15 files changed, 985 insertions, 40 deletions
diff --git a/engine/classes/ElggAutop.php b/engine/classes/ElggAutop.php
new file mode 100644
index 000000000..fa0c34225
--- /dev/null
+++ b/engine/classes/ElggAutop.php
@@ -0,0 +1,315 @@
+<?php
+
+/**
+ * Create wrapper P and BR elements in HTML depending on newlines. Useful when
+ * users use newlines to signal line and paragraph breaks. In all cases output
+ * should be well-formed markup.
+ *
+ * In DIV elements, Ps are only added when there would be at
+ * least two of them.
+ *
+ * @author Steve Clay <steve@mrclay.org>
+ * @license http://www.opensource.org/licenses/mit-license.php  MIT License
+ */
+class ElggAutop {
+
+	public $encoding = 'UTF-8';
+
+	/**
+	 * @var DOMDocument
+	 */
+	protected $_doc = null;
+
+	/**
+	 * @var DOMXPath
+	 */
+	protected $_xpath = null;
+
+	protected $_blocks = 'address article area aside blockquote caption col colgroup dd 
+		details div dl dt fieldset figure figcaption footer form h1 h2 h3 h4 h5 h6 header 
+		hr hgroup legend map math menu nav noscript p pre section select style summary
+		table tbody td tfoot th thead tr ul ol option li';
+
+	/**
+	 * @var array
+	 */
+	protected $_inlines = 'a abbr audio b button canvas caption cite code command datalist
+		del dfn em embed i iframe img input ins kbd keygen label map mark meter object
+		output progress q rp rt ruby s samp script select small source span strong style
+		sub sup textarea time var video wbr';
+
+	/**
+	 * Descend into these elements to add Ps
+	 *
+	 * @var array
+	 */
+	protected $_descendList = 'article aside blockquote body details div footer form
+		header section';
+
+	/**
+	 * Add Ps inside these elements
+	 *
+	 * @var array
+	 */
+	protected $_alterList = 'article aside blockquote body details div footer header
+		section';
+
+	protected $_unique = '';
+
+	public function __construct()
+	{
+		$this->_blocks = preg_split('@\\s+@', $this->_blocks);
+		$this->_descendList = preg_split('@\\s+@', $this->_descendList);
+		$this->_alterList = preg_split('@\\s+@', $this->_alterList);
+		$this->_inlines = preg_split('@\\s+@', $this->_inlines);
+		$this->_unique = md5(__FILE__);
+	}
+
+	/**
+	 * Intance of class for singleton pattern.
+	 * @var ElggAutop
+	 */
+	private static $instance;
+	
+	/**
+	 * Singleton pattern.
+	 * @return ElggAutop
+	 */
+	public static function getInstance() {
+		$className = __CLASS__;
+		if (!(self::$instance instanceof $className)) {
+			self::$instance = new $className();
+		}
+		return self::$instance;
+	}
+	
+	/**
+	 * Create wrapper P and BR elements in HTML depending on newlines. Useful when
+	 * users use newlines to signal line and paragraph breaks. In all cases output
+	 * should be well-formed markup.
+	 *
+	 * In DIV, LI, TD, and TH elements, Ps are only added when their would be at
+	 * least two of them.
+	 *
+	 * @param string $html snippet
+	 * @return string|false output or false if parse error occurred
+	 */
+	public function process($html)
+	{
+		// normalize whitespace
+		$html = str_replace(array("\r\n", "\r"), "\n", $html);
+
+		// allows preserving entities untouched
+		$html = str_replace('&', $this->_unique . 'AMP', $html);
+
+		$this->_doc = new DOMDocument();
+	   
+		// parse to DOM, suppressing loadHTML warnings
+		// http://www.php.net/manual/en/domdocument.loadhtml.php#95463
+		libxml_use_internal_errors(true);
+		if (! @$this->_doc->loadHTML("<html><meta http-equiv='content-type' " 
+				. "content='text/html; charset={$this->encoding}'><body>{$html}</body>"
+				. "</html>")) {
+			return false;
+		}
+
+		$this->_xpath = new DOMXPath($this->_doc);
+		// start processing recursively at the BODY element
+		$nodeList = $this->_xpath->query('//body[1]');
+		$this->_addParagraphs($nodeList->item(0));
+
+		// serialize back to HTML
+		$html = $this->_doc->saveHTML();
+
+		// split AUTOPs into multiples at /\n\n+/
+		$html = preg_replace('/(' . $this->_unique . 'NL){2,}/', '</autop><autop>', $html);
+		$html = str_replace(array($this->_unique . 'BR', $this->_unique . 'NL', '<br>'), 
+				'<br />',
+				$html);
+		$html = str_replace('<br /></autop>', '</autop>', $html);
+
+		// re-parse so we can handle new AUTOP elements
+
+		if (! @$this->_doc->loadHTML($html)) {
+			return false;
+		}
+		// must re-create XPath object after DOM load
+		$this->_xpath = new DOMXPath($this->_doc);
+
+		// strip AUTOPs that only have comments/whitespace
+		foreach ($this->_xpath->query('//autop') as $autop) {
+			$hasContent = false;
+			if (trim($autop->textContent) !== '') {
+				$hasContent = true;
+			} else {
+				foreach ($autop->childNodes as $node) {
+					if ($node->nodeType === XML_ELEMENT_NODE) {
+						$hasContent = true;
+						break;
+					}
+				}
+			}
+			if (! $hasContent) {
+				// strip w/ preg_replace later (faster than moving nodes out)
+				$autop->setAttribute("r", "1");
+			}
+		}
+
+		// remove a single AUTOP inside certain elements
+		
+		foreach ($this->_xpath->query('//div') as $el) {
