html-proofer 4.0.0.rc3 → 4.1.0

data/lib/html_proofer/cache.rb

@@ -1,8 +1,8 @@
 # frozen_string_literal: true
 
-require 'date'
-require 'json'
-require 'uri'
+require "date"
+require "json"
+require "uri"
 
 module HTMLProofer
   class Cache
@@ -10,8 +10,8 @@ module HTMLProofer
 
     CACHE_VERSION = 2
 
-    DEFAULT_STORAGE_DIR = File.join('tmp', '.htmlproofer')
-    DEFAULT_CACHE_FILE_NAME = 'cache.json'
+    DEFAULT_STORAGE_DIR = File.join("tmp", ".htmlproofer")
+    DEFAULT_CACHE_FILE_NAME = "cache.json"
 
     URI_REGEXP = URI::DEFAULT_PARSER.make_regexp
 
@@ -21,7 +21,7 @@ module HTMLProofer
       @runner = runner
       @logger = @runner.logger
 
-      @cache_datetime = DateTime.now
+      @cache_datetime = Time.now
       @cache_time = @cache_datetime.to_time
 
       if blank?(options)
@@ -29,28 +29,25 @@ module HTMLProofer
       else
         define_singleton_method(:enabled?) { true }
         setup_cache!(options)
-        @parsed_timeframe = parsed_timeframe(options[:timeframe])
-      end
-    end
 
-    def within_timeframe?(time)
-      return false if time.nil?
-
-      time = Time.parse(time) if time.is_a?(String)
-      (@parsed_timeframe..@cache_time).cover?(time)
+        @external_timeframe = parsed_timeframe(options[:timeframe][:external])
+        @internal_timeframe = parsed_timeframe(options[:timeframe][:internal])
+      end
     end
 
     def parsed_timeframe(timeframe)
+      return nil if timeframe.nil?
+
       time, date = timeframe.match(/(\d+)(\D)/).captures
       time = time.to_i
       case date
-      when 'M'
+      when "M"
         time_ago(time, :months)
-      when 'w'
+      when "w"
         time_ago(time, :weeks)
-      when 'd'
+      when "d"
         time_ago(time, :days)
-      when 'h'
+      when "h"
         time_ago(time, :hours)
       else
         raise ArgumentError, "#{date} is not a valid timeframe!"
@@ -71,7 +68,8 @@ module HTMLProofer
       found = status_code.between?(200, 299)
 
       clean_url = cleaned_url(url)
-      @cache_log[:external][clean_url] = { time: @cache_time.to_s, found: found, status_code: status_code, message: msg, metadata: filenames }
+      @cache_log[:external][clean_url] =
+        { time: @cache_time.to_s, found: found, status_code: status_code, message: msg, metadata: filenames }
     end
 
     def detect_url_changes(urls_detected, type)
@@ -82,39 +80,104 @@ module HTMLProofer
       additions
     end
 
+    def write
+      return unless enabled?
+
+      File.write(@cache_file, @cache_log.to_json)
+    end
+
+    def retrieve_urls(urls_detected, type)
+      # if there are no urls, bail
+      return {} if urls_detected.empty?
+
+      urls_detected = urls_detected.transform_keys do |url|
+        cleaned_url(url)
+      end
+
+      urls_to_check = detect_url_changes(urls_detected, type)
+
+      @cache_log[type].each_pair do |url, cache|
+        within_timeframe = type == :external ? within_external_timeframe?(cache[:time]) : within_internal_timeframe?(cache[:time])
+        next if within_timeframe
+
+        urls_to_check[url] = cache[:metadata] # recheck expired links
+      end
+
+      urls_to_check
+    end
+
+    def within_external_timeframe?(time)
+      within_timeframe?(time, @external_timeframe)
+    end
+
+    def within_internal_timeframe?(time)
+      within_timeframe?(time, @internal_timeframe)
+    end
+
+    def empty?
+      blank?(@cache_log) || (@cache_log[:internal].empty? && @cache_log[:external].empty?)
+    end
+
+    def size(type)
+      @cache_log[type].size
+    end
+
     private def construct_internal_link_metadata(metadata, found)
       {
         source: metadata[:source],
-        current_path: metadata[:current_path],
+        filename: metadata[:filename],
         line: metadata[:line],
         base_url: metadata[:base_url],
-        found: found
+        found: found,
       }
     end
 
