link-checker 0.5.2 → 0.6.0

data/VERSION

@@ -1 +1 @@
-0.5.2
+0.6.0

data/bin/check-links

@@ -6,6 +6,7 @@ require 'trollop'
 options = Trollop::options do
 	opt :no_warnings, "Don't warn about redirects to valid links"
 	opt :warnings_are_errors, "Treat any warning as an error and produce an error return code"
+	opt :max_threads, "The maximum number of threads to start.", :default => 100
 end
 
 exit LinkChecker.new(:options => options, :target => ARGV[0]).check_uris

data/lib/link_checker.rb

@@ -9,9 +9,16 @@ require 'anemone'
 class LinkChecker
 
   def initialize(params)
-    @options = params[:options] || {}
+    @options = params[:options] || { }
     @target =  params[:target] || './'
+
+    @html_files = []
+    @links = []
+    @errors = []
+    @warnings = []
     @return_code = 0
+
+    @options[:max_threads] ||= 100 # Only happens in testing.
   end
 
   def html_file_paths
@@ -43,7 +50,7 @@ class LinkChecker
         when Net::HTTPRedirection then
           return self.check_uri(URI(response['location']), true)
         else
-          return Error.new(:uri_string => uri.to_s, :response => response)
+          return Error.new(:uri_string => uri.to_s, :error => response)
         end
       end
     end
@@ -59,6 +66,21 @@ class LinkChecker
     rescue => error
       puts "Error: #{error.to_s}".red
     end
+
+    # Report the final results.
+    unless @html_files.empty?
+      file_pluralized = (@html_files.size.eql? 1) ? 'file' : 'files'
+      link_pluralized = (@links.size.eql? 1) ? 'link' : 'links'
+      if @errors.empty?
+        puts ("Checked #{@links.size} #{link_pluralized} in #{@html_files.size} " +
+          "HTML #{file_pluralized} and found no errors.").green
+      else
+        error_pluralized = (@errors.size.eql? 1) ? 'error' : 'errors'
+        puts ("Checked #{@links.size} #{link_pluralized} in #{@html_files.size} " +
+          "HTML #{file_pluralized} and found #{@errors.size} #{error_pluralized}.").red
+      end
+    end
+
     @return_code
   end
 
@@ -69,6 +91,7 @@ class LinkChecker
       anemone.on_every_page do |crawled_page|
         raise StandardError.new(crawled_page.error) if crawled_page.error
         threads << start_link_check_thread(crawled_page.body, crawled_page.url.to_s)
+        @html_files << crawled_page
       end
     end
     threads.each{|thread| thread.join }
@@ -77,23 +100,33 @@ class LinkChecker
   def check_uris_in_files
     threads = []
     html_file_paths.each do |file|
+      wait_to_spawn_thread
       threads << start_link_check_thread(open(file), file)
+      @html_files << file
     end
     threads.each{|thread| thread.join }
   end
 
   def start_link_check_thread(source, source_name)
     Thread.new do
-      results = self.class.external_link_uri_strings(source).map do |uri_string|
-        begin
-          uri = URI(uri_string)
-          response = self.class.check_uri(uri)
-          response.uri_string = uri_string
-          response
-        rescue => error
-          Error.new(:error => error.to_s, :uri_string => uri_string)
+      threads = []
+      results = []
+      self.class.external_link_uri_strings(source).each do |uri_string|
+        Thread.exclusive { @links << source }
+        wait_to_spawn_thread
+        threads << Thread.new do
+          begin
+            uri = URI(uri_string)
+            response = self.class.check_uri(uri)
+            response.uri_string = uri_string
+            Thread.exclusive { results << response }
+          rescue => error
+            Thread.exclusive { results <<
+              Error.new(:error => error.to_s, :uri_string => uri_string) }
+          end
         end
       end
+      threads.each {|thread| thread.join }
       report_results(source_name, results)
     end
   end
@@ -108,6 +141,11 @@ class LinkChecker
       warnings = []
     end
     Thread.exclusive do
+      # Store the results in the LinkChecker instance.
+      # This must be thread-exclusive to avoid a race condition.
+      @errors = @errors.concat(errors)
+      @warnings = @warnings.concat(warnings)
+
       if errors.empty?
         message = "Checked: #{file}"
         if warnings.empty? || @options[:no_warnings]
@@ -164,4 +202,15 @@ class LinkChecker
     end
   end
 
