RubyGems - raakt - Versions diffs - 0.1 - Mend

raakt 0.1

Files changed (49) hide show

	@@ -0,0 +1 @@
1	+ <empty>Blank document</empty>

data/tests/emptytitledoc.htm ADDED

@@ -0,0 +1,8 @@
+<html lang="en">
+  <head>
+    <title></title>
+  </head>
+  <body>
+    <p>This is a document with en amty title element.</p>
+  </body>
+</html>

data/tests/fielddoc1.htm ADDED

	@@ -0,0 +1,2 @@
1	+ <label for="myid">My label</label>
2	+ <input type="text" id="myid" />

data/tests/fielddoc2.htm ADDED

@@ -0,0 +1,11 @@
+<html>
+  <body>
+    <h3>First heading</h3>
+    <label for="textid">My label</label>
+    <p id="pid">This is a minimal <a href="http://www.w3.org/TR/xhtml1/">image document.</a>
+    <input type="text" id="textid" />
+    <input type="hidden" id="hiddenid"/>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">image document.</a>
+    <INPUT TYPE='text' ID='myid'>
+  </body>
+</html>

data/tests/fielddoc3.htm ADDED

@@ -0,0 +1,14 @@
+<LABEL for='myid' lang='en'>My input
+</LABEL>
+<input
+        type="text"
+        id="myid"
+>
+<textarea>value</textarea>
+<label for="selectid">
+    My select
+</label>
+<select lang='en'
+        id='selectid'>
+        <option>My value</opiton>
+</select>

data/tests/flickerdoc1.htm ADDED

File without changes

data/tests/framedoc1.htm ADDED

@@ -0,0 +1,22 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Frameset//EN"
+   "http://www.w3.org/TR/html4/frameset.dtd">
+<HTML>
+<HEAD>
+<TITLE>A simple frameset document</TITLE>
+</HEAD>
+<FRAMESET cols="20%, 80%">
+  <FRAMESET rows="100, 200">
+      <FRAME src="contents_of_frame1.html">
+      <FRAME src="contents_of_frame2.gif">
+  </FRAMESET>
+  <FRAME src="contents_of_frame3.html">
+  <NOFRAMES>
+      <P>This frameset document contains:
+      <UL>
+         <LI><A href="contents_of_frame1.html">Some neat contents</A>
+         <LI><IMG src="contents_of_frame2.gif" alt="A neat image">
+         <LI><A href="contents_of_frame3.html">Some other neat contents</A>
+      </UL>
+  </NOFRAMES>
+</FRAMESET>
+</HTML>

data/tests/framedoc2.htm ADDED

@@ -0,0 +1,8 @@
+<frameset
+cols="20%, 80%">
+  <FRAMESET rows="100, 200">
+      <FRAME src="contents_of_frame1.html" title="menu">
+      <FRAME src="contents_of_frame2.gif" title="content">
+  </FRAMESET>
+</frameset>
+</HTML>