+			$autops = $this->_xpath->query('./autop', $el);
+			if ($autops->length === 1) {
+				// strip w/ preg_replace later (faster than moving nodes out)
+				$autops->item(0)->setAttribute("r", "1");
+			}
+		}
+		
+		$html = $this->_doc->saveHTML();
+		
+		// trim to the contents of BODY
+		$bodyStart = strpos($html, '<body>');
+		$bodyEnd = strpos($html, '</body>', $bodyStart + 6);
+		$html = substr($html, $bodyStart + 6, $bodyEnd - $bodyStart - 6);
+		
+		// strip AUTOPs that should be removed
+		$html = preg_replace('@<autop r="1">(.*?)</autop>@', '\\1', $html);
+
+		// commit to converting AUTOPs to Ps
+		$html = str_replace('<autop>', "\n<p>", $html);
+		$html = str_replace('</autop>', "</p>\n", $html);
+		
+		$html = str_replace('<br>', '<br />', $html);
+		$html = str_replace($this->_unique . 'AMP', '&', $html);
+		return $html;
+	}
+
+	/**
+	 * Add P and BR elements as necessary
+	 *
+	 * @param DOMElement $el
+	 */
+	protected function _addParagraphs(DOMElement $el)
+	{
+		// no need to recurse, just queue up
+		$elsToProcess = array($el);
+		$inlinesToProcess = array();
+		while ($el = array_shift($elsToProcess)) {
+			// if true, we can alter all child nodes, if not, we'll just call
+			// _addParagraphs on each element in the descendInto list
+			$alterInline = in_array($el->nodeName, $this->_alterList);
+
+			// inside affected elements, we want to trim leading whitespace from
+			// the first text node
+			$ltrimFirstTextNode = true;
+
+			// should we open a new AUTOP element to move inline elements into?
+			$openP = true;
+			$autop = null;
+
+			// after BR, ignore a newline
+			$isFollowingBr = false;
+
+			$node = $el->firstChild;
+			while (null !== $node) {
+				if ($alterInline) {
+					if ($openP) {
+						$openP = false;
+						// create a P to move inline content into (this may be removed later)
+						$autop = $el->insertBefore($this->_doc->createElement('autop'), $node);
+					}
+				}
+
+				$isElement = ($node->nodeType === XML_ELEMENT_NODE);
+				if ($isElement) {
+					$elName = $node->nodeName;
+				}
+				$isBlock = ($isElement && in_array($elName, $this->_blocks));
+
+				if ($alterInline) {
+					$isInline = $isElement && ! $isBlock;
+					$isText = ($node->nodeType === XML_TEXT_NODE);
+					$isLastInline = (! $node->nextSibling
+								   || ($node->nextSibling->nodeType === XML_ELEMENT_NODE
+									   && in_array($node->nextSibling->nodeName, $this->_blocks)));
+					if ($isElement) {
+						$isFollowingBr = ($node->nodeName === 'br');
+					}
+
+					if ($isText) {
+						$nodeText = $node->nodeValue;
+						if ($ltrimFirstTextNode) {
+							$nodeText = ltrim($nodeText);
+							$ltrimFirstTextNode = false;
+						}
+						if ($isFollowingBr && preg_match('@^[ \\t]*\\n[ \\t]*@', $nodeText, $m)) {
+							// if a user ends a line with <br>, don't add a second BR
+							$nodeText = substr($nodeText, strlen($m[0]));
+						}
+						if ($isLastInline) {
+							$nodeText = rtrim($nodeText);
+						}
+						$nodeText = str_replace("\n", $this->_unique . 'NL', $nodeText);
+						$tmpNode = $node;
+						$node = $node->nextSibling; // move loop to next node
+
+						// alter node in place, then move into AUTOP
+						$tmpNode->nodeValue = $nodeText;
+						$autop->appendChild($tmpNode);
+
+						continue;
+					}
+				}
+				if ($isBlock || ! $node->nextSibling) {
+					if ($isBlock) {
+						if (in_array($node->nodeName, $this->_descendList)) {
+							$elsToProcess[] = $node;
+							//$this->_addParagraphs($node);
+						}
+					}
+					$openP = true;
+					$ltrimFirstTextNode = true;
+				}
+				if ($alterInline) {
+					if (! $isBlock) {
+						$tmpNode = $node;
+						if ($isElement && false !== strpos($tmpNode->textContent, "\n")) {
+							$inlinesToProcess[] = $tmpNode;
+						}
+						$node = $node->nextSibling;
+						$autop->appendChild($tmpNode);
+						continue;
+					}
+				}
+
+				$node = $node->nextSibling;
+			}
+		}
+
+		// handle inline nodes
+		// no need to recurse, just queue up
+		while ($el = array_shift($inlinesToProcess)) {
+			$ignoreLeadingNewline = false;
+			foreach ($el->childNodes as $node) {
+				if ($node->nodeType === XML_ELEMENT_NODE) {
+					if ($node->nodeValue === 'BR') {
+						$ignoreLeadingNewline = true;
+					} else {
+						$ignoreLeadingNewline = false;
+						if (false !== strpos($node->textContent, "\n")) {
+							$inlinesToProcess[] = $node;
+						}
+					}
+					continue;
+				} elseif ($node->nodeType === XML_TEXT_NODE) {
+					$text = $node->nodeValue;
+					if ($text[0] === "\n" && $ignoreLeadingNewline) {
+						$text = substr($text, 1);
+						$ignoreLeadingNewline = false;
+					}
+					$node->nodeValue = str_replace("\n", $this->_unique . 'BR', $text);
+				}
+			}
+		}
+	}
+}
diff --git a/engine/lib/output.php b/engine/lib/output.php
index 0069360f0..d50576b44 100644
--- a/engine/lib/output.php
+++ b/engine/lib/output.php
@@ -43,51 +43,25 @@ function parse_urls($text) {
 
 /**
  * Create paragraphs from text with line spacing
- * Borrowed from Wordpress.
