html-proofer 5.0.0 → 5.2.2

data/lib/html_proofer/configuration.rb

@@ -13,7 +13,7 @@ module HTMLProofer
       check_external_hash: true,
       check_internal_hash: true,
       checks: DEFAULT_TESTS,
-      directory_index_file: "index.html",
+      directory_index_files: ["index.html"],
       disable_external: false,
       ignore_empty_alt: true,
       ignore_empty_mailto: false,
@@ -47,6 +47,12 @@ module HTMLProofer
 
     class << self
       def generate_defaults(opts)
+        # If `:directory_index_file` (singular) is given, convert it into an
+        # array for `:directory_index_files` (plural) instead.
+        if opts.key?(:directory_index_file)
+          opts[:directory_index_files] = [opts.delete(:directory_index_file)]
+        end
+
         options = PROOFER_DEFAULTS.merge(opts)
 
         options[:typhoeus] = HTMLProofer::Configuration::TYPHOEUS_DEFAULTS.merge(opts[:typhoeus] || {})
@@ -89,6 +95,10 @@ module HTMLProofer
             @options[long_opt_symbol] = arg
           end
 
+          set_option(opts, "--directory-index-files [FILENAME1,FILENAME2,...]") do |long_opt_symbol, list|
+            @options[long_opt_symbol] = list.nil? ? [] : list.split(",")
+          end
+
           set_option(opts, "--extensions [EXT1,EXT2,...]") do |long_opt_symbol, list|
             @options[long_opt_symbol] = list.nil? ? [] : list.split(",")
           end
@@ -134,7 +144,17 @@ module HTMLProofer
 
         section(opts, "Ignore Configuration") do
           set_option(opts, "--ignore-files [FILE1,FILE2,...]") do |long_opt_symbol, list|
-            @options[long_opt_symbol] = list.nil? ? [] : list.split(",")
+            @options[long_opt_symbol] = if list.nil?
+              []
+            else
+              list.split(",").map.each do |l|
+                if l.start_with?("/") && l.end_with?("/")
+                  Regexp.new(l[1...-1])
+                else
+                  l
+                end
+              end
+            end
           end
 
           set_option(opts, "--[no-]ignore-empty-alt") do |long_opt_symbol, arg|
@@ -187,12 +207,12 @@ module HTMLProofer
             @options[long_opt_symbol] = parse_json_option("typhoeus", arg, symbolize_names: false)
           end
 
-          set_option(opts, "--hydra <CONFIG>") do |long_opt_symbol, _list|
-            @options[long_opt_symbol] = parse_json_option("hydra", arg, symbolize_names: false)
+          set_option(opts, "--hydra <CONFIG>") do |long_opt_symbol, arg|
+            @options[long_opt_symbol] = parse_json_option("hydra", arg, symbolize_names: true)
           end
 
-          set_option(opts, "--cache <CONFIG>") do |long_opt_symbol, _list|
-            @options[long_opt_symbol] = parse_json_option("cache", arg, symbolize_names: false)
+          set_option(opts, "--cache <CONFIG>") do |long_opt_symbol, arg|
+            @options[long_opt_symbol] = parse_json_option("cache", arg, symbolize_names: true)
           end
         end
 
@@ -223,8 +243,8 @@ module HTMLProofer
       arg.split(",").each_with_object({}) do |s, hsh|
         split = s.split(/(?<!\\):/, 2)
 
-        re = split[0].gsub(/\\:/, ":")
-        string = split[1].gsub(/\\:/, ":")
+        re = split[0].gsub("\\:", ":")
+        string = split[1].gsub("\\:", ":")
         hsh[Regexp.new(re)] = string
       end
     end
@@ -270,49 +290,70 @@ module HTMLProofer
     module ConfigurationHelp
       TEXT = {
         as_links: ["Assumes that `PATH` is a comma-separated array of links to check."],
-        assume_extension: ["Automatically add specified extension to files for internal links, ",
-                           "to allow extensionless URLs (as supported by most servers) (default: `.html`).",],
+        assume_extension: [
+          "Automatically add specified extension to files for internal links, ",
+          "to allow extensionless URLs (as supported by most servers) (default: `.html`).",
+        ],
         directory_index_file: ["Sets the file to look for when a link refers to a directory. (default: `index.html`)."],
-        extensions: ["A comma-separated list of Strings indicating the file extensions you",
-                     "would like to check (default: `.html`)",],
+        directory_index_files: ["Sets the files to look for when a link refers to a directory. (default: `[\"index.html\"]`)."],
+        extensions: [
+          "A comma-separated list of Strings indicating the file extensions you",
+          "would like to check (default: `.html`)",
+        ],
 
