html-proofer 2.6.4 → 3.0.0

data/lib/{html/proofer/checks → html-proofer/check}/links.rb

@@ -1,72 +1,58 @@
-class LinkCheckable < ::HTML::Proofer::Checkable
-
-  def href
-    real_attr @href
-  end
-
-  def id
-    real_attr @id
-  end
-
-  def name
-    real_attr @name
-  end
+class LinkCheck < ::HTMLProofer::Check
+  include HTMLProofer::Utils
 
   def missing_href?
-    href.nil? && name.nil? && id.nil?
+    blank?(@link.href) && blank?(@link.name) && blank?(@link.id)
   end
 
   def placeholder?
-    (id || name) && href.nil?
+    (!blank?(@link.id) || !blank?(@link.name)) && @link.href.nil?
   end
-end
-
-class LinkCheck < ::HTML::Proofer::CheckRunner
-  include HTML::Proofer::Utils
 
   def run
     @html.css('a, link').each do |node|
-      link = LinkCheckable.new(node, self)
-      line = node.line
+      @link = create_element(node)
+      line = @node.line
+
+      next if @link.ignore?
 
-      next if link.ignore?
-      next if link.href =~ /^javascript:/ # can't put this in ignore? because the URI does not parse
-      next if link.placeholder?
-      next if link.allow_hash_href? && link.href == '#'
+      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)
+      unless @link.valid?
+        add_issue("#{@link.href} is an invalid URL", line: line)
         next
       end
 
-      check_schemes(link, line)
+      check_schemes(@link, line)
 
-      # is there even a href?
-      if link.missing_href?
-        add_issue('anchor has no href attribute', line)
+      # is there even an href?
+      if missing_href?
+        # HTML5 allows dropping the href: http://git.io/vBX0z
+        next if @html.internal_subset.name == 'html' && @html.internal_subset.external_id.nil?
+        add_issue('anchor has no href attribute', line: line)
         next
       end
 
       # intentionally here because we still want valid? & missing_href? to execute
-      next if link.non_http_remote?
-
+      next if @link.non_http_remote?
       # does the file even exist?
-      if link.remote?
-        add_to_external_urls(link.href, line)
+      if @link.remote?
+        add_to_external_urls(@link.href, line)
         next
-      elsif !link.internal?
-        add_issue("internally linking to #{link.href}, which does not exist", line) unless link.exists?
+      elsif !@link.internal? && !@link.exists?
+        add_issue("internally linking to #{@link.href}, which does not exist", line: line)
       end
 
       # 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", line)
+      if @link.unslashed_directory? @link.absolute_path
+        add_issue("internally linking to a directory #{@link.absolute_path} without trailing slash", line: line)
         next
       end
 
       # verify the target hash
-      handle_hash(link, line) if link.hash
+      handle_hash(@link, line) if @link.hash
     end
 
     external_urls
@@ -79,26 +65,27 @@ class LinkCheck < ::HTML::Proofer::CheckRunner
     when 'tel'
       handle_tel(link, line)
     when 'http'
-      add_issue("#{link.href} is not an HTTPS link", line) if @options[:enforce_https]
+      return unless @options[:enforce_https]
+      add_issue("#{link.href} is not an HTTPS link", line: line)
     end
   end
 
   def handle_mailto(link, line)
     if link.path.empty?
-      add_issue("#{link.href} contains no email address", line)
+      add_issue("#{link.href} contains no email address", line: line)
     elsif !link.path.include?('@')
-      add_issue("#{link.href} contains an invalid email address", line)
+      add_issue("#{link.href} contains an invalid email address", line: line)
     end
   end
 
   def handle_tel(link, line)
-    add_issue("#{link.href} contains no phone number", line) if link.path.empty?
+    add_issue("#{link.href} contains no phone number", line: line) if link.path.empty?
   end
 