  *
  * @param string $pee The string
- * @param bool   $br  Add BRs?
+ * @deprecated Use elgg_autop instead
  *
- * @todo Rewrite
  * @return string
  **/
-function autop($pee, $br = 1) {
-	$pee = $pee . "\n"; // just to make things a little easier, pad the end
-	$pee = preg_replace('|<br />\s*<br />|', "\n\n", $pee);
-	// Space things out a little
-	$allblocks = '(?:table|thead|tfoot|caption|colgroup|tbody|tr|td|th|div|dl|dd|dt|ul|ol|li|pre|select|form|map|area|blockquote|address|math|style|input|p|h[1-6]|hr)';
-	$pee = preg_replace('!(<' . $allblocks . '[^>]*>)!', "\n$1", $pee);
-	$pee = preg_replace('!(</' . $allblocks . '>)!', "$1\n\n", $pee);
-	$pee = str_replace(array("\r\n", "\r"), "\n", $pee); // cross-platform newlines
-	if (strpos($pee, '<object') !== false) {
-		$pee = preg_replace('|\s*<param([^>]*)>\s*|', "<param$1>", $pee); // no pee inside object/embed
-		$pee = preg_replace('|\s*</embed>\s*|', '</embed>', $pee);
-	}
-	$pee = preg_replace("/\n\n+/", "\n\n", $pee); // take care of duplicates
-	$pee = preg_replace('/\n?(.+?)(?:\n\s*\n|\z)/s', "<p>$1</p>\n", $pee); // make paragraphs, including one at the end
-	$pee = preg_replace('|<p>\s*?</p>|', '', $pee); // under certain strange conditions it could create a P of entirely whitespace
-	$pee = preg_replace('!<p>([^<]+)\s*?(</(?:div|address|form)[^>]*>)!', "<p>$1</p>$2", $pee);
-	$pee = preg_replace('|<p>|', "$1<p>", $pee);
-	$pee = preg_replace('!<p>\s*(</?' . $allblocks . '[^>]*>)\s*</p>!', "$1", $pee); // don't pee all over a tag
-	$pee = preg_replace("|<p>(<li.+?)</p>|", "$1", $pee); // problem with nested lists
-	$pee = preg_replace('|<p><blockquote([^>]*)>|i', "<blockquote$1><p>", $pee);
-	$pee = str_replace('</blockquote></p>', '</p></blockquote>', $pee);
-	$pee = preg_replace('!<p>\s*(</?' . $allblocks . '[^>]*>)!', "$1", $pee);
-	$pee = preg_replace('!(</?' . $allblocks . '[^>]*>)\s*</p>!', "$1", $pee);
-	if ($br) {
-		$pee = preg_replace_callback('/<(script|style).*?<\/\\1>/s', create_function('$matches', 'return str_replace("\n", "<WPPreserveNewline />", $matches[0]);'), $pee);
-		$pee = preg_replace('|(?<!<br />)\s*\n|', "<br />\n", $pee); // optionally make line breaks
-		$pee = str_replace('<WPPreserveNewline />', "\n", $pee);
-	}
-	$pee = preg_replace('!(</?' . $allblocks . '[^>]*>)\s*<br />!', "$1", $pee);
-	$pee = preg_replace('!<br />(\s*</?(?:p|li|div|dl|dd|dt|th|pre|td|ul|ol)[^>]*>)!', '$1', $pee);
-	//if (strpos($pee, '<pre') !== false) {
-	//	mind the space between the ? and >.  Only there because of the comment.