+  private
+
+  def wait_to_spawn_thread
+    # Never spawn more than the specified maximum number of threads.
+    until Thread.list.select {|thread| thread.status == "run"}.count <
+      (1 + @options[:max_threads]) do
+      # Wait 5 milliseconds before trying again.
+      sleep 0.005
+    end
+  end
+
 end

data/link-checker.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = "link-checker"
-  s.version = "0.5.2"
+  s.version = "0.6.0"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Ryan Alyn Porter"]
-  s.date = "2012-10-07"
+  s.date = "2012-10-08"
   s.description = "Check the links in a web site before deploying, using Nokogiri."
   s.executables = ["check-links"]
   s.extra_rdoc_files = [
@@ -221,7 +221,8 @@ Gem::Specification.new do |s|
     "spec/test-site/public/assets/jwplayer/glow/sharing/shareScreen.png",
     "spec/test-site/public/assets/jwplayer/player.swf",
     "spec/test-site/public/atom.xml",
-    "spec/test-site/public/blog/2012/10/02/a-list-of-links/index.html",
+    "spec/test-site/public/blog/2012/10/07/a-bad-link/index.html",
+    "spec/test-site/public/blog/2012/10/07/some-good-links/index.html",
     "spec/test-site/public/blog/archives/index.html",
     "spec/test-site/public/favicon.png",
     "spec/test-site/public/images/bird_32_gray.png",
@@ -304,7 +305,8 @@ Gem::Specification.new do |s|
     "spec/test-site/source/_layouts/default.html",
     "spec/test-site/source/_layouts/page.html",
     "spec/test-site/source/_layouts/post.html",
-    "spec/test-site/source/_posts/2012-10-02-a-list-of-links.markdown",
+    "spec/test-site/source/_posts/2012-10-07-a-bad-link.markdown",
+    "spec/test-site/source/_posts/2012-10-07-some-good-links.markdown",
     "spec/test-site/source/assets/jwplayer/glow/controlbar/background.png",
     "spec/test-site/source/assets/jwplayer/glow/controlbar/blankButton.png",
     "spec/test-site/source/assets/jwplayer/glow/controlbar/divider.png",

data/spec/link-checker_spec.rb

@@ -11,18 +11,18 @@ describe LinkChecker do
 
     it "finds all of the HTML files in the target path." do
       files = LinkChecker.new(:target => @site_path).html_file_paths
-      files.size.should == 3
+      files.size.should == 4
     end
 
     it "finds all of the external links in an HTML file." do
       links = LinkChecker.external_link_uri_strings(
-        open('spec/test-site/public/blog/2012/10/02/a-list-of-links/index.html'))
+        open('spec/test-site/public/blog/2012/10/07/some-good-links/index.html'))
       links.size.should == 4
     end
 
     it "finds all of the external links in a string." do
       links = LinkChecker.external_link_uri_strings(
-        open('spec/test-site/public/blog/2012/10/02/a-list-of-links/index.html').read)
+        open('spec/test-site/public/blog/2012/10/07/some-good-links/index.html').read)
       links.size.should == 4
     end
 