   def handle_hash(link, line)
     if link.internal?
       unless hash_check @html, link.hash
-        add_issue("linking to internal hash ##{link.hash} that does not exist", line)
+        add_issue("linking to internal hash ##{link.hash} that does not exist", line: line)
       end
     elsif link.external?
       external_link_check(link, line)
@@ -107,11 +94,11 @@ class LinkCheck < ::HTML::Proofer::CheckRunner
 
   def external_link_check(link, line)
     if !link.exists?
-      add_issue("trying to find hash of #{link.href}, but #{link.absolute_path} does not exist", line)
+      add_issue("trying to find hash of #{link.href}, but #{link.absolute_path} does not exist", line: line)
     else
       target_html = create_nokogiri link.absolute_path
       unless hash_check target_html, link.hash
-        add_issue("linking to #{link.href}, but #{link.hash} does not exist", line)
+        add_issue("linking to #{link.href}, but #{link.hash} does not exist", line: line)
       end
     end
   end
@@ -122,7 +109,12 @@ class LinkCheck < ::HTML::Proofer::CheckRunner
                "//*[case_insensitive_equals(@name, '#{href_hash}')]", \
                "//*[case_insensitive_equals(@id, '#{decoded_href_hash}')]", \
                "//*[case_insensitive_equals(@name, '#{decoded_href_hash}')]", \
-               HTML::Proofer::XpathFunctions.new).length > 0
+               XpathFunctions.new).length > 0
   end
 
+  class XpathFunctions
+    def case_insensitive_equals(node_set, str_to_match)
+      node_set.find_all {|node| node.to_s.downcase == str_to_match.to_s.downcase }
+    end
+  end
 end

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

@@ -0,0 +1,28 @@
+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
+
+      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)
+      elsif @script.remote?
+        add_to_external_urls(@script.src, line)
+      elsif !@script.exists?
+        add_issue("internal script #{@script.src} does not exist", line: line)
+      end
+    end
+
+    external_urls
+  end
+end

data/lib/html-proofer/configuration.rb

@@ -0,0 +1,56 @@
+module HTMLProofer
+  module Configuration
+    require_relative 'version'
+
+    PROOFER_DEFAULTS = {
+      :allow_hash_href => false,
+      :alt_ignore => [],
+      :assume_extension => false,
+      :check_external_hash => false,
+      :check_favicon => false,
+      :check_html => false,
+      :checks_to_ignore => [],
+      :directory_index_file => 'index.html',
+      :disable_external => false,
+      :empty_alt_ignore => false,
+      :enforce_https => false,
+      :error_sort => :path,
+      :extension => '.html',
+      :external_only => false,
+      :file_ignore => [],
+      :http_status_ignore => [],
+      :log_level => :info,
+      :only_4xx => false,
+      :url_ignore => [],
+      :url_swap => []
+    }
+
+    TYPHOEUS_DEFAULTS = {
+      :followlocation => true,
+      :headers => {
+        'User-Agent' => "Mozilla/5.0 (compatible; HTML Proofer/#{HTMLProofer::VERSION}; +https://github.com/gjtorikian/html-proofer)"
+      }
+    }
+
+    HYDRA_DEFAULTS = {
+      :max_concurrency => 50
+    }
+
+    PARALLEL_DEFAULTS = {}
+
+    VALIDATION_DEFAULTS = {
+      :report_script_embeds => false,
+      :report_invalid_tags => false
+    }
+
+    CACHE_DEFAULTS = {}
+
+    def self.to_regex?(item)
+      if item.start_with?('/') && item.end_with?('/')
+        Regexp.new item[1...-1]
+      else
+        item
+      end
+    end
+  end
+end