-	//	$pee = preg_replace_callback('!(<pre.*? >)(.*?)</pre>!is', 'clean_pre', $pee );
-	//}
-	$pee = preg_replace("|\n</p>$|", '</p>', $pee);
-
-	return $pee;
+function autop($pee) {
+	return elgg_autop($pee);
+}
+
+/**
+ * Create paragraphs from text with line spacing
+ *
+ * @param string $pee The string
+ *
+ * @return string
+ **/
+function elgg_autop($pee) {
+	return ElggAutop::getInstance()->process($pee);
 }
 
 /**
@@ -398,3 +372,32 @@ function elgg_strip_tags($string) {
 
 	return $string;
 }
+
+/**
+ * Unit tests for Output
+ *
+ * @param sting  $hook   unit_test
+ * @param string $type   system
+ * @param mixed  $value  Array of tests
+ * @param mixed  $params Params
+ *
+ * @return array
+ * @access private
+ */
+function output_unit_test($hook, $type, $value, $params) {
+	global $CONFIG;
+	$value[] = $CONFIG->path . 'engine/tests/api/output.php';
+	return $value;
+}
+
+/**
+ * Initialise the Output subsystem.
+ *
+ * @return void
+ * @access private
+ */
+function output_init() {
+	elgg_register_plugin_hook_handler('unit_test', 'system', 'output_unit_test');
+}
+
+elgg_register_event_handler('init', 'system', 'output_init');
diff --git a/engine/tests/api/output.php b/engine/tests/api/output.php
new file mode 100644
index 000000000..eb1a66b29
--- /dev/null
+++ b/engine/tests/api/output.php
@@ -0,0 +1,64 @@
+<?php
+/**
+ * Test case for ElggAutop functionality.
+ * @author Steve Clay <steve@mrclay.org>
+ */
+class ElggCoreOutputAutoPTest extends ElggCoreUnitTest {
+
+	/**
+	 * @var ElggAutop
+	 */
+	protected $_autop;
+
+	public function setUp() {
+		$this->_autop = new ElggAutop();
+	}
+	
+	public function testDomRoundtrip()
+	{
+		$d = dir(dirname(__DIR__) . '/test_files/output/autop');
+		$in = file_get_contents($d->path . "/domdoc_in.html");
+		$exp = file_get_contents($d->path . "/domdoc_exp.html");
+
+		$doc = new DOMDocument();
+		libxml_use_internal_errors(true);
+		$doc->loadHTML("<html><meta http-equiv='content-type' content='text/html; charset=utf-8'><body>"
+				. $in . '</body></html>');
+		$serialized = $doc->saveHTML();
+		list(,$out) = explode('<body>', $serialized, 2);
+		list($out) = explode('</body>', $out, 2);
+
+		$this->assertEqual($exp, $out, "DOMDocument's parsing/serialization roundtrip");
+	}
+
+	public function testProcess()
+	{
+		$data = $this->provider();
+		foreach ($data as $row) {
+			list($test, $in, $exp) = $row;
+			$out = $this->_autop->process($in);
+			$this->assertEqual($exp, $out, "Equality case {$test}");
+		}
+	}
+
+	public function provider()
+	{
+		$d = dir(dirname(__DIR__) . '/test_files/output/autop');
+		$tests = array();
+		while (false !== ($entry = $d->read())) {
+			if (preg_match('/^([a-z\\-]+)\.in\.html$/i', $entry, $m)) {
+				$tests[] = $m[1];
+			}
+		}
+
+		$data = array();
+		foreach ($tests as $test) {
+			$data[] = array(
+				$test,
+				file_get_contents($d->path . '/' . "{$test}.in.html"),
+				file_get_contents($d->path . '/' . "{$test}.exp.html"),
+			);
+		}
+		return $data;
+	}
+}
diff --git a/engine/tests/test_files/output/autop/block-a.exp.norun.html b/engine/tests/test_files/output/autop/block-a.exp.norun.html
new file mode 100644
index 000000000..addf29dec
--- /dev/null
+++ b/engine/tests/test_files/output/autop/block-a.exp.norun.html
@@ -0,0 +1,6 @@
+
+<p>HTML5 allows A to contain block-level content</p>
+<a href="foo"><h3>A treated as block</h3>
+<p>Read more</p>
+</a>
+<p><a href="foo">A treated as<br />    inline</a></p>
diff --git a/engine/tests/test_files/output/autop/block-a.in.norun.html b/engine/tests/test_files/output/autop/block-a.in.norun.html
new file mode 100644
index 000000000..fc2dac43a
--- /dev/null
+++ b/engine/tests/test_files/output/autop/block-a.in.norun.html
@@ -0,0 +1,9 @@
+HTML5 allows A to contain block-level content
+<a href="foo">
+
+    <h3>A treated as block</h3>
+
+    Read more
+</a>
+<a href="foo">A treated as
+    inline</a>
diff --git a/engine/tests/test_files/output/autop/domdoc_exp.html b/engine/tests/test_files/output/autop/domdoc_exp.html
new file mode 100644
index 000000000..8480c1083
--- /dev/null
+++ b/engine/tests/test_files/output/autop/domdoc_exp.html
@@ -0,0 +1,46 @@
+›
+ 
+Vietnamese - Tiếng Việt
+
+<h1>h1</h1>
+<p>Paragraph <a href="http://google.com/">link</a> <strong>Bold</strong> <em>italic</em> <em><strong>bolditalic</strong></em> <span style="background-color: #ffff00; "></span></p>
+<h2>h2</h2>
+<p>Paragraph <span style="font-size: xx-small;">size1</span> <span style="font-size: x-small;">size2</span> <span style="font-size: medium;">size4</span></p>