@@ -74,29 +74,36 @@ describe LinkChecker do
 
     before(:each) do
       LinkChecker.any_instance.stub(:html_file_paths) {
-        ['spec/test-site/public/blog/2012/10/02/a-list-of-links/index.html'] }
+        ['spec/test-site/public/blog/2012/10/07/some-good-links/index.html'] }
       LinkChecker.stub(:external_link_uri_strings).and_return(
-        ['http://something.com', 'http://something-else.com'])
+        (1..20).map{|i| "http://something-#{i}.com" } )
     end
 
     it "prints green when the links are good." do
       LinkChecker.stub(:check_uri) do
+        sleep 0.5 # Make LinkChecker#wait_to_spawn_thread wait a little.
         LinkChecker::Good.new(:uri_string => 'http://something.com')
       end
-      $stdout.should_receive(:puts).with(/Checked/i).once
-      LinkChecker.new(:target => @site_path).check_uris.should == 0 # Return value: good
+      $stdout.should_receive(:puts).with(/Checked\: .*\.html/).once
+      $stdout.should_receive(:puts).with(/Checked 20 links in 1 HTML file and found no errors/)
+      LinkChecker.new(
+        :target => @site_path,
+        # This is to make sure that the entire LinkChecker#wait_to_spawn_thread gets hit during testing.
+        :options => { :max_threads => 1 }
+      ).check_uris.should == 0 # Return value: good
     end
 
     it "prints red when the links are bad." do
       LinkChecker.stub(:check_uri) do
         LinkChecker::Error.new(
           :uri_string => 'http://something.com',
-          :response => 'No.'
+          :error => 'No.'
         )
       end
-      $stdout.should_receive(:puts).with(/Problem/i).once
-      $stdout.should_receive(:puts).with(/Link/i).twice
-      $stdout.should_receive(:puts).with(/Response/i).twice
+      $stdout.should_receive(:puts).with(/Problem\: .*\.html/).once
+      $stdout.should_receive(:puts).with(/Link\: http/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Response/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Checked 20 links in 1 HTML file and found 20 errors/)
       LinkChecker.new(:target => @site_path).check_uris.should == 1 # Return value: error
     end
 
@@ -107,9 +114,10 @@ describe LinkChecker do
           :final_desination => 'http://something-else.com'
         )
       end
-      $stdout.should_receive(:puts).with(/Checked/i).once
-      $stdout.should_receive(:puts).with(/Warning/i).twice
-      $stdout.should_receive(:puts).with(/Redirected/i).twice
+      $stdout.should_receive(:puts).with(/Checked\: .*\.html/).once
+      $stdout.should_receive(:puts).with(/Warning/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Redirected/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Checked 20 links in 1 HTML file and found no errors/)
       LinkChecker.new(:target => @site_path).check_uris.should == 0 # Return value: good
     end
 
@@ -120,9 +128,10 @@ describe LinkChecker do
           :final_destination_uri_string => 'http://something-else.com'
         )
       end
-      $stdout.should_receive(:puts).with(/Problem/i).once
-      $stdout.should_receive(:puts).with(/Link/i).twice
-      $stdout.should_receive(:puts).with(/Redirected/i).twice
+      $stdout.should_receive(:puts).with(/Problem\: .*\.html/).once
+      $stdout.should_receive(:puts).with(/Link/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Redirected/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Checked 20 links in 1 HTML file and found 20 errors/)
       LinkChecker.new(
         :target => @site_path,
         :options => { :warnings_are_errors => true }
@@ -136,9 +145,10 @@ describe LinkChecker do
           :final_destination_uri_string => 'http://something-else.com'
         )
       end
-      $stdout.should_receive(:puts).with(/Checked/i).once
-      $stdout.should_receive(:puts).with(/Warning/i).twice
-      $stdout.should_receive(:puts).with(/Redirected/i).twice
+      $stdout.should_receive(:puts).with(/Checked\: .*\.html/).once
+      $stdout.should_receive(:puts).with(/Warning/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Redirected/).exactly(20).times
+      $stdout.should_receive(:puts).with(/Checked 20 links in 1 HTML file and found no errors/)
       LinkChecker.new(:target => @site_path).check_uris.should == 0 # Return value: good
     end
 