data/lib/html-proofer/element.rb

@@ -0,0 +1,165 @@
+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
+
+    def initialize(obj, check)
+      # Contruct readable ivars for every element
+      obj.attributes.each_pair do |attribute, value|
+        name = "#{attribute.tr('-:.', '_')}".to_sym
+        (class << self; self; end).send(:attr_reader, name)
+        instance_variable_set("@#{name}", value.value)
+      end
+
+      @text = obj.content
+      @check = check
+      @checked_paths = {}
+      @type = check.class.name
+      @line = obj.line
+
+      # fix up missing protocols
+      @href.insert 0, 'http:' if @href =~ %r{^//}
+      @src.insert 0, 'http:' if @src =~ %r{^//}
+    end
+
+    def url
+      url = @src || @srcset || @href || ''
+      return url if @check.options[:url_swap].empty?
+      swap(url, @check.options[:url_swap])
+    end
+
+    def valid?
+      !parts.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 unless parts.nil?
+    end
+
+    def scheme
+      parts.scheme unless parts.nil?
+    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 url.match(/^javascript:/)
+
+      # ignore base64 encoded images
+      if %w(ImageCheck FaviconCheck).include? @type
+        return true if url.match(/^data:image/)
+      end
+
+      # ignore user defined URLs
+      return true if ignores_pattern_check(@check.options[:url_ignore])
+
+      # ignore user defined alts
+      return false unless 'ImageCheck' == @type
+      return true if ignores_pattern_check(@check.options[:alt_ignore])
+    end
+
+    def ignore_empty_alt?
+      @check.options[:empty_alt_ignore]
+    end
+
+    def allow_hash_href?
+      @check.options[:allow_hash_href]
+    end
+
+    # path is external to the file
+    def external?
+      !internal?
+    end
+
+    # path is an anchor or a query
+    def internal?
+      url.start_with? '#', '?'
+    end
+
+    def file_path
+      return if path.nil?
+
+      path_dot_ext = ''
+
+      if @check.options[:assume_extension]
+        path_dot_ext = path + @check.options[:extension]
+      end
+
+      if path =~ %r{^/} # path relative to root
+        base = 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
+        base = 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)) # relative links in nested dir, path is a file
+        base = File.dirname @check.path
+      else # relative link, path is a directory
+        base = @check.path
+      end
+
+      file = File.join base, path
+
+      # implicit index support
+      if File.directory?(file) && !unslashed_directory?(file)
+        file = File.join file, @check.options[:directory_index_file]
+      elsif @check.options[:assume_extension] && File.file?("#{file}#{@check.options[:extension]}")
+        file = "#{file}#{@check.options[:extension]}"
+      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)
+      links.each do |ignore|
+        if ignore.is_a? String
+          return true if ignore == url
+        elsif ignore.is_a? Regexp
+          return true if ignore =~ 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
+  end
+end

data/lib/{html/proofer/check_runner → html-proofer}/issue.rb

@@ -1,18 +1,16 @@
-# encoding: utf-8
-class HTML::Proofer::CheckRunner
-
+module HTMLProofer
   class Issue
-    attr_reader :path, :desc, :status, :line_number
+    attr_reader :path, :desc, :status, :line
 
-    def initialize(path, desc, line_number = nil, status = -1)
-      @line_number = line_number.nil? ? '' : " (line #{line_number})"
+    def initialize(path, desc, line: nil, status: -1)
+      @line = line.nil? ? '' : " (line #{line})"
       @path = path
       @desc = desc
       @status = status
     end
 
     def to_s
-      "#{@path}: #{@desc}#{@line_number}"
+      "#{@path}: #{@desc}#{@line}"
     end
   end
 
@@ -48,13 +46,13 @@ class HTML::Proofer::CheckRunner
 
       sorted_issues.each do |issue|
         if matcher != issue.send(first_report)
-          @logger.log :error, :red, "- #{issue.send(first_report)}"
+          @logger.log :error, "- #{issue.send(first_report)}"
           matcher = issue.send(first_report)
         end
         if first_report == :status
-          @logger.log :error, :red, "  *  #{issue}"
+          @logger.log :error, "  *  #{issue}"
         else
-          @logger.log :error, :red, "  *  #{issue.send(second_report)}#{issue.line_number}"
+          @logger.log :error, "  *  #{issue.send(second_report)}#{issue.line}"
         end
       end
     end