Added an html validation check by looking at Nokogiri errors

README.md

@@ -122,6 +122,11 @@ Project | Repository
 * Whether your internal script references are not broken
 * Whether external scripts are loading
 
+### HTML
+
+Nokogiri looks at the markup and [provides errors](http://www.nokogiri.org/tutorials/ensuring_well_formed_markup.html) when parsing your document.
+This is an optional feature, set the `validate_html` option to enable validation errors from Nokogiri.
+
 ## Configuration
 
 The `HTML::Proofer` constructor takes an optional hash of additional options:
@@ -138,6 +143,7 @@ The `HTML::Proofer` constructor takes an optional hash of additional options:
 | `href_swap` | A hash containing key-value pairs of `RegExp => String`. It transforms links that match `RegExp` into `String` via `gsub`. | `{}` |
 | `verbose` | If `true`, outputs extra information as the checking happens. Useful for debugging. | `false` |
 | `only_4xx` | Only reports errors for links that fall within the 4xx status code range. | `false` |
+| `validate_html` | Enables HTML validation errors from Nokogiri | `false` |
 
 ### Configuring Typhoeus
 

bin/htmlproof

@@ -24,6 +24,7 @@ Mercenary.program(:htmlproof) do |p|
   p.option 'only-4xx', '--only-4xx', 'Only reports errors for links that fall within the 4x status code range.'
   p.option 'verbose', '--verbose', 'Enables more verbose logging.'
   p.option 'directory_index_file', '--directory_index_file', 'Sets the file to look for when a link refers to a directory.'
+  p.option 'validate_html', '--validate_html', 'Enables HTML validation errors from Nokogiri (default: `false`).'
 
   p.action do |args, opts|
     args = ["."] if args.empty?
@@ -45,6 +46,7 @@ Mercenary.program(:htmlproof) do |p|
     options[:favicon] = opts["favicon"] unless opts["favicon"].nil?
     options[:verbose] = opts["verbose"] unless opts["verbose"].nil?
     options[:directory_index_file] = opts["directory_index_file"] unless opts["directory_index_file"].nil?
+    options[:validate_html] = opts["validate_html"] unless opts["validate_html"].nil?
 
     path = path.delete(' ').split(",") if opts["as-links"]
 

lib/html/proofer.rb

@@ -41,7 +41,8 @@ def initialize(src, opts={})
         :disable_external => false,
         :verbose => false,
         :only_4xx => false,
-        :directory_index_file => "index.html"
+        :directory_index_file => "index.html",
+        :validate_html => false
       }
 
       @typhoeus_opts = {
@@ -191,6 +192,7 @@ def self.create_nokogiri(path)
     def get_checks
       checks = HTML::Proofer::Checks::Check.subclasses.map { |c| c.name }
       checks.delete("Favicons") unless @options[:favicon]
+      checks.delete("Html") unless @options[:validate_html]
       checks
     end
 

lib/html/proofer/checks.rb

@@ -7,7 +7,8 @@ class Checks
         "checks/images",
         "checks/links",
         "checks/scripts",
-        "checks/favicon"
+        "checks/favicon",
+        "checks/html"
       ].each { |r| require File.join(File.dirname(__FILE__), r) }
     end
   end

lib/html/proofer/checks/html.rb

@@ -0,0 +1,24 @@
+# encoding: utf-8
+
+class Html < ::HTML::Proofer::Checks::Check
+
+  # new html5 tags (source: http://www.w3schools.com/html/html5_new_elements.asp)
+  HTML5_TAGS = %w(article aside bdi details dialog figcaption 
+                  figure footer header main mark menuitem meter 
+                  nav progress rp rt ruby section summary 
+                  time wbr datalist keygen output color date 
+                  datetime datetime-local email month number 
+                  range search tel time url week canvas 
+                  svg audio embed source track video)
+
+  def run
+    @html.errors.each do |e|
+
+        # Nokogiri (or rather libxml2 underhood) only recognizes html4 tags,
+        # so we need to skip errors caused by the new tags in html5
+        next if HTML5_TAGS.include? e.to_s[/Tag ([\w-]+) invalid/o, 1]
+
+        self.add_issue(e.to_s)
+    end
+  end
+end

spec/html/proofer/fixtures/html/div_inside_head.html

@@ -0,0 +1,6 @@
+<html>
+<head>
+<div>
+</div>
+</head>
+</html>

spec/html/proofer/fixtures/html/invalid_tag.html

@@ -0,0 +1,3 @@
+<html>
+<myfancytag></myfancytag>
+</html>

spec/html/proofer/fixtures/html/missing_closing_quotes.html

@@ -0,0 +1,5 @@
+<html>
+<body>
+<a href="/test>A test link</a>
+</body>
+</html>

spec/html/proofer/fixtures/html/opening_and_ending_tag_mismatch.html

@@ -0,0 +1,7 @@
+<html>
+<body>
+<p>The quick <strong>brown fox</p>
+jumped over the
+<p>lazy</strong> dog.</p>
+</body>
+</html>

spec/html/proofer/fixtures/html/unescaped_ampersand_in_attribute.html

@@ -0,0 +1,4 @@
+<html>
+<body data-something="up&down">
+</body>
+</html>

spec/html/proofer/fixtures/html/unmatched_end_tag.html

@@ -0,0 +1,5 @@
+<html>
+<body>
+</div>
+</body>
+</html>

spec/html/proofer/html_spec.rb

@@ -0,0 +1,45 @@
+require "spec_helper"
+
+describe "Html test" do
+  it "ignores an invalid tag by default" do
+    html = "#{FIXTURES_DIR}/html/invalid_tag.html"
+    output = capture_stderr { HTML::Proofer.new(html).run }
+    output.should == ""
+  end
+
+  it "fails for an invalid tag" do
+    html = "#{FIXTURES_DIR}/html/invalid_tag.html"
+    output = capture_stderr { HTML::Proofer.new(html, {:validate_html => true}).run }
+    output.should match /Tag myfancytag invalid/
+  end
+
+  it "fails for an unmatched end tag" do
+    html = "#{FIXTURES_DIR}/html/unmatched_end_tag.html"
+    output = capture_stderr { HTML::Proofer.new(html, {:validate_html => true}).run }
+    output.should match /Unexpected end tag : div/
+  end
+
+  it "fails for an unescaped ampersand in attribute" do
+    html = "#{FIXTURES_DIR}/html/unescaped_ampersand_in_attribute.html"
+    output = capture_stderr { HTML::Proofer.new(html, {:validate_html => true}).run }
+    output.should match /htmlParseEntityRef: expecting ';'/
+  end
+
+  it "fails for mismatch between opening and ending tag" do
+    html = "#{FIXTURES_DIR}/html/opening_and_ending_tag_mismatch.html"
+    output = capture_stderr { HTML::Proofer.new(html, {:validate_html => true}).run }
+    output.should match /Opening and ending tag mismatch: p and strong/
+  end
+
+  it "fails for div inside head" do
+    html = "#{FIXTURES_DIR}/html/div_inside_head.html"
+    output = capture_stderr { HTML::Proofer.new(html, {:validate_html => true}).run }
+    output.should match /Unexpected end tag : head/
+  end
+
+  it "fails for missing closing quotation mark in href" do
+    html = "#{FIXTURES_DIR}/html/missing_closing_quotes.html"
+    output = capture_stderr { HTML::Proofer.new(html, {:validate_html => true}).run }
+    output.should match /Couldn't find end of Start Tag a/
+  end
+end