         allow_hash_href: ['"If `true`, assumes `href="#"` anchors are valid (default: `true`)"'],
-        allow_missing_href: ["If `true`, does not flag `a` tags missing `href`. In HTML5, this is technically ",
-                             "allowed, but could also be human error. (default: `false`)",],
-        checks: ["A comma-separated list of Strings indicating which checks you",
-                 "want to run (default: `[\"Links\", \"Images\", \"Scripts\"]",],
+        allow_missing_href: [
+          "If `true`, does not flag `a` tags missing `href`. In HTML5, this is technically ",
+          "allowed, but could also be human error. (default: `false`)",
+        ],
+        checks: [
+          "A comma-separated list of Strings indicating which checks you",
+          "want to run (default: `[\"Links\", \"Images\", \"Scripts\"]",
+        ],
         check_external_hash: ["Checks whether external hashes exist (even if the webpage exists) (default: `true`)."],
         check_internal_hash: ["Checks whether internal hashes exist (even if the webpage exists) (default: `true`)."],
         check_sri: ["Check that `<link>` and `<script>` external resources use SRI (default: `false`)."],
         disable_external: ["If `true`, does not run the external link checker (default: `false`)."],
-        enforce_https: ["Fails a link if it\'s not marked as `https` (default: `true`)."],
+        enforce_https: ["Fails a link if it's not marked as `https` (default: `true`)."],
         root_dir: ["The absolute path to the directory serving your html-files."],
 
-        ignore_empty_alt: ["If `true`, ignores images with empty/missing ",
-                           "alt tags (in other words, `<img alt>` and `<img alt=\"\">`",
-                           "are valid; set this to `false` to flag those) (default: `true`).",],
-        ignore_empty_mailto: ["If `true`, allows `mailto:` `href`s which don't",
-                              "contain an email address (default: `false`)'.",],
+        ignore_empty_alt: [
+          "If `true`, ignores images with empty/missing ",
+          "alt tags (in other words, `<img alt>` and `<img alt=\"\">`",
+          "are valid; set this to `false` to flag those) (default: `true`).",
+        ],
+        ignore_empty_mailto: [
+          "If `true`, allows `mailto:` `href`s which don't",
+          "contain an email address (default: `false`)'.",
+        ],
         ignore_missing_alt: ["If `true`, ignores images with missing alt tags (default: `false`)."],
         ignore_status_codes: ["A comma-separated list of numbers representing status codes to ignore."],
         ignore_files: ["A comma-separated list of Strings or RegExps containing file paths that are safe to ignore"],
-        ignore_urls: ["A comma-separated list of Strings or RegExps containing URLs that are",
-                      "safe to ignore. This affects all HTML attributes, such as `alt` tags on images.",],
+        ignore_urls: [
+          "A comma-separated list of Strings or RegExps containing URLs that are",
+          "safe to ignore. This affects all HTML attributes, such as `alt` tags on images.",
+        ],
         only_status_codes: ["A comma-separated list of numbers representing the only status codes to report on."],
         only_4xx: ["Only reports errors for links that fall within the 4xx status code range."],
 
-        swap_attributes: ["JSON-formatted config that maps element names to the",
-                          "preferred attribute to check (default: `{}`).",],
-        swap_urls: ["A comma-separated list containing key-value pairs of `RegExp => String`.",
-                    "It transforms URLs that match `RegExp` into `String` via `gsub`.",
-                    "The escape sequences `\\:` should be used to produce literal `:`s.",],
+        swap_attributes: [
+          "JSON-formatted config that maps element names to the",
+          "preferred attribute to check (default: `{}`).",
+        ],
+        swap_urls: [
+          "A comma-separated list containing key-value pairs of `RegExp => String`.",
+          "It transforms URLs that match `RegExp` into `String` via `gsub`.",
+          "The escape sequences `\\:` should be used to produce literal `:`s.",
+        ],
 
         typhoeus: ["JSON-formatted string of Typhoeus config; if set, overrides the html-proofer defaults."],
         hydra: ["JSON-formatted string of Hydra config; if set, overrides the html-proofer defaults."],
         cache: ["JSON-formatted string of cache config; if set, overrides the html-proofer defaults."],
 