+<h3>h3</h3>
+<p>Paragraph <span style="text-decoration: underline;">underline</span> <span style="text-decoration: line-through;">strikethrough</span> <span style="color: #ff0000;">color</span> <span style="background-color: #ffff00; ">background</span></p>
+<blockquote>
+    <p>Blockquoted paragraph</p>
+</blockquote>
+<p>Paragraph following blockquote</p>
+<ul><li>Unordered</li>
+    <li>List</li>
+</ul><p>Paragraph between lists</p>
+<ol><li>Ordered</li>
+    <li>List</li>
+</ol><p>Paragraph between lists</p>
+<ul><li>OL list</li>
+    <li>nested<ol><li>inside a</li>
+            <li>UL list</li>
+        </ol></li>
+</ul><p>Paragraph between lists</p>
+<table border="0"><tbody><tr><td>Table with</td>
+            <td></td>
+        </tr><tr><td></td>
+            <td>border=0</td>
+        </tr></tbody></table><p>Paragraph</p>
+<ol><li>UL list</li>
+    <li>nested
+        <ul><li>inside a</li>
+            <li>OL list</li>
+        </ul></li>
+</ol><p>Paragraph between tables</p>
+<table border="1" cellpadding="5"><tbody><tr><td>Table with border=1</td>
+            <td></td>
+        </tr><tr><td></td>
+            <td>cellpadding = 5</td>
+        </tr></tbody></table><p>Paragraph between tables</p>
+<table border="2"><tbody><tr><td>Table with</td>
+            <td></td>
+        </tr><tr><td></td>
+            <td>border=2</td>
+        </tr></tbody></table>
+\ No newline at end of file
diff --git a/engine/tests/test_files/output/autop/domdoc_in.html b/engine/tests/test_files/output/autop/domdoc_in.html
new file mode 100644
index 000000000..4c465b435
--- /dev/null
+++ b/engine/tests/test_files/output/autop/domdoc_in.html
@@ -0,0 +1,80 @@
+&#8250;
+&nbsp;
+Vietnamese - Tiếng Việt
+
+<h1>h1</h1>
+<p>Paragraph <a href="http://google.com/">link</a> <strong>Bold</strong> <em>italic</em> <em><strong>bolditalic</strong></em>&nbsp;<span style="background-color: #ffff00; "></span></p>
+<h2>h2</h2>
+<p>Paragraph <span style="font-size: xx-small;">size1</span> <span style="font-size: x-small;">size2</span> <span style="font-size: medium;">size4</span></p>
+<h3>h3</h3>
+<p>Paragraph <span style="text-decoration: underline;">underline</span> <span style="text-decoration: line-through;">strikethrough</span> <span style="color: #ff0000;">color</span> <span style="background-color: #ffff00; ">background</span></p>
+<blockquote>
+    <p>Blockquoted paragraph</p>
+</blockquote>
+<p>Paragraph following blockquote</p>
+<ul>
+    <li>Unordered</li>
+    <li>List</li>
+</ul>
+<p>Paragraph between lists</p>
+<ol>
+    <li>Ordered</li>
+    <li>List</li>
+</ol>
+<p>Paragraph between lists</p>
+<ul>
+    <li>OL list</li>
+    <li>nested<ol>
+            <li>inside a</li>
+            <li>UL list</li>
+        </ol></li>
+</ul>
+<p>Paragraph between lists</p>
+<table border="0">
+    <tbody>
+        <tr>
+            <td>Table with</td>
+            <td></td>
+        </tr>
+        <tr>
+            <td></td>
+            <td>border=0</td>
+        </tr>
+    </tbody>
+</table>
+<p>Paragraph</p>
+<ol>
+    <li>UL list</li>
+    <li>nested
+        <ul>
+            <li>inside a</li>
+            <li>OL list</li>
+        </ul>
+    </li>
+</ol>
+<p>Paragraph between tables</p>
+<table border="1" cellpadding="5">
+    <tbody>
+        <tr>
+            <td>Table with border=1</td>
+            <td></td>
+        </tr>
+        <tr>
+            <td></td>
+            <td>cellpadding = 5</td>
+        </tr>
+    </tbody>
+</table>
+<p>Paragraph between tables</p>
+<table border="2">
+    <tbody>
+        <tr>
+            <td>Table with</td>
+            <td></td>
+        </tr>
+        <tr>
+            <td></td>
+            <td>border=2</td>
+        </tr>
+    </tbody>
+</table>
+\ No newline at end of file
diff --git a/engine/tests/test_files/output/autop/typical-post.exp.html b/engine/tests/test_files/output/autop/typical-post.exp.html
new file mode 100644
index 000000000..f9d75a114
--- /dev/null
+++ b/engine/tests/test_files/output/autop/typical-post.exp.html
@@ -0,0 +1,84 @@
+<h2>Lorem ipsum dolor sit amet, consectetur adipiscing elit.</h2>
+<p><img class="alignright size-thumbnail wp-image-905" title="Surest Things mixing session in Adobe Audition" src="http://www.mrclay.org/wp-content/uploads/2010/09/surestThings_audition-150x150.png" alt="screenshot of Audition mixing session" width="150" height="150">Vivamus enim ante, <em>mattis eget imperdiet nec, pharetra vel velit.</em> Sed at euismod nibh. Praesent lacus tellus, <a href="http://google.com/">posuere et convallis</a> a, <strong>mollis et tellus. Suspendisse potenti</strong>. Phasellus tincidunt dignissim est eget mattis. Vestibulum lacinia <del>condimentum tellus, non vestibulum erat dapibus</del> quis. Aliquam arcu nibh, viverra adipiscing eleifend quis, pretium vitae ipsum.</p>