@@ -149,9 +159,9 @@ describe LinkChecker do
     it "declares them to be bad." do
       LinkChecker.stub(:external_link_uri_strings).and_return(
         ['hQQp://!!!.com', 'hOOp://???.com'])
-      $stdout.should_receive(:puts).with(/Problem/i).once
-      $stdout.should_receive(:puts).with(/Link/i).twice
-      $stdout.should_receive(:puts).with(/Response/i).twice
+      $stdout.should_receive(:puts).with(/Problem\: .*\.html/).once
+      $stdout.should_receive(:puts).with(/Link/).twice
+      $stdout.should_receive(:puts).with(/Response/).twice
       thread = LinkChecker.new(:target => @site_path).start_link_check_thread('<html></html>', 'source.html')
       thread.join
     end
@@ -164,7 +174,8 @@ describe LinkChecker do
     LinkChecker.stub(:check_uri) do
       LinkChecker::Good.new(:uri_string => 'http://something.com')
     end
-    $stdout.should_receive(:puts).with(/Checked/i).once
+    $stdout.should_receive(:puts).with(/Checked\: http/).once
+    $stdout.should_receive(:puts).with(/Checked 1 link in 1 HTML file and found no errors/)
     LinkChecker.new(:target => 'http://some-target.com').check_uris
   end
 

data/spec/test-site/public/atom.xml

@@ -4,7 +4,7 @@
   <title><![CDATA[My Octopress Blog]]></title>
   <link href="http://yoursite.com/atom.xml" rel="self"/>
   <link href="http://yoursite.com/"/>
-  <updated>2012-10-02T22:10:18-04:00</updated>
+  <updated>2012-10-07T13:09:13-04:00</updated>
   <id>http://yoursite.com/</id>
   <author>
     <name><![CDATA[Your Name]]></name>
@@ -14,13 +14,22 @@
 
   
   <entry>
-    <title type="html"><![CDATA[A list of links]]></title>
-    <link href="http://yoursite.com/blog/2012/10/02/a-list-of-links/"/>
-    <updated>2012-10-02T14:44:00-04:00</updated>
-    <id>http://yoursite.com/blog/2012/10/02/a-list-of-links</id>
+    <title type="html"><![CDATA[Some good links]]></title>
+    <link href="http://yoursite.com/blog/2012/10/07/some-good-links/"/>
+    <updated>2012-10-07T00:00:00-04:00</updated>
+    <id>http://yoursite.com/blog/2012/10/07/some-good-links</id>
     <content type="html"><![CDATA[<p><a href="http://goodlink.com">A good link</a></p>
 
-<p><a href="http://brokenlink.com">A broken link</a></p>
+<p><a href="http://anothergoodlink.com">Another good link</a></p>
+]]></content>
+  </entry>
+  
+  <entry>
+    <title type="html"><![CDATA[Some good links]]></title>
+    <link href="http://yoursite.com/blog/2012/10/07/a-bad-link/"/>
+    <updated>2012-10-07T00:00:00-04:00</updated>
+    <id>http://yoursite.com/blog/2012/10/07/a-bad-link</id>
+    <content type="html"><![CDATA[<p><a href="http://badlink.com">A bad link</a></p>
 ]]></content>
   </entry>
   

