img.check_img_http? && @img.scheme == 'http'
-    end
-
-    external_urls
-  end
-end

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

@@ -1,182 +0,0 @@
-# frozen_string_literal: true
-
-class LinkCheck < ::HTMLProofer::Check
-  include HTMLProofer::Utils
-
-  def missing_href?
-    return blank?(@link.src) if @node.name == 'source'
-
-    blank?(@link.href) && blank?(@link.name) && blank?(@link.id)
-  end
-
-  def placeholder?
-    (!blank?(@link.id) || !blank?(@link.name)) && @link.href.nil?
-  end
-
-  def run
-    @html.css('a, link, source').each do |node|
-      @link = create_element(node)
-      line = node.line
-      content = node.to_s
-
-      next if @link.ignore?
-
-      next if placeholder?
-      next if @link.allow_hash_href? && @link.href == '#'
-
-      # is it even a valid URL?
-      unless @link.valid?
-        add_issue("#{@link.href} is an invalid URL", line: line, content: content)
-        next
-      end
-
-      check_schemes(@link, line, content)
-
-      # is there even an href?
-      if missing_href?
-        next if @link.allow_missing_href?
-        # HTML5 allows dropping the href: http://git.io/vBX0z
-        next if @html.internal_subset.nil? || (@html.internal_subset.name == 'html' && @html.internal_subset.external_id.nil?)
-
-        add_issue('anchor has no href attribute', line: line, content: content)
-        next
-      end
-
-      # intentionally here because we still want valid? & missing_href? to execute
-      next if @link.non_http_remote?
-
-      if !@link.href&.start_with?('#') && !@link.internal? && @link.remote?
-        check_sri(line, content) if @link.check_sri? && node.name == 'link'
-        # 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.respond_to?(:rel) && @link.rel == 'dns-prefetch'
-
-        unless @link.path?
-          add_issue("#{@link.href} is an invalid URL", line: line, content: content)
-          next
-        end
-
-        add_to_external_urls(@link.href || @link.src)
-        next
-      elsif @link.internal?
-        add_to_internal_urls(@link.href, InternalLink.new(@link, @path, line, content))
-        add_issue("internally linking to #{@link.href}, which does not exist", line: line, content: content) if !@link.exists? && !@link.hash
-      end
-    end
-
-    external_urls
-  end
-
-  def check_internal_link(link, path, line, content)
-    # does the local directory have a trailing slash?
-    if link.unslashed_directory?(link.absolute_path)
-      add_issue("internally linking to a directory #{link.absolute_path} without trailing slash", path: path, line: line, content: content)
-      return false
-    end
-
-    return true unless link.hash
-
-    # verify the target hash
-    handle_hash(link, path, line, content)
-  end
-
-  def check_schemes(link, line, content)
-    case link.scheme
-    when 'mailto'
-      handle_mailto(link, line, content)
-    when 'tel'
-      handle_tel(link, line, content)
-    when 'http'
-      return unless @options[:enforce_https]
-
-      add_issue("#{link.href} is not an HTTPS link", line: line, content: content)
-    end
-  end
-
-  def handle_mailto(link, line, content)
-    if link.path.empty?
-      add_issue("#{link.href} contains no email address", line: line, content: content) unless link.ignore_empty_mailto?
-    elsif !link.path.include?('@')
-      add_issue("#{link.href} contains an invalid email address", line: line, content: content)
-    end
-  end
-
-  def handle_tel(link, line, content)
-    add_issue("#{link.href} contains no phone number", line: line, content: content) if link.path.empty?
-  end
-
-  def handle_hash(link, path, line, content)
-    if link.internal? && !hash_exists?(link.html, link.hash) # rubocop:disable Style/GuardClause
-      return add_issue("linking to internal hash ##{link.hash} that does not exist", path: path, line: line, content: content)
-    elsif link.external?
-      return external_link_check(link, line, content)
-    end
-
-    true
-  end
-
-  def external_link_check(link, line, content)
-    if link.exists? # rubocop:disable Style/GuardClause
-      target_html = create_nokogiri(link.absolute_path)
-      return add_issue("linking to #{link.href}, but #{link.hash} does not exist", line: line, content: content) unless hash_exists?(target_html, link.hash)
-    else
-      return add_issue("trying to find hash of #{link.href}, but #{link.absolute_path} does not exist", line: line, content: content)
-    end
-
-    true
-  end
-
-  def hash_exists?(html, href_hash)
-    decoded_href_hash = Addressable::URI.unescape(href_hash)
-    fragment_ids = [href_hash, decoded_href_hash]
-    # https://www.w3.org/TR/html5/single-page.html#scroll-to-fragid
-    fragment_ids.include?('top') || !find_fragments(html, fragment_ids).empty?
-  end
-
-  def find_fragments(html, fragment_ids)
-    xpaths = fragment_ids.flat_map do |frag_id|
-      escaped_frag_id = "'#{frag_id.split("'").join("', \"'\", '")}', ''"
-      [
-        "//*[case_sensitive_equals(@id, concat(#{escaped_frag_id}))]",
-        "//*[case_sensitive_equals(@name, concat(#{escaped_frag_id}))]"
-      ]
-    end
-    xpaths << XpathFunctions.new
-
-    html.xpath(*xpaths)
-  end
-
-  # Whitelist for affected elements from Subresource Integrity specification
-  # https://w3c.github.io/webappsec-subresource-integrity/#link-element-for-stylesheets
-  SRI_REL_TYPES = %(stylesheet)
-
-  def check_sri(line, content)
-    return unless SRI_REL_TYPES.include?(@link.rel)
-
-    if !defined?(@link.integrity) && !defined?(@link.crossorigin)
-      add_issue("SRI and CORS not provided in: #{@link.src}", line: line, content: content)
-    elsif !defined?(@link.integrity)
-      add_issue("Integrity is missing in: #{@link.src}", line: line, content: content)
-    elsif !defined?(@link.crossorigin)
-      add_issue("CORS not provided for external resource in: #{@link.src}", line: line, content: content)
-    end
-  end
-
-  class XpathFunctions
-    def case_sensitive_equals(node_set, str_to_match)
-      node_set.find_all { |node| node.to_s.== str_to_match.to_s }
-    end
-  end
-
-  class InternalLink
-    attr_reader :link, :href, :path, :line, :content
-
-    def initialize(link, path, line, content)
-      @link = link
-      @href = @link.href
-      @path = path
-      @line = line
-      @content = content
-    end
-  end
-end