     # prepare to add new URLs detected
     private def determine_additions(urls_detected, type)
-      additions = urls_detected.reject do |url, metadata|
-        if @cache_log[type].include?(url)
-          @cache_log[type][url][:metadata] = metadata
-
-          # if this is false, we're trying again
-          if type == :external
-            @cache_log[type][url][:found]
-          else
-            @cache_log[type][url][:metadata].none? { |m| m[:found] }
-          end
+      additions = type == :external ? determine_external_additions(urls_detected) : determine_internal_additions(urls_detected)
+
+      new_link_count = additions.length
+      new_link_text = pluralize(new_link_count, "new #{type} link", "new #{type} links")
+      @logger.log(:debug, "Adding #{new_link_text} to the cache")
+
+      additions
+    end
+
+    private def determine_external_additions(urls_detected)
+      urls_detected.reject do |url, _metadata|
+        if @cache_log[:external].include?(url)
+          @cache_log[:external][url][:found] # if this is false, we're trying again
         else
-          @logger.log :debug, "Adding #{url} to #{type} cache"
+          @logger.log(:debug, "Adding #{url} to external cache")
           false
         end
       end
+    end
 
-      new_link_count = additions.length
-      new_link_text = pluralize(new_link_count, "new #{type} link", "new #{type} links")
-      @logger.log :debug, "Adding #{new_link_text} to the cache"
+    private def determine_internal_additions(urls_detected)
+      urls_detected.each_with_object({}) do |(url, metadata), hsh|
+        # url is not even in cache
+        if @cache_log[:internal][url].nil?
+          hsh[url] = metadata
+          next
+        end
 
-      additions
+        cache_metadata = @cache_log[:internal][url][:metadata]
+        incoming_metadata = urls_detected[url].each_with_object([]) do |incoming_url, arr|
+          existing_cache_metadata = cache_metadata.find { |k, _| k[:filename] == incoming_url[:filename] }
+
+          # cache for this url, from an existing path, exists as found
+          if !existing_cache_metadata.nil? && !existing_cache_metadata.empty? && existing_cache_metadata[:found]
+            metadata.find { |m| m[:filename] == existing_cache_metadata[:filename] }[:found] = true
+            next
+          end
+
+          @logger.log(:debug, "Adding #{incoming_url} to internal cache")
+          arr << incoming_url
+        end
+
+        hsh[url] = incoming_metadata
+      end
     end
 
     # remove from cache URLs that no longer exist
@@ -125,54 +188,21 @@ module HTMLProofer
         if urls_detected.include?(url)
           false
         elsif url_matches_type?(url, type)
-          @logger.log :debug, "Removing #{url} from #{type} cache"
+          @logger.log(:debug, "Removing #{url} from #{type} cache")
           deletions += 1
           true
         end
       end
 
       del_link_text = pluralize(deletions, "outdated #{type} link", "outdated #{type} links")
-      @logger.log :debug, "Removing #{del_link_text} from the cache"
-    end
-
-    def write
-      return unless enabled?
-
-      File.write(@cache_file, @cache_log.to_json)
-    end
-
-    def retrieve_urls(urls_detected, type)
-      # if there are no urls, bail
-      return {} if urls_detected.empty?
-
-      urls_detected = urls_detected.transform_keys do |url|
-        cleaned_url(url)
-      end
-
-      urls_to_check = detect_url_changes(urls_detected, type)
-
-      @cache_log[type].each_pair do |url, cache|
-        next if within_timeframe?(cache[:time])
-
-        urls_to_check[url] = cache[:metadata] # recheck expired links
-      end
-
-      urls_to_check
-    end
-
-    def empty?
-      blank?(@cache_log) || (@cache_log[:internal].empty? && @cache_log[:external].empty?)
-    end
-
-    def size(type)
-      @cache_log[type].size
+      @logger.log(:debug, "Removing #{del_link_text} from the cache")
     end
 
     private def setup_cache!(options)
       default_structure = {
         version: CACHE_VERSION,
         internal: {},
-        external: {}
+        external: {},
       }
 
       @storage_dir = options[:storage_dir] || DEFAULT_STORAGE_DIR
@@ -193,26 +223,32 @@ module HTMLProofer
 
       old_cache = (cache_version = log[:version]).nil?
       @cache_log = if old_cache # previous cache version, create a new one
-                     default_structure
-                   elsif cache_version != CACHE_VERSION
-                   # if cache version is newer...do something
-                   else
-                     log[:internal] = log[:internal].transform_keys(&:to_s)
-                     log[:external] = log[:external].transform_keys(&:to_s)
-                     log
-                   end
+        default_structure
+      elsif cache_version != CACHE_VERSION
+      # if cache version is newer...do something
+      else
+        log[:internal] = log[:internal].transform_keys(&:to_s)
+        log[:external] = log[:external].transform_keys(&:to_s)
+        log
+      end
     end
 