data/tests/full_google.htm ADDED

@@ -0,0 +1,17 @@
+<html><head><meta http-equiv="content-type" content="text/html; charset=UTF-8"><title>Google</title><style><!--
+body,td,a,p,.h{font-family:arial,sans-serif;}
+.h{font-size: 20px;}
+.q{color:#0000cc;}
+-->
+</style>
+<script>
+<!--
+function sf(){document.f.q.focus();}
+function asq(event,el,oi,cad,ct,cd,sg){if(window.XMLHttpRequest){if(el.handledFirstTime){el.handledFirstTime=false;return false;}el.handledFirstTime=true;var e = window.encodeURIComponent ? encodeURIComponent : escape;var oi_param="";var cad_param="";if (oi) oi_param="&oi="+e(oi);if (cad) cad_param="&cad="+e(cad);var x=new XMLHttpRequest();x.open("GET","/url?sa=T"+oi_param+cad_param+"&ct="+e(ct)+"&cd="+e(cd)+"&url="+e(el.href).replace(/\+/g,"%2B")+"&ei="+sg,true);var m=event.altKey||event.metaKey;if(!m){x.onreadystatechange=function(){if(x.readyState==4){clearTimeout(timeoutid);el.dispatchEvent(event);}};var timeoutid=setTimeout(function(){x.abort();el.dispatchEvent(event);},2000);}x.send(null);return m;}return true;}
+// -->
+</script>
+</head><body bgcolor=#ffffff text=#000000 link=#0000cc vlink=#551a8b alink=#ff0000 onLoad=sf() topmargin=3 marginheight=3><center><table border=0 cellspacing=0 cellpadding=0 width=100%><tr><td align=right nowrap><font size=-1><b>...</b>&nbsp;|&nbsp;<a href="/url?sa=p&pref=ig&pval=3&q=http://www.google.com/" onmousedown="return asq(event,this,'promos','hppphou:def','pro','1','&sig2=')">Personalized Home</a>&nbsp;|&nbsp;<a href="/searchhistory/?hl=en">Search History</a>&nbsp;|&nbsp;<a href="https://www.google.com/accounts/ManageAccount">My Account</a>&nbsp;|&nbsp;<a href="http://www.google.com/accounts/Logout?continue=http://www.google.com/">Sign out</a></font></td></tr><tr height=4><td><img alt="" width=1 height=1></td></tr></table><img src="/intl/en/images/logo.gif" width=276 height=110 alt="Google"><br><br>
+<form action=/search name=f><script><!--
+function qs(el) {if (window.RegExp && window.encodeURIComponent) {var ue=el.href;var qe=encodeURIComponent(document.f.q.value);if(ue.indexOf("q=")!=-1){el.href=ue.replace(new RegExp("q=[^&$]*"),"q="+qe);}else{el.href=ue+"&q="+qe;}}return 1;}
+// -->
+</script><table border=0 cellspacing=0 cellpadding=4><tr><td nowrap><font size=-1><b>Web</b>&nbsp;&nbsp;&nbsp;&nbsp;<a id=1a class=q href="/imghp?hl=en&tab=wi" onClick="return qs(this);">Images</a>&nbsp;&nbsp;&nbsp;&nbsp;<a id=2a class=q href="http://groups.google.com/grphp?hl=en&tab=wg" onClick="return qs(this);">Groups</a>&nbsp;&nbsp;&nbsp;&nbsp;<a id=4a class=q href="http://news.google.com/nwshp?hl=en&tab=wn" onClick="return qs(this);">News</a>&nbsp;&nbsp;&nbsp;&nbsp;<a id=5a class=q href="http://froogle.google.com/frghp?hl=en&tab=wf" onClick="return qs(this);">Froogle</a>&nbsp;&nbsp;&nbsp;&nbsp;<a id=7a class=q href="/maphp?hl=en&tab=wl" onClick="return qs(this);">Maps</a>&nbsp;&nbsp;&nbsp;&nbsp;<b><a href="/intl/en/options/" class=q>more&nbsp;&raquo;</a></b></font></td></tr></table><table cellspacing=0 cellpadding=0><tr><td width=25%>&nbsp;</td><td align=center><input type=hidden name=hl value=en><input maxlength=2048 size=55 name=q value="" title="Google Search"><br><input type=submit value="Google Search" name=btnG><input type=submit value="I'm Feeling Lucky" name=btnI></td><td valign=top nowrap width=25%><font size=-2>&nbsp;&nbsp;<a href=/advanced_search?hl=en>Advanced Search</a><br>&nbsp;&nbsp;<a href=/preferences?hl=en>Preferences</a><br>&nbsp;&nbsp;<a href=/language_tools?hl=en>Language Tools</a></font></td></tr></table></form><br><br><font size=-1><a href="/intl/en/ads/">Advertising&nbsp;Programs</a> - <a href=/intl/en/services/>Business Solutions</a> - <a href=/intl/en/about.html>About Google</a> - <b><a href=http://www.google.se/>Go to Google Sverige</a></b></font><p><font size=-2>&copy;2006 Google</font></p></center></body></html>

data/tests/headingsdoc1.htm ADDED

@@ -0,0 +1,17 @@
+<!DOCTYPE html
+     PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
+    "DTD/xhtml1-strict.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+  <head>
+    <title>This is the title</title>
+  </head>
+  <body>
+    <h1>First h1 heading</h1>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <h1>Second h1
+    heading</h1>
+    <table><tr><td>Test table</td></tr></table>
+    <H1>Third h1 heading</H1>
+  </body>
+</html>

