html-proofer 4.3.2 → 4.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f917378eca9c4bf25268ef711b047db8762a6eda7dd613bc75b66520f67cee0e
-  data.tar.gz: 909509bebddb27a6857775642123f0cc2f97d8bf3982ae262e6ba70497f255a4
+  metadata.gz: 4043366fa1c0d7aaa58fc0b577d25d2d0fba7e72afca2b43ae688e90c55dddef
+  data.tar.gz: a93621d2169757d3665e5950e8c99d23a1b64cc9014fbc806a986fcdcb906e26
 SHA512:
-  metadata.gz: 84f7ea7cf3eb93b39c0882fa8694027dce632b6e47a8e9fb194dbb901c64b14e82d33ff036f31f7786ab782671e9861c8a991688af7a2cbf5e67c50c591f5d91
-  data.tar.gz: feb87b5fc56c88b44fc0924e1e7f3e3a9d96a580920cd07080b2633bc75c3fb99543fdf449e49b5975c3292ff250c39f029118141cf60bd72b0cd4ba01d07723
+  metadata.gz: 1636df520fa7d4035ca482eeafd0b5e059b0fc77a0219f340e7c47b8f2105c8dd107bf03e0d78ca582f5d7abef9d5853edae498a2133b75eab6f92b7e77b636f
+  data.tar.gz: 5310df2250a5db3c8864b12ae01e49a0a2d043fc15f5bece45615cbad1d2b9100c899fa35902db5536b2be85908f9208ebfe65e31318cf6a6db47e945a745f17

data/lib/html_proofer/attribute/url.rb

@@ -20,12 +20,13 @@ module HTMLProofer
 
           swap_urls!
           clean_url!
-
-          # convert "//" links to "https://"
-          @url.start_with?("//") ? @url = "https:#{@url}" : @url
         end
       end
 
+      def protocol_relative?
+        url.start_with?("//")
+      end
+
       def to_s
         @url
       end

data/lib/html_proofer/check/favicon.rb

@@ -16,7 +16,10 @@ module HTMLProofer
         return if immediate_redirect?
 
         if found
-          if @favicon.url.remote?
+          if @favicon.url.protocol_relative?
+            add_failure("favicon link #{@favicon.url} is a protocol-relative URL, use explicit https:// instead",
+              line: @favicon.line, content: @favicon.content)
+          elsif @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,

data/lib/html_proofer/check/images.rb

@@ -18,6 +18,9 @@ module HTMLProofer
           # 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.protocol_relative?
+            add_failure("image link #{@img.url} is a protocol-relative URL, use explicit https:// instead",
+              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?
@@ -27,7 +30,10 @@ module HTMLProofer
             @img.srcsets_wo_sizes.each do |srcset|
               srcset_url = HTMLProofer::Attribute::Url.new(@runner, srcset, base_url: @img.base_url, extract_size: true)
 
-              if srcset_url.remote?
+              if srcset_url.protocol_relative?
+                add_failure("image link #{srcset_url.url} is a protocol-relative URL, use explicit https:// instead",
+                  line: @img.line, content: @img.content)
+              elsif 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)

data/lib/html_proofer/check/links.rb

@@ -28,6 +28,12 @@ module HTMLProofer
             next
           end
 
+          if @link.url.protocol_relative?
+            add_failure("#{@link.url} is a protocol-relative URL, use explicit https:// instead",
+              line: @link.line, content: @link.content)
+            next
+          end
+
           check_schemes
 
           # intentionally down here because we still want valid? & missing_href? to execute

data/lib/html_proofer/check/open_graph.rb

@@ -16,6 +16,9 @@ module HTMLProofer
             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.protocol_relative?
+            add_failure("open graph link #{@open_graph.url} is a protocol-relative URL, use explicit https:// instead",
+              line: @open_graph.line, content: @open_graph.content)
           elsif @open_graph.url.remote?
             add_to_external_urls(@open_graph.url, @open_graph.line)
           else