+    # https://github.com/rails/rails/blob/3872bc0e54d32e8bf3a6299b0bfe173d94b072fc/activesupport/lib/active_support/duration.rb#L112-L117
+    SECONDS_PER_HOUR   = 3600
+    SECONDS_PER_DAY    = 86400
+    SECONDS_PER_WEEK   = 604800
+    SECONDS_PER_MONTH  = 2629746  # 1/12 of a gregorian year
+
     private def time_ago(measurement, unit)
       case unit
       when :months
-        @cache_datetime >> -measurement
+        @cache_datetime - (SECONDS_PER_MONTH * measurement)
       when :weeks
-        @cache_datetime - (measurement * 7)
+        @cache_datetime - (SECONDS_PER_WEEK * measurement)
       when :days
-        @cache_datetime - measurement
+        @cache_datetime - (SECONDS_PER_DAY * measurement)
       when :hours
-        @cache_datetime - Rational(measurement / 24.0)
+        @cache_datetime - Rational(SECONDS_PER_HOUR * measurement)
       end.to_time
     end
 
@@ -224,7 +260,7 @@ module HTMLProofer
     private def cleaned_url(url)
       cleaned_url = escape_unescape(url)
 
-      return cleaned_url unless cleaned_url.end_with?('/', '#', '?') && cleaned_url.length > 1
+      return cleaned_url unless cleaned_url.end_with?("/", "#", "?") && cleaned_url.length > 1
 
       cleaned_url[0..-2]
     end
@@ -232,5 +268,12 @@ module HTMLProofer
     private def escape_unescape(url)
       Addressable::URI.parse(url).normalize.to_s
     end
+
+    private def within_timeframe?(current_time, parsed_timeframe)
+      return false if current_time.nil? || parsed_timeframe.nil?
+
+      current_time = Time.parse(current_time) if current_time.is_a?(String)
+      (parsed_timeframe..@cache_time).cover?(current_time)
+    end
   end
 end

data/lib/html_proofer/check/favicon.rb

@@ -1,35 +1,40 @@
 # frozen_string_literal: true
 
-class HTMLProofer::Check::Favicon < HTMLProofer::Check
-  def run
-    found = false
-    @html.css('link').each do |node|
-      @favicon = create_element(node)
+module HTMLProofer
+  class Check
+    class Favicon < HTMLProofer::Check
+      def run
+        found = false
+        @html.css("link").each do |node|
+          @favicon = create_element(node)
 
-      next if @favicon.ignore?
+          next if @favicon.ignore?
 
-      break if (found = @favicon.node['rel'].split.last.eql? 'icon')
-    end
+          break if (found = @favicon.node["rel"].split.last.eql?("icon"))
+        end
 
-    return if immediate_redirect?
+        return if immediate_redirect?
 
-    if found
-      if @favicon.url.remote?
-        add_to_external_urls(@favicon.url, @favicon.line)
-      elsif !@favicon.url.exists?
-        add_failure("internal favicon #{@favicon.url.raw_attribute} does not exist", line: @favicon.line, content: @favicon.content)
+        if found
+          if @favicon.url.remote?
+            add_to_external_urls(@favicon.url, @favicon.line)
+          elsif !@favicon.url.exists?
+            add_failure("internal favicon #{@favicon.url.raw_attribute} does not exist", line: @favicon.line,
+              content: @favicon.content)
+          end
+        else
+          add_failure("no favicon provided")
+        end
       end
-    else
-      add_failure('no favicon provided')
-    end
-  end
 
-  private
+      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
+      # 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
   end
 end