data/tests/headingsdoc2.htm ADDED

@@ -0,0 +1,14 @@
+<html>
+  <head>
+    <title>This is the title</title>
+  </head>
+  <body>
+    <h1>First h1 heading</h1>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <h2>First h2
+    heading</h2>
+    <table><tr><td>Test table</td></tr></table>
+    <H3>First h3 heading</H3>
+  </body>
+</html>

data/tests/headingsdoc3.htm ADDED

@@ -0,0 +1,6 @@
+<html>
+  <body>
+    <h3>First heading</h3>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+  </body>
+</html>

data/tests/headingsdoc4.htm ADDED

@@ -0,0 +1,9 @@
+<html>
+  <body>
+    <h1>First h1 heading</h1>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <table><tr><td>Test table</td></tr></table>
+    <H3>First h3 heading</H3>
+  </body>
+</html>

data/tests/headingsdoc5.htm ADDED

@@ -0,0 +1,9 @@
+<html>
+  <body>
+    <h2>First h1 heading</h2>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <table><tr><td>Test table</td></tr></table>
+    <H4>First h3 heading</H4>
+  </body>
+</html>

data/tests/headingsdoc6.htm ADDED

@@ -0,0 +1,6 @@
+<h1>h1 heading</h1>
+<h2>h2 heading</h2>
+<h1>h1 heading 2</h1>
+<h2>h2 heading 2</h2>
+<h3>h3 heading</h3>
+<h1>h1 heading 3</h1>

data/tests/headingsdoc7.htm ADDED

@@ -0,0 +1,8 @@
+<html>
+  <body>
+    <h2></h2>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <table><tr><td>Test table</td></tr></table>
+  </body>
+</html>

data/tests/headingsdoc8.htm ADDED

@@ -0,0 +1,12 @@
+<script type="text/javascript">
+function printpopup(url){
+        MyWin = window.open("","","scrollbars=no,resizable=yes,toolbar=no,location=no,directories=no,status=no,menubar=no,width=100,height=100");
+        with(MyWin.document){
+        open();
+        write("<html>\n<head><scr"+"ipt>awidth=(document.layers)?0:8;awidth+=(document.all)?4:0;aheight=(document.layers)?0:29;</scr"+"ipt>\n<title>This is not the title</title>\n</head>\n<body onLoad=\"window.resizeTo(document.images[0].width+awidth,document.images[0].height+aheight)\"marginwidth=0 marginheight=0 leftmargin=0 topmargin=0 rightmargin=0 style=\"overflow:hidden;\">\n");
+        write("<h1>This is not a document heading</h1>\n");
+        write("</body>\n</html>\n");
+        close();
+        }
+}
+</script>

data/tests/headingsdoc9.htm ADDED

@@ -0,0 +1,20 @@
+<html lang="en">
+  <head>
+    <title>Sample</title>
+  </head>
+  <body>
+    <p>This is a document with an empty title element.</p>
+	<script type="text/javascript">
+	function printpopup(url){
+	        MyWin = window.open("","","");
+	        with(MyWin.document){
+	        open();
+	        write("<html><head><title>This is not the title</title>\n</head>\n<body>\n");
+	        write("<h1>This is not a document heading</h1>\n");
+	        write("</body>\n</html>\n");
+	        close();
+	        }
+	}
+	</script>
+  </body>
+</html>

data/tests/imagedoc1.htm ADDED

@@ -0,0 +1,8 @@
+<img
+    src="image.png" />
+<img src="image.png" alt="">
+<img src=image.png alt=mytext>
+<IMG
+    SRC="image.png"
+    ALT="mytext"
+/>

data/tests/imagedoc2.htm ADDED

	@@ -0,0 +1 @@
1	+ <p><img src="image.png" alt=''></p>

data/tests/imagedoc3.htm ADDED

@@ -0,0 +1,11 @@
+<html>
+  <body>
+    <h3>First heading</h3>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">image document.</a>
+    <img src="noimagealt.png" border="0">
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">image document.</a>
+    <img src="/noimagealt2.png">
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">image document.</a>
+    <img src="../folder/noimagealt3.png" border="0">
+  </body>
+</html>