data/lib/html_proofer/check/scripts.rb

@@ -13,8 +13,11 @@ module HTMLProofer
           # 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.protocol_relative?
+            add_failure("script link #{@script.url} is a protocol-relative URL, use explicit https:// instead",
+              line: @script.line, content: @script.content)
           elsif @script.url.remote?
-            add_to_external_urls(@script.src, @script.line)
+            add_to_external_urls(@script.url, @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,

data/lib/html_proofer/check.rb

@@ -29,28 +29,10 @@ module HTMLProofer
         content: content)
     end
 
-    def self.subchecks(runner_options)
-      # grab all known checks
-      checks = ObjectSpace.each_object(Class).select do |klass|
-        klass < self
-      end
-
-      # remove any checks not explicitly included
-      checks.each_with_object([]) do |check, arr|
-        next unless runner_options[:checks].include?(check.short_name)
-
-        arr << check
-      end
-    end
-
     def short_name
       self.class.name.split("::").last
     end
 
-    def self.short_name
-      name.split("::").last
-    end
-
     def add_to_internal_urls(url, line)
       url_string = url.raw_attribute
 
@@ -74,6 +56,26 @@ module HTMLProofer
       @external_urls[url_string] << { filename: @runner.current_filename, line: line }
     end
 
+    class << self
+      def subchecks(runner_options)
+        # grab all known checks
+        checks = ObjectSpace.each_object(Class).select do |klass|
+          klass < self
+        end
+
+        # remove any checks not explicitly included
+        checks.each_with_object([]) do |check, arr|
+          next unless runner_options[:checks].include?(check.short_name)
+
+          arr << check
+        end
+      end
+
+      def short_name
+        name.split("::").last
+      end
+    end
+
     private def base_url
       return @base_url if defined?(@base_url)
 

data/lib/html_proofer/configuration.rb

@@ -47,42 +47,44 @@ module HTMLProofer
 
     CACHE_DEFAULTS = {}.freeze
 
-    def self.generate_defaults(opts)
-      options = PROOFER_DEFAULTS.merge(opts)
+    class << self
+      def generate_defaults(opts)
+        options = PROOFER_DEFAULTS.merge(opts)
 
-      options[:typhoeus] = HTMLProofer::Configuration::TYPHOEUS_DEFAULTS.merge(opts[:typhoeus] || {})
-      options[:hydra] = HTMLProofer::Configuration::HYDRA_DEFAULTS.merge(opts[:hydra] || {})
+        options[:typhoeus] = HTMLProofer::Configuration::TYPHOEUS_DEFAULTS.merge(opts[:typhoeus] || {})
+        options[:hydra] = HTMLProofer::Configuration::HYDRA_DEFAULTS.merge(opts[:hydra] || {})
 
-      options[:parallel] = HTMLProofer::Configuration::PARALLEL_DEFAULTS.merge(opts[:parallel] || {})
-      options[:cache] = HTMLProofer::Configuration::CACHE_DEFAULTS.merge(opts[:cache] || {})
+        options[:parallel] = HTMLProofer::Configuration::PARALLEL_DEFAULTS.merge(opts[:parallel] || {})
+        options[:cache] = HTMLProofer::Configuration::CACHE_DEFAULTS.merge(opts[:cache] || {})
 
-      options.delete(:src)
+        options.delete(:src)
 
-      options
-    end
+        options
+      end
 
-    def self.to_regex?(item)
-      if item.start_with?("/") && item.end_with?("/")
-        Regexp.new(item[1...-1])
-      else
-        item
+      def to_regex?(item)
+        if item.start_with?("/") && item.end_with?("/")
+          Regexp.new(item[1...-1])
+        else
+          item
+        end
       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?
+      def 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?
 
-      return {} if config.nil?
+        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?
+        return {} if config.strip.empty?
 