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

@@ -1,46 +0,0 @@
-# frozen_string_literal: true
-
-class OpenGraphElement < ::HTMLProofer::Element
-  attr_reader :src
-
-  def initialize(obj, check, logger)
-    super(obj, check, logger)
-    # Fake up src from the content attribute
-    instance_variable_set('@src', @content)
-
-    @src.insert 0, 'http:' if %r{^//}.match?(@src)
-  end
-end
-
-class OpenGraphCheck < ::HTMLProofer::Check
-  def missing_src?
-    !@opengraph.src
-  end
-
-  def empty_src?
-    blank?(@opengraph.src)
-  end
-
-  def run
-    @html.css('meta[property="og:url"], meta[property="og:image"]').each do |m|
-      @opengraph = OpenGraphElement.new(m, self, @logger)
-
-      next if @opengraph.ignore?
-
-      # does the opengraph exist?
-      if missing_src?
-        add_issue('open graph has no content attribute', line: m.line, content: m.content)
-      elsif empty_src?
-        add_issue('open graph content attribute is empty', line: m.line, content: m.content)
-      elsif !@opengraph.valid?
-        add_issue("#{@opengraph.src} is an invalid URL", line: m.line)
-      elsif @opengraph.remote?
-        add_to_external_urls(@opengraph.url)
-      else
-        add_issue("internal open graph #{@opengraph.url} does not exist", line: m.line, content: m.content) unless @opengraph.exists?
-      end
-    end
-
-    external_urls
-  end
-end

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