data/tests/imagedoc4.htm ADDED

@@ -0,0 +1,7 @@
+<html>
+  <body>
+    <h3>First heading</h3>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">image document.</a>
+    <IMG src="noimagealt.png" border="0">
+  </body>
+</html>

data/tests/invalidelements1.htm ADDED

@@ -0,0 +1,18 @@
+<html lang="en">
+  <head>
+    <title>This is the title</title>
+  </head>
+  <body>
+    <h1>First h1 heading</h1>
+    <p>This is a <b>minimal</b> <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <p>This is a <font style="bold">bold</font> <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <h1>Second h1
+    heading</h1>
+	<blink>No blinking text!</blink>
+    <table><tr><td>Test table</td></tr></table>
+    <H1>Third h1 heading</H1>
+	<marquee>Marquees are so nineties!</marquee>
+  </body>
+</html>

data/tests/invalidhtmldoc1.htm ADDED

@@ -0,0 +1,10 @@
+<HTML>
+  <head>
+    <TITLE>This is the title</title>
+  </head>
+  <body>
+      <H1>Heading</H1>
+    <p>S small document</p>
+    <!-- comment -->
+  </body>
+</HTML>

data/tests/invalidhtmldoc2.htm ADDED

@@ -0,0 +1,20 @@
+<HTML>
+  <head>
+    <TITLE>This is the
+    title
+    </title>
+    <link rel="schema.DC" href="http://purl.org/DC/elements/1.0">
+  </head>
+  <body>
+      <!-- h3>Heading 3
+      </h3>
+      <h1>Heading 1</h1 -->
+    <p>S small document</p>
+    <p> more content </p>
+    <!-- second comment
+    -->
+  </body>
+</HTML>

data/tests/invalidxhtmldoc1.htm ADDED

@@ -0,0 +1,17 @@
+<!DOCTYPE html
+     PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
+    "DTD/xhtml1-strict.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+  <head>
+    <?xml version="1.0" encoding="UTF-8"?>
+    <title>This is the title</title>
+    <link rel="schema.DC" href="http://purl.org/DC/elements/1.0/" />
+  </head>
+  <body>
+      <h1>First h1 heading</h1>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+    <h1>Second h1 heading</h1>
+    <table><tr><td>Test table</td></tr></table>
+  </body>
+</html>

data/tests/linkdoc1.htm ADDED

@@ -0,0 +1,18 @@
+<html>
+  <head>
+    <title>This is a link document</title>
+  </head>
+  <body>
+	<h1>The first heading</h1>
+	<h2><a href="/news1">New sitem 1</a></h2>
+    <p>This is the text for the first news item. <a href="/news1">Read more </a></p>
+	<h2><a href="/news2">New sitem 2</a></h2>
+    <p>This is the text for the second news item. <a href="/news2">
+	Read more</a></p>
+	<h2><a href="/news1">New sitem 3</a></h2>
+    <p>This is the text for the third news item. <a href="/news3">Read
+	more</a></p>
+	<h2><a href="/news1">New sitem 4</a></h2>
+    <p>This is the text for the fourth news item. <a href="/news4">Read&nbsp;more</a></p>
+  </body>
+</html>

data/tests/linkdoc2.htm ADDED

@@ -0,0 +1,12 @@
+<html>
+  <head>
+    <title>This is a link document</title>
+  </head>
+  <body>
+	<h1>The first heading</h1>
+	<h2><a href="/news1">New sitem 1</a></h2>
+    <p>This is the text for the first news item. <a href="/news1" title="More on item 1">Read more</a></p>
+	<h2><a href="/news2">New sitem 2</a></h2>
+    <p>This is the text for the second news item. <a href="/news2">Read more</a></p>
+  </body>
+</html>

data/tests/linkdoc3.htm ADDED

@@ -0,0 +1,16 @@
+<html>
+  <head>
+    <title>This is a link document</title>
+  </head>
+  <body>
+	<h1>The first heading</h1>
+	<h2><a href="/news1">New sitem 1</a></h2>
+    <p>This is the text for the first news item.</p>
+	<h2><a href="/news2">New sitem 2</a></h2>
+    <p>This is the text for the second news item.</p>
+	<h2><a href="/news1">New sitem 3</a></h2>
+    <p>This is the text for the third news item.</p>
+	<h2><a href="/news1">New sitem 4</a></h2>
+    <p>This is the text for the fourth news item.</p>
+  </body>
+</html>