-        log_level: ["Sets the logging level. One of `:debug`, `:info`, ",
-                    "`:warn`, `:error`, or `:fatal`. (default: `:info`)",],
+        log_level: [
+          "Sets the logging level. One of `:debug`, `:info`, ",
+          "`:warn`, `:error`, or `:fatal`. (default: `:info`)",
+        ],
 
         version: ["Prints the version of html-proofer."],
       }.freeze

data/lib/html_proofer/element.rb

@@ -16,7 +16,7 @@ module HTMLProofer
       swap_attributes!
 
       @base_url = base_url
-      @url = Attribute::Url.new(runner, link_attribute, base_url: base_url)
+      @url = Attribute::Url.new(runner, link_attribute, base_url: base_url, source: @runner.current_source, filename: @runner.current_filename)
 
       @line = node.line
       @content = node.content
@@ -27,7 +27,7 @@ module HTMLProofer
     end
 
     def meta_content
-      return nil unless meta_tag?
+      return unless meta_tag?
 
       @node["content"]
     end
@@ -37,7 +37,7 @@ module HTMLProofer
     end
 
     def src
-      return nil if !img_tag? && !script_tag? && !source_tag?
+      return if !img_tag? && !script_tag? && !source_tag?
 
       @node["src"]
     end
@@ -51,7 +51,7 @@ module HTMLProofer
     end
 
     def srcset
-      return nil if !img_tag? && !source_tag?
+      return if !img_tag? && !source_tag?
 
       @node["srcset"]
     end
@@ -61,7 +61,7 @@ module HTMLProofer
     end
 
     def href
-      return nil if !a_tag? && !link_tag?
+      return if !a_tag? && !link_tag?
 
       @node["href"]
     end
@@ -82,10 +82,25 @@ module HTMLProofer
       !blank?(srcset) && srcset.split(",").size > 1
     end
 
+    # From https://github.com/sindresorhus/srcset/blob/f7c48acd7facf18e94dec47e6b96e84e0f0e69dc/index.js#LL1-L16C71
+    # This regex represents a loose rule of an “image candidate string”; see https://html.spec.whatwg.org/multipage/images.html#srcset-attribute
+    # An “image candidate string” roughly consists of the following:
+    # 1. Zero or more whitespace characters.
+    # 2. A non-empty URL that does not start or end with `,`.
+    # 3. Zero or more whitespace characters.
+    # 4. An optional “descriptor” that starts with a whitespace character.
+    # 5. Zero or more whitespace characters.
+    # 6. Each image candidate string is separated by a `,`.
+    # We intentionally implement a loose rule here so that we can perform more aggressive error handling and reporting in the below code.
+
+    IMAGE_CANDIDATE_REGEX = /\s*([^,]\S*[^,](?:\s+[^,]+)?)\s*(?:,|$)/
+
     def srcsets
-      return nil if blank?(srcset)
+      return if blank?(srcset)
 
-      srcset.split(",").map(&:strip)
+      srcset.split(IMAGE_CANDIDATE_REGEX).select.with_index do |_part, idx|
+        idx.odd?
+      end.map(&:strip)
     end
 
     def multiple_sizes?
@@ -97,7 +112,7 @@ module HTMLProofer
     end
 
     def srcsets_wo_sizes
-      return nil if blank?(srcsets)
+      return if blank?(srcsets)
 
       srcsets.map do |srcset|
         srcset.split(" ").first
@@ -118,7 +133,7 @@ module HTMLProofer
 
       attrs = @runner.options[:swap_attributes][@node.name]
 
-      return true unless blank?(attrs)
+      true unless blank?(attrs)
     end
 
     private def swap_attributes!

data/lib/html_proofer/failure.rb

@@ -2,9 +2,9 @@
 
 module HTMLProofer
   class Failure
-    attr_reader :path, :check_name, :description, :status, :line, :content
+    attr_reader :path, :check_name, :description, :status, :line, :content, :element
 