-      begin
-        JSON.parse(config, { symbolize_names: symbolize_names })
-      rescue StandardError
-        raise ArgumentError, "Option '#{option_name} did not contain valid JSON."
+        begin
+          JSON.parse(config, { symbolize_names: symbolize_names })
+        rescue StandardError
+          raise ArgumentError, "Option '#{option_name} did not contain valid JSON."
+        end
       end
     end
   end

data/lib/html_proofer/url_validator/internal.rb

@@ -22,22 +22,39 @@ module HTMLProofer
       end
 
       def run_internal_link_checker(links)
+        # collect urls and metadata for hashes to be checked in the same target file
+        file_paths_hashes_to_check = {}
         to_add = []
-        links.each_pair do |link, matched_files|
+        links.each_with_index do |(link, matched_files), i|
+          matched_count_to_log = pluralize(matched_files.count, "reference", "references")
+          @logger.log(:debug, "(#{i + 1} / #{links.count}) Internal link #{link}: Checking #{matched_count_to_log}")
           matched_files.each do |metadata|
             url = HTMLProofer::Attribute::Url.new(@runner, link, base_url: metadata[:base_url])
 
             @runner.current_source = metadata[:source]
             @runner.current_filename = metadata[:filename]
 
-            unless file_exists?(url)
+            target_file_path = url.absolute_path
+            unless file_exists?(target_file_path)
               @failed_checks << Failure.new(@runner.current_filename, "Links > Internal",
                 "internally linking to #{url}, which does not exist", line: metadata[:line], status: nil, content: nil)
               to_add << [url, metadata, false]
               next
             end
 
-            unless hash_exists?(url)
+            hash_exists = hash_exists_for_url?(url)
+            if hash_exists.nil?
+              # the hash needs to be checked in the target file, we collect the url and metadata
+              unless file_paths_hashes_to_check.key?(target_file_path)
+                file_paths_hashes_to_check[target_file_path] = {}
+              end
+              unless file_paths_hashes_to_check[target_file_path].key?(url.hash)
+                file_paths_hashes_to_check[target_file_path][url.hash] = []
+              end
+              file_paths_hashes_to_check[target_file_path][url.hash] << [url, metadata]
+              next
+            end
+            unless hash_exists
               @failed_checks << Failure.new(@runner.current_filename, "Links > Internal",
                 "internally linking to #{url}; the file exists, but the hash '#{url.hash}' does not", line: metadata[:line], status: nil, content: nil)
               to_add << [url, metadata, false]
@@ -48,6 +65,24 @@ module HTMLProofer
           end
         end
 
+        # check hashes by target file
+        @logger.log(:info, "Checking internal link hashes in #{pluralize(file_paths_hashes_to_check.count, "file", "files")}")
+        file_paths_hashes_to_check.each_with_index do |(file_path, hashes_to_check), i|
+          hash_count_to_log = pluralize(hashes_to_check.count, "hash", "hashes")
+          @logger.log(:debug, "(#{i + 1} / #{file_paths_hashes_to_check.count}) Checking #{hash_count_to_log} in #{file_path}")
+          html = create_nokogiri(file_path)
+          hashes_to_check.each_pair do |href_hash, url_metadata|
+            exists = hash_exists_in_html?(href_hash, html)
+            url_metadata.each do |(url, metadata)|
+              unless exists
+                @failed_checks << Failure.new(metadata[:filename], "Links > Internal",
+                  "internally linking to #{url}; the file exists, but the hash '#{href_hash}' does not", line: metadata[:line], status: nil, content: nil)
+              end
+              to_add << [url, metadata, exists]
+            end
+          end
+        end
+
         # adding directly to the cache above results in an endless loop
         to_add.each do |(url, metadata, exists)|
           @cache.add_internal(url.to_s, metadata, exists)
@@ -56,15 +91,15 @@ module HTMLProofer
         @failed_checks
       end
 
-      private def file_exists?(url)
-        absolute_path = url.absolute_path
-        return @runner.checked_paths[url.absolute_path] if @runner.checked_paths.key?(absolute_path)
+      private def file_exists?(absolute_path)
+        return @runner.checked_paths[absolute_path] if @runner.checked_paths.key?(absolute_path)
 
-        @runner.checked_paths[url.absolute_path] = File.exist?(absolute_path)
+        @runner.checked_paths[absolute_path] = File.exist?(absolute_path)
       end
 