+
+<p>Curabitur turpis ante, <span style="color: #993300;">congue ac dapibus quis, vehicula ac orci.</span> Nunc luctus neque non massa porta sed pharetra ante accumsan. <a href="http://google.com/">Nam suscipit</a> risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor.</p>
+<h3>Donec at massa ante, sagittis fermentum urna.</h3><blockquote>
+<p>Mauris volutpat est id massa volutpat lacinia. Class aptent taciti sociosqu ad litora torquent per conubia nostra, per inceptos himenaeos. In in nisl mauris. In aliquet pretium nisl, vel convallis neque cursus vitae. Curabitur id mauris in urna gravida ornare.</p>
+
+<p>[caption id="attachment_719" align="alignleft" width="150" caption="Ibanez AGB140 Bass"]<img class="size-thumbnail wp-image-719" title="Ibanez AGB140 Bass" src="http://www.mrclay.org/wp-content/uploads/2010/04/agb140-e1271773766573-150x150.jpg" alt="Ibanez AGB140 Bass" width="150" height="150">[/caption]</p>
+
+<p>Aenean <a href="http://google.com/">aliquet cursus purus sed gravida. Cras auctor euismod justo, ac dictum purus facilisis dignissim.</a> Quisque facilisis porta sem, ac suscipit quam molestie nec. Pellentesque quis hendrerit enim. Vivamus tempor erat diam. Sed eu felis nunc. Cras posuere lorem commodo turpis mollis sagittis. Mauris lobortis nunc felis.</p>
+
+<p>Maecenas elit lorem, varius sed condimentum ac, cursus et magna. Nam ut massa id augue consectetur porttitor eleifend in nunc. Curabitur cursus varius dictum. Vestibulum vel justo et neque tempus placerat a vel sapien.</p>
+</blockquote>
+<p>Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vivamus enim ante, mattis eget imperdiet nec, <a href="http://google.com/">pharetra </a>vel velit. Sed at euismod nibh. Praesent lacus tellus, posuere et convallis a, mollis et tellus.</p>
+<pre><code>&lt;?php
+class DataTest extends PHPUnit_Framework_TestCase
+{
+    /**
+     * @dataProvider provider
+     */
+    public function testAdd($a, $b, $c)
+    {
+        $this-&gt;assertEquals($c, $a + $b);
+    }
+
+    public function provider()
+    {
+        return array(
+          array(0, 0, 0),
+          array(0, 1, 1),
+          array(1, 0, 1),
+          array(1, 1, 3)
+        );
+    }
+}</code></pre><ul><li>Suspendisse potenti. Phasellus tincidunt dignissim est eget mattis.</li>
+	<li>Vestibulum lacinia condimentum tellus, non vestibulum erat dapibus quis.</li>
+	<li>Aliquam arcu nibh, <a href="http://google.com/">viverra</a> adipiscing eleifend quis, pretium vitae ipsum.</li>
+	<li>Curabitur turpis ante, congue ac <a href="http://google.com/">dapibus quis</a>, vehicula ac orci.</li>
+</ul>
+<p>Nunc luctus neque non massa porta sed pharetra ante accumsan. Nam suscipit risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor. Donec at massa ante, sagittis fermentum urna.</p>
+
+<p><object width="480" height="390"><param name="movie" value="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US"><param name="allowFullScreen" value="true"><param name="allowscriptaccess" value="always"><embed type="application/x-shockwave-flash" width="480" height="390" src="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US" allowfullscreen="true" allowscriptaccess="always"></embed></object></p>
+<h2>Lorem ipsum dolor sit amet, consectetur adipiscing elit.</h2>
+<p><img class="alignright size-thumbnail wp-image-905" title="Surest Things mixing session in Adobe Audition" src="http://www.mrclay.org/wp-content/uploads/2010/09/surestThings_audition-150x150.png" alt="screenshot of Audition mixing session" width="150" height="150">Vivamus enim ante, <em>mattis eget imperdiet nec, pharetra vel velit.</em> Sed at euismod nibh. Praesent lacus tellus, <a href="http://google.com/">posuere et convallis</a> a, <strong>mollis et tellus. Suspendisse potenti</strong>. Phasellus tincidunt dignissim est eget mattis. Vestibulum lacinia <del>condimentum tellus, non vestibulum erat dapibus</del> quis. Aliquam arcu nibh, viverra adipiscing eleifend quis, pretium vitae ipsum.</p>
+
+<p>Curabitur turpis ante, <span style="color: #993300;">congue ac dapibus quis, vehicula ac orci.</span> Nunc luctus neque non massa porta sed pharetra ante accumsan. <a href="http://google.com/">Nam suscipit</a> risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor.</p>