-    def initialize(path, check_name, description, line: nil, status: nil, content: nil)
+    def initialize(path, check_name, description, line: nil, status: nil, content: nil, element: nil)
       @path = path
       @check_name = check_name
       @description = description
@@ -12,6 +12,7 @@ module HTMLProofer
       @line = line
       @status = status
       @content = content
+      @element = element
     end
   end
 end

data/lib/html_proofer/log.rb

@@ -11,9 +11,11 @@ module HTMLProofer
     STDERR_LEVELS = [:error, :fatal].freeze
 
     def initialize(log_level)
-      @logger = Yell.new(format: false, \
-        name: "HTMLProofer", \
-        level: "gte.#{log_level}") do |l|
+      @logger = Yell.new(
+        format: false,
+        name: "HTMLProofer",
+        level: "gte.#{log_level}",
+      ) do |l|
         l.adapter(:stdout, level: "lte.warn")
         l.adapter(:stderr, level: "gte.error")
       end
@@ -39,8 +41,8 @@ module HTMLProofer
         :red
       end
 
-      if (STDOUT_LEVELS.include?(level) && $stdout.isatty) || \
-          (STDERR_LEVELS.include?(level) && $stderr.isatty)
+      if STDOUT_LEVELS.include?(level) ||
+          STDERR_LEVELS.include?(level)
         Rainbow(message).send(color)
       else
         message

data/lib/html_proofer/reporter.rb

@@ -11,8 +11,8 @@ module HTMLProofer
     end
 
     def failures=(failures)
-      @failures = failures.group_by(&:check_name) \
-        .transform_values { |issues| issues.sort_by { |issue| [issue.path, issue.line] } } \
+      @failures = failures.group_by(&:check_name)
+        .transform_values { |issues| issues.sort_by { |issue| [issue.path, issue.line] } }
         .sort
     end
 

data/lib/html_proofer/runner.rb

@@ -6,8 +6,8 @@ module HTMLProofer
   class Runner
     include HTMLProofer::Utils
 
-    attr_reader :options, :cache, :logger, :internal_urls, :external_urls, :checked_paths, :current_check
-    attr_accessor :current_filename, :current_source, :reporter
+    attr_reader :options, :cache, :logger, :internal_urls, :external_urls, :resolved_paths, :current_check, :current_filename, :current_source
+    attr_accessor :reporter
 
     URL_TYPES = [:external, :internal].freeze
 
@@ -26,7 +26,7 @@ module HTMLProofer
 
       @before_request = []
 
-      @checked_paths = {}
+      @resolved_paths = {}
 
       @current_check = nil
       @current_source = nil
@@ -42,8 +42,10 @@ module HTMLProofer
         @logger.log(:info, "Running #{check_text} (#{format_checks_list(checks)}) on #{@source} ...\n\n")
         check_list_of_links unless @options[:disable_external]
       else
-        @logger.log(:info,
-          "Running #{check_text} (#{format_checks_list(checks)}) in #{@source} on *#{@options[:extensions].join(", ")} files ...\n\n")
+        @logger.log(
+          :info,
+          "Running #{check_text} (#{format_checks_list(checks)}) in #{@source} on *#{@options[:extensions].join(", ")} files ...\n\n",
+        )
 
         check_files
         @logger.log(:info, "Ran on #{pluralize(files.length, "file", "files")}!\n\n")
@@ -166,8 +168,14 @@ module HTMLProofer
 
     def ignore_file?(file)
       @options[:ignore_files].each do |pattern|
-        return true if pattern.is_a?(String) && pattern == file
-        return true if pattern.is_a?(Regexp) && pattern =~ file
+        if pattern.is_a?(String) && pattern == file
+          @logger.log(:debug, "Ignoring #{file} because it matches #{pattern}")
+          return true
+        end
+        next unless pattern.is_a?(Regexp) && pattern.match(file)
+
+        @logger.log(:debug, "Ignoring #{file} because it matches regexp #{pattern}")
+        return true
       end
 
       false
@@ -192,7 +200,7 @@ module HTMLProofer
     end
 
     def failed_checks
-      @reporter.failures.flatten.select { |f| f.is_a?(Failure) }
+      @reporter.failures.flatten.grep(Failure)
     end
 