data/tests/linkdoc4.htm ADDED

@@ -0,0 +1,10 @@
+<html>
+  <head>
+    <title>This is a link document</title>
+  </head>
+  <body>
+	<h1>The first heading</h1>
+    <p><a href="/news1"><img src="/example.gif" alt="Read more"></a></p>
+	<p><a href="/news2">Read more</a></p>
+  </body>
+</html>

data/tests/metarefreshdoc1.htm ADDED

@@ -0,0 +1,10 @@
+<html lang="en">
+  <head>
+    <title>This is the title</title>
+    <meta http-equiv="refresh" content="5"/>
+  </head>
+  <body>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+  </body>
+</html>

data/tests/metarefreshdoc2.htm ADDED

@@ -0,0 +1,14 @@
+<html lang="en">
+  <head>
+    <title>This is the title</title>
+    <meta http-equiv="cache-control" content="no-cache" >
+	<meta http-equiv=
+			"refresh"
+		  content=
+			"5" />
+  </head>
+  <body>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+  </body>
+</html>

data/tests/metarefreshdoc3.htm ADDED

@@ -0,0 +1,10 @@
+<html lang="en">
+  <head>
+    <title>This is the title</title>
+    <META HTTP-equiv="refresh" CONTENT="5">
+  </head>
+  <body>
+    <p>This is a minimal <a href="http://www.w3.org/TR/xhtml1/">XHTML 1.0</a>
+    document.</p>
+  </body>
+</html>

data/tests/nestedcomment.htm ADDED

@@ -0,0 +1,7 @@
+<script language="JavaScript">
+<!--	function BrowserCheck() {
+    test = "<!--my" + "broken"
+    + " comment" +
+    "-->
+-->
+</script>