+<h3>Donec at massa ante, sagittis fermentum urna.</h3><blockquote>
+<p>Mauris volutpat est id massa volutpat lacinia. Class aptent taciti sociosqu ad litora torquent per conubia nostra, per inceptos himenaeos. In in nisl mauris. In aliquet pretium nisl, vel convallis neque cursus vitae. Curabitur id mauris in urna gravida ornare.</p>
+
+<p>[caption id="attachment_719" align="alignleft" width="150" caption="Ibanez AGB140 Bass"]<img class="size-thumbnail wp-image-719" title="Ibanez AGB140 Bass" src="http://www.mrclay.org/wp-content/uploads/2010/04/agb140-e1271773766573-150x150.jpg" alt="Ibanez AGB140 Bass" width="150" height="150">[/caption]</p>
+
+<p>Aenean <a href="http://google.com/">aliquet cursus purus sed gravida. Cras auctor euismod justo, ac dictum purus facilisis dignissim.</a> Quisque facilisis porta sem, ac suscipit quam molestie nec. Pellentesque quis hendrerit enim. Vivamus tempor erat diam. Sed eu felis nunc. Cras posuere lorem commodo turpis mollis sagittis. Mauris lobortis nunc felis.</p>
+
+<p>Maecenas elit lorem, varius sed condimentum ac, cursus et magna. Nam ut massa id augue consectetur porttitor eleifend in nunc. Curabitur cursus varius dictum. Vestibulum vel justo et neque tempus placerat a vel sapien.</p>
+</blockquote>
+<p>Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vivamus enim ante, mattis eget imperdiet nec, <a href="http://google.com/">pharetra </a>vel velit. Sed at euismod nibh. Praesent lacus tellus, posuere et convallis a, mollis et tellus.</p>
+<pre><code>&lt;?php
+class DataTest extends PHPUnit_Framework_TestCase
+{
+    /**
+     * @dataProvider provider
+     */
+    public function testAdd($a, $b, $c)
+    {
+        $this-&gt;assertEquals($c, $a + $b);
+    }
+
+    public function provider()
+    {
+        return array(
+          array(0, 0, 0),
+          array(0, 1, 1),
+          array(1, 0, 1),
+          array(1, 1, 3)
+        );
+    }
+}</code></pre><ul><li>Suspendisse potenti. Phasellus tincidunt dignissim est eget mattis.</li>
+	<li>Vestibulum lacinia condimentum tellus, non vestibulum erat dapibus quis.</li>
+	<li>Aliquam arcu nibh, <a href="http://google.com/">viverra</a> adipiscing eleifend quis, pretium vitae ipsum.</li>
+	<li>Curabitur turpis ante, congue ac <a href="http://google.com/">dapibus quis</a>, vehicula ac orci.</li>
+</ul>
+<p>Nunc luctus neque non massa porta sed pharetra ante accumsan. Nam suscipit risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor. Donec at massa ante, sagittis fermentum urna.</p>
+
+<p><object width="480" height="390"><param name="movie" value="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US"><param name="allowFullScreen" value="true"><param name="allowscriptaccess" value="always"><embed type="application/x-shockwave-flash" width="480" height="390" src="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US" allowfullscreen="true" allowscriptaccess="always"></embed></object></p>
diff --git a/engine/tests/test_files/output/autop/typical-post.in.html b/engine/tests/test_files/output/autop/typical-post.in.html
new file mode 100644
index 000000000..6e4984cc4
--- /dev/null
+++ b/engine/tests/test_files/output/autop/typical-post.in.html
@@ -0,0 +1,89 @@
+<h2>Lorem ipsum dolor sit amet, consectetur adipiscing elit.</h2>
+<img class="alignright size-thumbnail wp-image-905" title="Surest Things mixing session in Adobe Audition" src="http://www.mrclay.org/wp-content/uploads/2010/09/surestThings_audition-150x150.png" alt="screenshot of Audition mixing session" width="150" height="150" />Vivamus enim ante, <em>mattis eget imperdiet nec, pharetra vel velit.</em> Sed at euismod nibh. Praesent lacus tellus, <a href="http://google.com/">posuere et convallis</a> a, <strong>mollis et tellus. Suspendisse potenti</strong>. Phasellus tincidunt dignissim est eget mattis. Vestibulum lacinia <del>condimentum tellus, non vestibulum erat dapibus</del> quis. Aliquam arcu nibh, viverra adipiscing eleifend quis, pretium vitae ipsum.
+
+Curabitur turpis ante, <span style="color: #993300;">congue ac dapibus quis, vehicula ac orci.</span> Nunc luctus neque non massa porta sed pharetra ante accumsan. <a href="http://google.com/">Nam suscipit</a> risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor.
+<h3>Donec at massa ante, sagittis fermentum urna.</h3>
+<blockquote>Mauris volutpat est id massa volutpat lacinia. Class aptent taciti sociosqu ad litora torquent per conubia nostra, per inceptos himenaeos. In in nisl mauris. In aliquet pretium nisl, vel convallis neque cursus vitae. Curabitur id mauris in urna gravida ornare.