data/spec/test-site/public/blog/2012/10/07/a-bad-link/index.html

@@ -0,0 +1,200 @@
+
+<!DOCTYPE html>
+<!--[if IEMobile 7 ]><html class="no-js iem7"><![endif]-->
+<!--[if lt IE 9]><html class="no-js lte-ie8"><![endif]-->
+<!--[if (gt IE 8)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <title>Some good links - My Octopress Blog</title>
+  <meta name="author" content="Your Name">
+
+  
+  <meta name="description" content="A bad link
+">
+  
+
+  <!-- http://t.co/dKP3o1e -->
+  <meta name="HandheldFriendly" content="True">
+  <meta name="MobileOptimized" content="320">
+  <meta name="viewport" content="width=device-width, initial-scale=1">
+
+  
+  <link rel="canonical" href="http://yoursite.com/blog/2012/10/07/a-bad-link/">
+  <link href="/favicon.png" rel="icon">
+  <link href="/stylesheets/screen.css" media="screen, projection" rel="stylesheet" type="text/css">
+  <script src="/javascripts/modernizr-2.0.js"></script>
+  <script src="/javascripts/ender.js"></script>
+  <script src="/javascripts/octopress.js" type="text/javascript"></script>
+  <link href="/atom.xml" rel="alternate" title="My Octopress Blog" type="application/atom+xml">
+  <!--Fonts from Google"s Web font directory at http://google.com/webfonts -->
+<link href="http://fonts.googleapis.com/css?family=PT+Serif:regular,italic,bold,bolditalic" rel="stylesheet" type="text/css">
+<link href="http://fonts.googleapis.com/css?family=PT+Sans:regular,italic,bold,bolditalic" rel="stylesheet" type="text/css">
+
+  
+
+</head>
+
+<body   >
+  <header role="banner"><hgroup>
+  <h1><a href="/">My Octopress Blog</a></h1>
+  
+    <h2>A blogging framework for hackers.</h2>
+  
+</hgroup>
+
+</header>
+  <nav role="navigation"><ul class="subscription" data-subscription="rss">
+  <li><a href="/atom.xml" rel="subscribe-rss" title="subscribe via RSS">RSS</a></li>
+  
+</ul>
+  
+<form action="http://google.com/search" method="get">
+  <fieldset role="search">
+    <input type="hidden" name="q" value="site:yoursite.com" />
+    <input class="search" type="text" name="q" results="0" placeholder="Search"/>
+  </fieldset>
+</form>
+  
+<ul class="main-navigation">
+  <li><a href="/">Blog</a></li>
+  <li><a href="/blog/archives">Archives</a></li>
+</ul>
+
+</nav>
+  <div id="main">
+    <div id="content">
+      <div>
+<article class="hentry" role="article">
+  
+  <header>
+    
+      <h1 class="entry-title">Some Good Links</h1>
+    
+    
+      <p class="meta">
+        
+
+
+
+
+
+
+
+
+  
+
+
+<time datetime="2012-10-07T00:00:00-04:00" pubdate data-updated="true">Oct 7<span>th</span>, 2012</time>
+        
+      </p>
+    
+  </header>
+
+
+<div class="entry-content"><p><a href="http://badlink.com">A bad link</a></p>
+</div>
+
+
+  <footer>
+    <p class="meta">
+      
+  
+
+<span class="byline author vcard">Posted by <span class="fn">Your Name</span></span>
+
+      
+
+
+
+
+
+
+
+
+  
+
+
+<time datetime="2012-10-07T00:00:00-04:00" pubdate data-updated="true">Oct 7<span>th</span>, 2012</time>
+      
+
+
+    </p>
+    
+      <div class="sharing">
+  
+  <a href="http://twitter.com/share" class="twitter-share-button" data-url="http://yoursite.com/blog/2012/10/07/a-bad-link/" data-via="" data-counturl="http://yoursite.com/blog/2012/10/07/a-bad-link/" >Tweet</a>
+  
+  
+  
+</div>
+
+    
+    <p class="meta">
+      
+      
+        <a class="basic-alignment right" href="/blog/2012/10/07/some-good-links/" title="Next Post: Some good links">Some good links &raquo;</a>
+      
+    </p>
+  </footer>
+</article>
+
+</div>
+
+<aside class="sidebar">
+  
+    <section>
+  <h1>Recent Posts</h1>
+  <ul id="recent_posts">
+    
+      <li class="post">
+        <a href="/blog/2012/10/07/some-good-links/">Some good links</a>
+      </li>
+    
+      <li class="post">
+        <a href="/blog/2012/10/07/a-bad-link/">Some good links</a>
+      </li>
+    
+  </ul>
+</section>
+
+
+
+
+
+
+  
+</aside>
+
+
+    </div>
+  </div>
+  <footer role="contentinfo"><p>
+  Copyright &copy; 2012 - Your Name -
+  <span class="credit">Powered by <a href="http://octopress.org">Octopress</a></span>
+</p>
+
+</footer>
+  
+
+
+
+
+
+
+
+  <script type="text/javascript">
+    (function(){
+      var twitterWidgets = document.createElement('script');
+      twitterWidgets.type = 'text/javascript';
+      twitterWidgets.async = true;
+      twitterWidgets.src = 'http://platform.twitter.com/widgets.js';
+      document.getElementsByTagName('head')[0].appendChild(twitterWidgets);
+    })();
+  </script>
+
+
+
+
+
+</body>
+</html>

data/spec/test-site/public/blog/2012/10/{02/a-list-of-links → 07/some-good-links}/index.html