-      # verify the target hash
-      private def hash_exists?(url)
+      # verify the hash w/o just based on the URL, w/o looking at the target file
+      # => returns nil if the has could not be verified
+      private def hash_exists_for_url?(url)
         href_hash = url.hash
         return true if blank?(href_hash)
         return true unless @runner.options[:check_internal_hash]
@@ -76,10 +111,18 @@ module HTMLProofer
         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(fragment_ids, url).empty?
+        return true if fragment_ids.include?("top")
+
+        nil
+      end
+
+      private def hash_exists_in_html?(href_hash, html)
+        decoded_href_hash = Addressable::URI.unescape(href_hash)
+        fragment_ids = [href_hash, decoded_href_hash]
+        !find_fragments(fragment_ids, html).empty?
       end
 
-      private def find_fragments(fragment_ids, url)
+      private def find_fragments(fragment_ids, html)
         xpaths = fragment_ids.uniq.flat_map do |frag_id|
           escaped_frag_id = "'#{frag_id.split("'").join("', \"'\", '")}', ''"
           [
@@ -89,7 +132,6 @@ module HTMLProofer
         end
         xpaths << XpathFunctions.new
 
-        html = create_nokogiri(url.absolute_path)
         html.xpath(*xpaths)
       end
     end

data/lib/html_proofer/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module HTMLProofer
-  VERSION = "4.3.2"
+  VERSION = "4.4.1"
 end

data/lib/html_proofer.rb

@@ -4,7 +4,7 @@ require "zeitwerk"
 lib_dir = File.join(File.dirname(__dir__), "lib")
 gem_loader = Zeitwerk::Loader.for_gem
 gem_loader.inflector.inflect(
-  "html_proofer" => "HTMLProofer"
+  "html_proofer" => "HTMLProofer",
 )
 gem_loader.ignore(File.join(lib_dir, "html-proofer.rb"))
 gem_loader.setup
@@ -20,37 +20,39 @@ if ENV.fetch("DEBUG", false)
 end
 
 module HTMLProofer
-  def self.check_file(file, options = {})
-    raise ArgumentError unless file.is_a?(String)
-    raise ArgumentError, "#{file} does not exist" unless File.exist?(file)
+  class << self
+    def check_file(file, options = {})
+      raise ArgumentError unless file.is_a?(String)
+      raise ArgumentError, "#{file} does not exist" unless File.exist?(file)
 
-    options[:type] = :file
-    HTMLProofer::Runner.new(file, options)
-  end
+      options[:type] = :file
+      HTMLProofer::Runner.new(file, options)
+    end
 
-  def self.check_directory(directory, options = {})
-    raise ArgumentError unless directory.is_a?(String)
-    raise ArgumentError, "#{directory} does not exist" unless Dir.exist?(directory)
+    def check_directory(directory, options = {})
+      raise ArgumentError unless directory.is_a?(String)
+      raise ArgumentError, "#{directory} does not exist" unless Dir.exist?(directory)
 
-    options[:type] = :directory
-    HTMLProofer::Runner.new([directory], options)
-  end
+      options[:type] = :directory
+      HTMLProofer::Runner.new([directory], options)
+    end
 
-  def self.check_directories(directories, options = {})
-    raise ArgumentError unless directories.is_a?(Array)
+    def check_directories(directories, options = {})
+      raise ArgumentError unless directories.is_a?(Array)
 
-    options[:type] = :directory
-    directories.each do |directory|
-      raise ArgumentError, "#{directory} does not exist" unless Dir.exist?(directory)
+      options[:type] = :directory
+      directories.each do |directory|
+        raise ArgumentError, "#{directory} does not exist" unless Dir.exist?(directory)
+      end
+      HTMLProofer::Runner.new(directories, options)
     end
-    HTMLProofer::Runner.new(directories, options)
-  end
 
-  def self.check_links(links, options = {})
-    raise ArgumentError unless links.is_a?(Array)
+    def check_links(links, options = {})
+      raise ArgumentError unless links.is_a?(Array)
 
-    options[:type] = :links
-    HTMLProofer::Runner.new(links, options)
+      options[:type] = :links
+      HTMLProofer::Runner.new(links, options)
+    end
   end
 end
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: html-proofer
 version: !ruby/object:Gem::Version
-  version: 4.3.2
+  version: 4.4.1
 platform: ruby
 authors:
 - Garen Torikian
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-08-03 00:00:00.000000000 Z
+date: 2022-09-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: addressable