     def report_failed_checks
@@ -228,7 +236,7 @@ module HTMLProofer
     end
 
     private def load_cache(type)
-      ivar = instance_variable_get("@#{type}_urls")
+      ivar = instance_variable_get(:"@#{type}_urls")
 
       existing_urls_count = @cache.size(type)
       cache_text = pluralize(existing_urls_count, "#{type} link", "#{type} links")
@@ -241,7 +249,7 @@ module HTMLProofer
 
     private def format_checks_list(checks)
       checks.map do |check|
-        check.sub(/HTMLProofer::Check::/, "")
+        check.sub("HTMLProofer::Check::", "")
       end.sort.join(", ")
     end
   end

data/lib/html_proofer/url_validator/external.rb

@@ -1,7 +1,9 @@
 # frozen_string_literal: true
 
 require "typhoeus"
-require "uri"
+require "open-uri"
+# require "uri"
+require "pdf-reader"
 
 module HTMLProofer
   class UrlValidator
@@ -88,8 +90,12 @@ module HTMLProofer
         return if @runner.options[:ignore_status_codes].include?(response_code)
 
         if response_code.between?(200, 299)
-          @cache.add_external(href, filenames, response_code, "OK", true) unless check_hash_in_2xx_response(href, url,
-            response, filenames)
+          @cache.add_external(href, filenames, response_code, "OK", true) unless check_hash_in_2xx_response(
+            href,
+            url,
+            response,
+            filenames,
+          )
         elsif response.timed_out?
           handle_timeout(href, filenames, response_code)
         elsif response_code.zero?
@@ -115,6 +121,31 @@ module HTMLProofer
         return false unless url.hash?
 
         hash = url.hash
+        # `response.headers` covers both live responses (which expose the headers
+        # via :response_headers) and stubbed ones (which set :headers directly);
+        # reading options[:headers] saw nothing but stubs.
+        headers = response.headers || {}
+        content_type = headers.find { |k, _| k.casecmp("content-type").zero? }
+
+        # attempt to verify PDF hash ref; see #787 for more details
+        # FIXME: this is re-reading the PDF response
+        if content_type && content_type[1].include?("pdf")
+          io = URI.parse(url.to_s).open
+          reader = PDF::Reader.new(io)
+
+          pages = reader.pages
+          if hash =~ /\Apage=(\d+)\z/
+            page = Regexp.last_match[1].to_i
+
+            unless pages[page - 1]
+              msg = "External link #{href} failed: #{url.without_hash} exists, but the hash '#{hash}' does not"
+              add_failure(filenames, msg, response.code)
+              @cache.add_external(href, filenames, response.code, msg, false)
+            end
+
+            return true
+          end
+        end
 
         body_doc = create_nokogiri(response.body)
 
@@ -130,7 +161,7 @@ module HTMLProofer
 
         return unless body_doc.xpath(xpath.join("|")).empty?
 
-        msg = "External link #{href} failed: #{url.sans_hash} exists, but the hash '#{hash}' does not"
+        msg = "External link #{href} failed: #{url.without_hash} exists, but the hash '#{hash}' does not"
         add_failure(filenames, msg, response.code)
         @cache.add_external(href, filenames, response.code, msg, false)
         true
@@ -167,7 +198,15 @@ module HTMLProofer
           @failed_checks << Failure.new("", "Links > External", description, status: status)
         else
           metadata.each do |m|
-            @failed_checks << Failure.new(m[:filename], "Links > External", description, line: m[:line], status: status)
+            @failed_checks << Failure.new(
+              m[:filename],
+              "Links > External",
+              description,
+              line: m[:line],
+              status: status,
+              content: m[:element]&.content,
+              element: m[:element],
+            )
           end
         end
       end