@@ -5,11 +5,11 @@
 <!--[if (gt IE 8)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]-->
 <head>
   <meta charset="utf-8">
-  <title>A list of links - My Octopress Blog</title>
+  <title>Some good links - My Octopress Blog</title>
   <meta name="author" content="Your Name">
 
   
-  <meta name="description" content="A good link A broken link
+  <meta name="description" content="A good link Another good link
 ">
   
 
@@ -19,7 +19,7 @@
   <meta name="viewport" content="width=device-width, initial-scale=1">
 
   
-  <link rel="canonical" href="http://yoursite.com/blog/2012/10/02/a-list-of-links/">
+  <link rel="canonical" href="http://yoursite.com/blog/2012/10/07/some-good-links/">
   <link href="/favicon.png" rel="icon">
   <link href="/stylesheets/screen.css" media="screen, projection" rel="stylesheet" type="text/css">
   <script src="/javascripts/modernizr-2.0.js"></script>
@@ -68,7 +68,7 @@
   
   <header>
     
-      <h1 class="entry-title">A List of Links</h1>
+      <h1 class="entry-title">Some Good Links</h1>
     
     
       <p class="meta">
@@ -84,7 +84,7 @@
   
 
 
-<time datetime="2012-10-02T14:44:00-04:00" pubdate data-updated="true">Oct 2<span>nd</span>, 2012</time>
+<time datetime="2012-10-07T00:00:00-04:00" pubdate data-updated="true">Oct 7<span>th</span>, 2012</time>
         
       </p>
     
@@ -93,7 +93,7 @@
 
 <div class="entry-content"><p><a href="http://goodlink.com">A good link</a></p>
 
-<p><a href="http://brokenlink.com">A broken link</a></p>
+<p><a href="http://anothergoodlink.com">Another good link</a></p>
 </div>
 
 
@@ -116,7 +116,7 @@
   
 
 
-<time datetime="2012-10-02T14:44:00-04:00" pubdate data-updated="true">Oct 2<span>nd</span>, 2012</time>
+<time datetime="2012-10-07T00:00:00-04:00" pubdate data-updated="true">Oct 7<span>th</span>, 2012</time>
       
 
 
@@ -124,7 +124,7 @@
     
       <div class="sharing">
   
-  <a href="http://twitter.com/share" class="twitter-share-button" data-url="http://yoursite.com/blog/2012/10/02/a-list-of-links/" data-via="" data-counturl="http://yoursite.com/blog/2012/10/02/a-list-of-links/" >Tweet</a>
+  <a href="http://twitter.com/share" class="twitter-share-button" data-url="http://yoursite.com/blog/2012/10/07/some-good-links/" data-via="" data-counturl="http://yoursite.com/blog/2012/10/07/some-good-links/" >Tweet</a>
   
   
   
@@ -133,6 +133,8 @@
     
     <p class="meta">
       
+        <a class="basic-alignment left" href="/blog/2012/10/07/a-bad-link/" title="Previous Post: Some good links">&laquo; Some good links</a>
+      
       
     </p>
   </footer>
@@ -147,7 +149,11 @@
   <ul id="recent_posts">
     
       <li class="post">
-        <a href="/blog/2012/10/02/a-list-of-links/">A list of links</a>
+        <a href="/blog/2012/10/07/some-good-links/">Some good links</a>
+      </li>
+    
+      <li class="post">
+        <a href="/blog/2012/10/07/a-bad-link/">Some good links</a>
       </li>
     