data/lib/html_proofer/check/images.rb

@@ -1,62 +1,102 @@
 # frozen_string_literal: true
 
-class HTMLProofer::Check::Images < HTMLProofer::Check
-  SCREEN_SHOT_REGEX = /Screen(?: |%20)Shot(?: |%20)\d+-\d+-\d+(?: |%20)at(?: |%20)\d+.\d+.\d+/.freeze
-
-  def run
-    @html.css('img').each do |node|
-      @img = create_element(node)
-
-      next if @img.ignore?
-
-      # screenshot filenames should return because of terrible names
-      add_failure("image has a terrible filename (#{@img.url.raw_attribute})", line: @img.line, content: @img.content) if terrible_filename?
-
-      # does the image exist?
-      if missing_src?
-        add_failure('image has no src or srcset attribute', line: @img.line, content: @img.content)
-      elsif @img.url.remote?
-        add_to_external_urls(@img.url, @img.line)
-      elsif !@img.url.exists? && !@img.multiple_srcsets?
-        add_failure("internal image #{@img.url.raw_attribute} does not exist", line: @img.line, content: @img.content)
-      elsif @img.multiple_srcsets?
-        srcsets = @img.srcset.split(',').map(&:strip)
-        srcsets.each do |srcset|
-          srcset_url = HTMLProofer::Attribute::Url.new(@runner, srcset, base_url: @img.base_url)
-
-          if srcset_url.remote?
-            add_to_external_urls(srcset_url.url, @img.line)
-          elsif !srcset_url.exists?
-            add_failure("internal image #{srcset} does not exist", line: @img.line, content: @img.content)
+module HTMLProofer
+  class Check
+    class Images < HTMLProofer::Check
+      SCREEN_SHOT_REGEX = /Screen(?: |%20)Shot(?: |%20)\d+-\d+-\d+(?: |%20)at(?: |%20)\d+.\d+.\d+/.freeze
+
+      def run
+        @html.css("img").each do |node|
+          @img = create_element(node)
+
+          next if @img.ignore?
+
+          # screenshot filenames should return because of terrible names
+          add_failure("image has a terrible filename (#{@img.url.raw_attribute})", line: @img.line,
+            content: @img.content) if terrible_filename?
+
+          # does the image exist?
+          if missing_src?
+            add_failure("image has no src or srcset attribute", line: @img.line, content: @img.content)
+          elsif @img.url.remote?
+            add_to_external_urls(@img.url, @img.line)
+          elsif !@img.url.exists? && !@img.multiple_srcsets? && !@img.multiple_sizes?
+            add_failure("internal image #{@img.url.raw_attribute} does not exist", line: @img.line,
+              content: @img.content)
+          elsif @img.multiple_srcsets?
+            @img.srcsets.each do |srcset|
+              srcset_url = HTMLProofer::Attribute::Url.new(@runner, srcset, base_url: @img.base_url)
+
+              if srcset_url.remote?
+                add_to_external_urls(srcset_url.url, @img.line)
+              elsif !srcset_url.exists?
+                add_failure("internal image #{srcset} does not exist", line: @img.line, content: @img.content)
+              end
+            end
+          elsif @img.multiple_sizes?
+            @img.srcsets_wo_sizes.each do |srcset|
+              srcset_url = HTMLProofer::Attribute::Url.new(@runner, srcset, base_url: @img.base_url)
+
+              if srcset_url.remote?
+                add_to_external_urls(srcset_url.url, @img.line)
+              elsif !srcset_url.exists?
+                add_failure("internal image #{srcset} does not exist", line: @img.line, content: @img.content)
+              end
+            end
+          end
+
+          unless ignore_element?
+            if missing_alt_tag? && !ignore_missing_alt?
+              add_failure("image #{@img.url.raw_attribute} does not have an alt attribute", line: @img.line,
+                content: @img.content)
+            elsif (empty_alt_tag? || alt_all_spaces?) && !ignore_empty_alt?
+              add_failure("image #{@img.url.raw_attribute} has an alt attribute, but no content", line: @img.line,
+                content: @img.content)
+            end
           end
+
+          add_failure("image #{@img.url.raw_attribute} uses the http scheme", line: @img.line,
+            content: @img.content) if @runner.enforce_https? && @img.url.http?
         end
+
+        external_urls
       end
 
-      add_failure("image #{@img.url.raw_attribute} does not have an alt attribute", line: @img.line, content: @img.content) if empty_alt_tag? && !ignore_missing_alt? && !ignore_alt?
+      def ignore_missing_alt?
+        @runner.options[:ignore_missing_alt]
+      end
 
-      add_failure("image #{@img.url.raw_attribute} uses the http scheme", line: @img.line, content: @img.content) if @runner.enforce_https? && @img.url.http?
-    end
+      def ignore_empty_alt?
+        @runner.options[:ignore_empty_alt]
+      end
 
-    external_urls
-  end
+      def ignore_element?
+        @img.url.ignore? || @img.aria_hidden?
+      end
 
-  def ignore_missing_alt?
-    @runner.options[:ignore_missing_alt]
-  end
+      def missing_alt_tag?
+        @img.node["alt"].nil?
+      end
 
-  def ignore_alt?
-    @img.url.ignore? || @img.aria_hidden?
-  end
+      def empty_alt_tag?
+        !missing_alt_tag? && @img.node["alt"].empty?
+      end
 
-  def empty_alt_tag?
-    @img.node['alt'].nil? || @img.node['alt'].strip.empty?
-  end
+      def empty_whitespace_alt_tag?
+        !missing_alt_tag? && @img.node["alt"].strip.empty?
+      end
 
-  def terrible_filename?
-    @img.url.to_s =~ SCREEN_SHOT_REGEX
-  end
+      def alt_all_spaces?
+        !missing_alt_tag? && @img.node["alt"].split.all?(" ")
+      end
+
+      def terrible_filename?
+        @img.url.to_s =~ SCREEN_SHOT_REGEX
+      end
 
-  def missing_src?
-    blank?(@img.url.to_s)
+      def missing_src?
+        blank?(@img.url.to_s)
+      end
+    end
   end
 end