data/lib/html_proofer/url_validator/internal.rb

@@ -29,15 +29,18 @@ module HTMLProofer
           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]
-
-            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)
+            url = HTMLProofer::Attribute::Url.new(@runner, link, base_url: metadata[:base_url], source: metadata[:source], filename: metadata[:filename])
+
+            unless url.exists?
+              @failed_checks << Failure.new(
+                metadata[:filename],
+                "Links > Internal",
+                "internally linking to #{url}, which does not exist",
+                line: metadata[:line],
+                status: nil,
+                content: metadata[:element]&.content,
+                element: metadata[:element],
+              )
               to_add << [url, metadata, false]
               next
             end
@@ -45,6 +48,7 @@ module HTMLProofer
             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
+              target_file_path = url.resolved_path
               unless file_paths_hashes_to_check.key?(target_file_path)
                 file_paths_hashes_to_check[target_file_path] = {}
               end
@@ -55,8 +59,15 @@ module HTMLProofer
               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)
+              @failed_checks << Failure.new(
+                metadata[:filename],
+                "Links > Internal",
+                "internally linking to #{url}; the file exists, but the hash '#{url.hash}' does not",
+                line: metadata[:line],
+                status: nil,
+                content: metadata[:element]&.content,
+                element: metadata[:element],
+              )
               to_add << [url, metadata, false]
               next
             end
@@ -75,8 +86,15 @@ module HTMLProofer
             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)
+                @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: metadata[:element]&.content,
+                  element: metadata[:element],
+                )
               end
               to_add << [url, metadata, exists]
             end
@@ -91,12 +109,6 @@ module HTMLProofer
         @failed_checks
       end
 
-      private def file_exists?(absolute_path)
-        return @runner.checked_paths[absolute_path] if @runner.checked_paths.key?(absolute_path)
-
-        @runner.checked_paths[absolute_path] = File.exist?(absolute_path)
-      end
-
       # 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)
@@ -104,16 +116,17 @@ module HTMLProofer
         return true if blank?(href_hash)
         return true unless @runner.options[:check_internal_hash]
 
-        # prevents searching files we didn't ask about
-        return false unless url.known_extension?
+        # For hash links, we need to defer to file-based checking
+        # (even if URL doesn't have extension, it may resolve to index.html)
         return false unless url.has_hash?
 
+        # If URL has no known extension, defer to file-based checking
+        return false unless url.known_extension?
+
         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
-        return true if fragment_ids.include?("top")
-
-        nil
+        true if fragment_ids.include?("top")
       end
 
       private def hash_exists_in_html?(href_hash, html)
@@ -124,10 +137,18 @@ module HTMLProofer
 
       private def find_fragments(fragment_ids, html)
         xpaths = fragment_ids.uniq.flat_map do |frag_id|
-          escaped_frag_id = "'#{frag_id.split("'").join("', \"'\", '")}', ''"
+          # Build XPath string argument, handling single quotes
+          if frag_id.include?("'")
+            # Use concat() to handle single quotes: concat('part1', "'", 'part2')
+            escaped_frag_id = frag_id.split("'").map { |part| "'#{part}'" }.join(", \"'\", ")
+            xpath_arg = "concat(#{escaped_frag_id})"
+          else
+            # No single quotes, just use quoted string
+            xpath_arg = "'#{frag_id}'"
+          end
           [
-            "//*[case_sensitive_equals(@id, concat(#{escaped_frag_id}))]",
-            "//*[case_sensitive_equals(@name, concat(#{escaped_frag_id}))]",
+            "//*[case_sensitive_equals(@id, #{xpath_arg})]",
+            "//*[case_sensitive_equals(@name, #{xpath_arg})]",
           ]
         end
         xpaths << XpathFunctions.new

data/lib/html_proofer/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module HTMLProofer
-  VERSION = "5.0.0"
+  VERSION = "5.2.2"
 end

data/lib/html_proofer/xpath_functions.rb

@@ -4,7 +4,7 @@ module HTMLProofer
   # https://stackoverflow.com/a/8812293
   class XpathFunctions
     def case_sensitive_equals(node_set, str_to_match)
-      node_set.find_all { |node| node.to_s.== str_to_match.to_s }
+      node_set.find_all { |node| node.to_s == str_to_match.to_s }
     end
   end
 end

data/lib/html_proofer.rb

@@ -15,8 +15,11 @@ require "html_proofer/version"
 require "fileutils"
 
 if ENV.fetch("DEBUG", false)
-  require "awesome_print"
   require "debug"
+  begin
+    require "amazing_print"
+  rescue LoadError
+  end
 end
 
 module HTMLProofer