   </ul>

data/spec/test-site/public/blog/archives/index.html

@@ -9,8 +9,9 @@
   <meta name="author" content="Your Name">
 
   
-  <meta name="description" content=" Blog Archive 2012 A list of links
-Oct 02 2012 Recent Posts A list of links ">
+  <meta name="description" content=" Blog Archive 2012 Some good links
+Oct 07 2012 Some good links
+Oct 07 2012 Recent Posts Some good links Some good links ">
   
 
   <!-- http://t.co/dKP3o1e -->
@@ -80,8 +81,18 @@ Oct 02 2012 Recent Posts A list of links ">
 
 <article>
   
-<h1><a href="/blog/2012/10/02/a-list-of-links/">A list of links</a></h1>
-<time datetime="2012-10-02T14:44:00-04:00" pubdate><span class='month'>Oct</span> <span class='day'>02</span> <span class='year'>2012</span></time>
+<h1><a href="/blog/2012/10/07/some-good-links/">Some good links</a></h1>
+<time datetime="2012-10-07T00:00:00-04:00" pubdate><span class='month'>Oct</span> <span class='day'>07</span> <span class='year'>2012</span></time>
+
+
+</article>
+
+
+
+<article>
+  
+<h1><a href="/blog/2012/10/07/a-bad-link/">Some good links</a></h1>
+<time datetime="2012-10-07T00:00:00-04:00" pubdate><span class='month'>Oct</span> <span class='day'>07</span> <span class='year'>2012</span></time>
 
 
 </article>
@@ -100,7 +111,11 @@ Oct 02 2012 Recent Posts A list of links ">
   <ul id="recent_posts">
     
       <li class="post">
-        <a href="/blog/2012/10/02/a-list-of-links/">A list of links</a>
+        <a href="/blog/2012/10/07/some-good-links/">Some good links</a>
+      </li>
+    
+      <li class="post">
+        <a href="/blog/2012/10/07/a-bad-link/">Some good links</a>
       </li>
     
   </ul>

data/spec/test-site/public/index.html

@@ -9,7 +9,7 @@
   <meta name="author" content="Your Name">
 
   
-  <meta name="description" content="A good link A broken link
+  <meta name="description" content="A good link Another good link
 ">
   
 
@@ -71,7 +71,7 @@
       
   <header>
     
-      <h1 class="entry-title"><a href="/blog/2012/10/02/a-list-of-links/">A List of Links</a></h1>
+      <h1 class="entry-title"><a href="/blog/2012/10/07/some-good-links/">Some Good Links</a></h1>
     
     
       <p class="meta">
@@ -87,7 +87,7 @@
   
 
 
-<time datetime="2012-10-02T14:44:00-04:00" pubdate data-updated="true">Oct 2<span>nd</span>, 2012</time>
+<time datetime="2012-10-07T00:00:00-04:00" pubdate data-updated="true">Oct 7<span>th</span>, 2012</time>
         
       </p>
     
@@ -96,7 +96,43 @@
 
   <div class="entry-content"><p><a href="http://goodlink.com">A good link</a></p>
 
-<p><a href="http://brokenlink.com">A broken link</a></p>
+<p><a href="http://anothergoodlink.com">Another good link</a></p>
+</div>
+  
+  
+
+
+    </article>
+  
+  
+    <article>
+      
+  <header>
+    
+      <h1 class="entry-title"><a href="/blog/2012/10/07/a-bad-link/">Some Good Links</a></h1>
+    
+    
+      <p class="meta">
+        
+
+
+
+
+
+
+
+
+  
+
+
+<time datetime="2012-10-07T00:00:00-04:00" pubdate data-updated="true">Oct 7<span>th</span>, 2012</time>
+        
+      </p>
+    
+  </header>
+
+
+  <div class="entry-content"><p><a href="http://badlink.com">A bad link</a></p>
 </div>
   
   
@@ -117,7 +153,11 @@
   <ul id="recent_posts">
     
       <li class="post">
-        <a href="/blog/2012/10/02/a-list-of-links/">A list of links</a>
+        <a href="/blog/2012/10/07/some-good-links/">Some good links</a>
+      </li>
+    
+      <li class="post">
+        <a href="/blog/2012/10/07/a-bad-link/">Some good links</a>
       </li>
     