data/tests/newlinetext.txt ADDED

@@ -0,0 +1,3 @@
+line 1
+line 2
+line 3

data/tests/raakt_test.rb ADDED

@@ -0,0 +1,224 @@
+require File.dirname(__FILE__) + '/test_helper'
+require File.dirname(__FILE__) + '/../lib/raakt'
+require 'rubyful_soup'
+class RaaktTest < Test::Unit::TestCase
+  def setup
+    @raakt = Raakt::Test.new
+  end
+  def test_all
+    puts @raakt.all(data_full_google)
+  end
+  def test_check_images
+    assert_equal 1, @raakt.check_images(data_imagedoc1).length
+    assert_equal "missingalt", @raakt.check_images(data_imagedoc1)[0].eid
+    assert_equal 0, @raakt.check_images(data_imagedoc2).length
+    assert_equal 3, @raakt.check_images(data_imagedoc3).length
+    assert_equal 1, @raakt.check_images(data_imagedoc4).length
+  end
+  def test_check_images_in_blank_doc
+    assert_equal 0, @raakt.check_images(data_empty).length
+  end
+  def test_check_title
+    assert_equal 0, @raakt.check_title(data_xhtmldoc1).length
+    assert_equal 1, @raakt.check_title(data_empty).length
+    assert_equal "missingtitle", @raakt.check_title(data_empty)[0].eid
+    assert_equal 1, @raakt.check_title(data_emptytitledoc).length
+    assert_equal "emptytitle", @raakt.check_title(data_emptytitledoc)[0].eid
+    assert_equal 0, @raakt.check_title(data_invalidhtmldoc1).length
+    assert_equal 0, @raakt.check_title(data_invalidhtmldoc2).length
+  end
+  def test_headings
+    assert_equal 3, @raakt.headings(data_headingsdoc1).length
+    assert_equal 0, @raakt.headings(data_invalidhtmldoc2).length
+  end
+  def test_level
+    assert_equal 1, @raakt.level("h1")
+    assert_equal 2, @raakt.level("h2")
+    assert_equal 6, @raakt.level("h6")
+  end
+  def test_check_has_heading
+    assert_equal 1, @raakt.check_has_heading(data_empty).length
+    assert_equal "missingheading", @raakt.check_has_heading(data_empty)[0].eid
+    assert_equal 0, @raakt.check_has_heading(data_headingsdoc1).length
+    assert_equal 0, @raakt.check_has_heading(data_headingsdoc9).length
+    assert_equal 1, @raakt.check_has_heading(data_invalidhtmldoc2).length
+    assert_equal "missingheading", @raakt.check_has_heading(data_invalidhtmldoc2)[0].eid
+  end
+  def test_check_document_structure
+    assert_equal 0, @raakt.check_document_structure(data_headingsdoc1).length
+    assert_equal 1, @raakt.check_document_structure(data_headingsdoc3).length
+    assert_equal "firsthnoth1", @raakt.check_document_structure(data_headingsdoc3)[0].eid
+    assert_equal "wronghstructure", @raakt.check_document_structure(data_headingsdoc4)[0].eid
+    assert_equal "firsthnoth1", @raakt.check_document_structure(data_headingsdoc5)[0].eid
+    assert_equal "wronghstructure", @raakt.check_document_structure(data_headingsdoc5)[1].eid
+    assert_equal 0, @raakt.check_document_structure(data_headingsdoc6).length
+    assert_equal 0, @raakt.check_document_structure("").length
+  end
+  def test_check_for_nested_tables
+    assert_equal 0, @raakt.check_for_nested_tables(data_tabledoc1).length
+    assert_equal 0, @raakt.check_for_nested_tables(data_tabledoc2).length
+    assert_equal 1, @raakt.check_for_nested_tables(data_tabledoc3).length
+    assert_equal 0, @raakt.check_for_nested_tables(data_tabledoc4).length
+    assert_equal 1, @raakt.check_for_nested_tables(data_tabledoc5).length
+    assert_equal "hasnestedtables", @raakt.check_for_nested_tables(data_tabledoc3)[0].eid
+  end
+  def test_check_tables
+    puts @raakt.check_tables(data_tabledoc4).to_s
+    assert_equal 0, @raakt.check_tables(data_tabledoc4).length
+    assert_equal 0, @raakt.check_tables(data_tabledoc1).length
+    assert_equal 2, @raakt.check_tables(data_tabledoc2).length
+  end
+  def test_check_for_formatting_elements
+    assert_equal 1, @raakt.check_for_formatting_elements(data_invalidelements1).length
+    assert_equal "boldused", @raakt.check_for_formatting_elements(data_invalidelements1)[0].eid
+  end
+  def test_check_for_language_info
+    assert_equal 0, @raakt.check_for_language_info(data_xhtmldoc1).length
+    assert_equal 1, @raakt.check_for_language_info(data_tabledoc2).length
+    assert_equal 1, @raakt.check_for_language_info(data_tablelayoutdoc).length
+  end
+  def test_check_link_text
+    assert_equal 1, @raakt.check_link_text(data_linkdoc1).length
+    assert_equal "ambiguouslinktext", @raakt.check_link_text(data_linkdoc1)[0].eid
+    assert_equal 0, @raakt.check_link_text(data_linkdoc3).length
+    assert_equal 0, @raakt.check_link_text(data_linkdoc2).length