   </ul>

data/spec/test-site/public/sitemap.xml

@@ -1,15 +1,19 @@
 <?xml version='1.0' encoding='UTF-8'?>
 <urlset xmlns='http://www.sitemaps.org/schemas/sitemap/0.9'>
     <url>
-        <loc>http://yoursite.com/blog/2012/10/02/a-list-of-links/</loc>
-        <lastmod>2012-10-02T21:00:55-04:00</lastmod>
+        <loc>http://yoursite.com/blog/2012/10/07/a-bad-link/</loc>
+        <lastmod>2012-10-07T11:47:25-04:00</lastmod>
+    </url>
+    <url>
+        <loc>http://yoursite.com/blog/2012/10/07/some-good-links/</loc>
+        <lastmod>2012-10-07T11:47:18-04:00</lastmod>
     </url>
     <url>
         <loc>http://yoursite.com/blog/archives/</loc>
-        <lastmod>2012-10-02T21:00:55-04:00</lastmod>
+        <lastmod>2012-10-07T11:47:25-04:00</lastmod>
     </url>
     <url>
         <loc>http://yoursite.com/</loc>
-        <lastmod>2012-10-02T21:00:55-04:00</lastmod>
+        <lastmod>2012-10-07T11:47:25-04:00</lastmod>
     </url>
 </urlset>

data/spec/test-site/source/_posts/2012-10-07-a-bad-link.markdown

@@ -0,0 +1,9 @@
+---
+layout: post
+title: "Some good links"
+date: 2012-10-07
+comments: true
+categories: 
+---
+
+[A bad link](http://badlink.com)

data/spec/test-site/source/_posts/2012-10-07-some-good-links.markdown

@@ -0,0 +1,11 @@
+---
+layout: post
+title: "Some good links"
+date: 2012-10-07
+comments: true
+categories: 
+---
+
+[A good link](http://goodlink.com)
+
+[Another good link](http://anothergoodlink.com)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: link-checker
 version: !ruby/object:Gem::Version
-  version: 0.5.2
+  version: 0.6.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-10-07 00:00:00.000000000 Z
+date: 2012-10-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -400,7 +400,8 @@ files:
 - spec/test-site/public/assets/jwplayer/glow/sharing/shareScreen.png
 - spec/test-site/public/assets/jwplayer/player.swf
 - spec/test-site/public/atom.xml
-- spec/test-site/public/blog/2012/10/02/a-list-of-links/index.html
+- spec/test-site/public/blog/2012/10/07/a-bad-link/index.html
+- spec/test-site/public/blog/2012/10/07/some-good-links/index.html
 - spec/test-site/public/blog/archives/index.html
 - spec/test-site/public/favicon.png
 - spec/test-site/public/images/bird_32_gray.png
@@ -483,7 +484,8 @@ files:
 - spec/test-site/source/_layouts/default.html
 - spec/test-site/source/_layouts/page.html
 - spec/test-site/source/_layouts/post.html
-- spec/test-site/source/_posts/2012-10-02-a-list-of-links.markdown
+- spec/test-site/source/_posts/2012-10-07-a-bad-link.markdown
+- spec/test-site/source/_posts/2012-10-07-some-good-links.markdown
 - spec/test-site/source/assets/jwplayer/glow/controlbar/background.png
 - spec/test-site/source/assets/jwplayer/glow/controlbar/blankButton.png
 - spec/test-site/source/assets/jwplayer/glow/controlbar/divider.png
@@ -558,7 +560,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 3561793786704085147
+      hash: 1307597629832746042
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:

data/spec/test-site/source/_posts/2012-10-02-a-list-of-links.markdown

@@ -1,11 +0,0 @@
----
-layout: post
-title: "A list of links"
-date: 2012-10-02 14:44
-comments: true
-categories: 
----
-
-[A good link](http://goodlink.com)
-
-[A broken link](http://brokenlink.com)