+    assert_equal 1, @raakt.check_link_text(data_linkdoc4).length
+  end
+  def test_get_links
+    assert_equal 8, @raakt.get_links(data_linkdoc1).length
+    assert_equal 2, @raakt.get_links(data_linkdoc4).length
+    assert_equal "Read more", @raakt.get_links(data_linkdoc4)[0][3]
+  end
+  def test_img_to_text
+    assert_equal "Read more", @raakt.img_to_text(BeautifulSoup.new("<img src='123' alt='Read more' />").img)
+  end
+  def test_elements_to_text
+    assert_equal "Read more about", @raakt.elements_to_text(BeautifulSoup.new("<a href='rrr'>Read <img src='123' alt='more' /> about</a>").a)
+    assert_equal "A sample text here", @raakt.elements_to_text(BeautifulSoup.new("<a href='r'><strong>A</strong> sample <img src='123' alt='text' /> <b>here</b></a>").a)
+  end
+  def test_normalize_text
+    assert_equal "Read more", @raakt.normalize_text("Read&nbsp;more")
+    assert_equal "Read more", @raakt.normalize_text("Read&#160;more")
+    assert_equal "Read more", @raakt.normalize_text("Read  more")
+    assert_equal "Read more", @raakt.normalize_text("Read    more")
+    assert_equal "Read more", @raakt.normalize_text("Read     more")
+    assert_equal "Read more", @raakt.normalize_text("Read\n more")
+    assert_equal "L�s mer",   @raakt.normalize_text("L�s\n mer")
+    assert_equal "L�s mer",   @raakt.normalize_text("L�s \nmer")
+    assert_equal "Read more", @raakt.normalize_text("Read \n\n\nmore")
+    assert_equal "Read more", @raakt.normalize_text("Read \tmore")
+    assert_equal "Read more", @raakt.normalize_text("  Read more")
+  end
+  def test_is_ambiguous_link
+    link_a = [1, "/news/1", "", "Read more"]
+    link_b = [2, "/news/2", "", "Read more"]
+    assert_equal true, @raakt.is_ambiguous_link(link_a, link_b)
+    link_c = [1, "/news/1", "More about first news item", "Read more"]
+    link_d = [2, "/news/2", "More about second news item", "Read more"]
+    assert_equal false, @raakt.is_ambiguous_link(link_c, link_d)
+    link_a = [1, "/news/1", nil, "Read more"]
+    link_b = [2, "/news/2", nil, "Read more"]
+    assert_equal true, @raakt.is_ambiguous_link(link_a, link_b)
+    link_g = [1, "/news/1", "", "Read more"]
+    link_h = [2, "/news/1", "", "Read more"]
+    assert_equal false, @raakt.is_ambiguous_link(link_g, link_h)
+    link_i = [1, "/news/1", "", "L�s mer"]
+    link_j = [2, "/news/2", "", "L�s\n mer"]
+    assert_equal true, @raakt.is_ambiguous_link(link_i, link_j)
+    link_k = [1, "/news/1", "", "L�s mer"]
+    link_l = [2, "/news/2", "", "L�s \nmer"]
+    assert_equal true, @raakt.is_ambiguous_link(link_k, link_l)
+  end
+  def test_get_labels
+    assert_equal 1, @raakt.get_labels(data_fielddoc1).length
+    assert_equal 1, @raakt.get_labels(data_fielddoc2).length
+    assert_equal 2, @raakt.get_labels(data_fielddoc3).length
+  end
+  def test_get_editable_fields
+    assert_equal 1, @raakt.get_editable_fields(data_fielddoc1).length
+    assert_equal 2, @raakt.get_editable_fields(data_fielddoc2).length
+    assert_equal 3, @raakt.get_editable_fields(data_fielddoc3).length
+  end
+  def test_check_form
+    assert_equal 0, @raakt.check_form(data_fielddoc1).length
+    assert_equal 1, @raakt.check_form(data_fielddoc2).length
+    assert_equal "fieldmissinglabel", @raakt.check_form(data_fielddoc2)[0].eid
+    assert_equal 1, @raakt.check_form(data_fielddoc3).length
+    assert_equal "fieldmissinglabel", @raakt.check_form(data_fielddoc3)[0].eid
+  end
+  def test_is_frameset
+    assert @raakt.is_frameset(data_framedoc1)
+    assert @raakt.is_frameset(data_framedoc2)
+    assert !@raakt.is_frameset(data_xhtmldoc1)
+  end
+  def test_check_frames
+    assert_equal 3, @raakt.check_frames(data_framedoc1).length
+    assert_equal 0, @raakt.check_frames(data_framedoc2).length
+  end
+  def test_check_for_formatting_elements
+    invaliderrs = @raakt.check_for_formatting_elements(data_invalidelements1)
+    assert_equal 2, invaliderrs.length
+    assert_equal "missingsemantics", invaliderrs[0].eid
+    assert_equal "hasflicker", invaliderrs[1].eid
+    assert_equal 0, @raakt.check_for_formatting_elements(data_xhtmldoc1).length
+  end
+  def test_refresh
+    assert_equal 1, @raakt.check_refresh(data_metarefreshdoc1).length
+    assert_equal 1, @raakt.check_refresh(data_metarefreshdoc2).length
+    assert_equal 1, @raakt.check_refresh(data_metarefreshdoc3).length
+    assert_equal 0, @raakt.check_refresh(data_xhtmldoc1).length
+  end
+end