+
+[caption id="attachment_719" align="alignleft" width="150" caption="Ibanez AGB140 Bass"]<img class="size-thumbnail wp-image-719" title="Ibanez AGB140 Bass" src="http://www.mrclay.org/wp-content/uploads/2010/04/agb140-e1271773766573-150x150.jpg" alt="Ibanez AGB140 Bass" width="150" height="150" />[/caption]
+
+Aenean <a href="http://google.com/">aliquet cursus purus sed gravida. Cras auctor euismod justo, ac dictum purus facilisis dignissim.</a> Quisque facilisis porta sem, ac suscipit quam molestie nec. Pellentesque quis hendrerit enim. Vivamus tempor erat diam. Sed eu felis nunc. Cras posuere lorem commodo turpis mollis sagittis. Mauris lobortis nunc felis.
+
+Maecenas elit lorem, varius sed condimentum ac, cursus et magna. Nam ut massa id augue consectetur porttitor eleifend in nunc. Curabitur cursus varius dictum. Vestibulum vel justo et neque tempus placerat a vel sapien.</blockquote>
+Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vivamus enim ante, mattis eget imperdiet nec, <a href="http://google.com/">pharetra </a>vel velit. Sed at euismod nibh. Praesent lacus tellus, posuere et convallis a, mollis et tellus.
+
+<pre><code>&lt;?php
+class DataTest extends PHPUnit_Framework_TestCase
+{
+    /**
+     * @dataProvider provider
+     */
+    public function testAdd($a, $b, $c)
+    {
+        $this-&gt;assertEquals($c, $a + $b);
+    }
+
+    public function provider()
+    {
+        return array(
+          array(0, 0, 0),
+          array(0, 1, 1),
+          array(1, 0, 1),
+          array(1, 1, 3)
+        );
+    }
+}</code></pre>
+<ul>
+	<li>Suspendisse potenti. Phasellus tincidunt dignissim est eget mattis.</li>
+	<li>Vestibulum lacinia condimentum tellus, non vestibulum erat dapibus quis.</li>
+	<li>Aliquam arcu nibh, <a href="http://google.com/">viverra</a> adipiscing eleifend quis, pretium vitae ipsum.</li>
+	<li>Curabitur turpis ante, congue ac <a href="http://google.com/">dapibus quis</a>, vehicula ac orci.</li>
+</ul>
+Nunc luctus neque non massa porta sed pharetra ante accumsan. Nam suscipit risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor. Donec at massa ante, sagittis fermentum urna.
+
+<object width="480" height="390"><param name="movie" value="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US" /><param name="allowFullScreen" value="true" /><param name="allowscriptaccess" value="always" /><embed type="application/x-shockwave-flash" width="480" height="390" src="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US" allowfullscreen="true" allowscriptaccess="always"></embed></object>
+
+<h2>Lorem ipsum dolor sit amet, consectetur adipiscing elit.</h2>
+<img class="alignright size-thumbnail wp-image-905" title="Surest Things mixing session in Adobe Audition" src="http://www.mrclay.org/wp-content/uploads/2010/09/surestThings_audition-150x150.png" alt="screenshot of Audition mixing session" width="150" height="150" />Vivamus enim ante, <em>mattis eget imperdiet nec, pharetra vel velit.</em> Sed at euismod nibh. Praesent lacus tellus, <a href="http://google.com/">posuere et convallis</a> a, <strong>mollis et tellus. Suspendisse potenti</strong>. Phasellus tincidunt dignissim est eget mattis. Vestibulum lacinia <del>condimentum tellus, non vestibulum erat dapibus</del> quis. Aliquam arcu nibh, viverra adipiscing eleifend quis, pretium vitae ipsum.
+
+Curabitur turpis ante, <span style="color: #993300;">congue ac dapibus quis, vehicula ac orci.</span> Nunc luctus neque non massa porta sed pharetra ante accumsan. <a href="http://google.com/">Nam suscipit</a> risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor.
+<h3>Donec at massa ante, sagittis fermentum urna.</h3>
+<blockquote>Mauris volutpat est id massa volutpat lacinia. Class aptent taciti sociosqu ad litora torquent per conubia nostra, per inceptos himenaeos. In in nisl mauris. In aliquet pretium nisl, vel convallis neque cursus vitae. Curabitur id mauris in urna gravida ornare.
+
+[caption id="attachment_719" align="alignleft" width="150" caption="Ibanez AGB140 Bass"]<img class="size-thumbnail wp-image-719" title="Ibanez AGB140 Bass" src="http://www.mrclay.org/wp-content/uploads/2010/04/agb140-e1271773766573-150x150.jpg" alt="Ibanez AGB140 Bass" width="150" height="150" />[/caption]
+
+Aenean <a href="http://google.com/">aliquet cursus purus sed gravida. Cras auctor euismod justo, ac dictum purus facilisis dignissim.</a> Quisque facilisis porta sem, ac suscipit quam molestie nec. Pellentesque quis hendrerit enim. Vivamus tempor erat diam. Sed eu felis nunc. Cras posuere lorem commodo turpis mollis sagittis. Mauris lobortis nunc felis.
+
+Maecenas elit lorem, varius sed condimentum ac, cursus et magna. Nam ut massa id augue consectetur porttitor eleifend in nunc. Curabitur cursus varius dictum. Vestibulum vel justo et neque tempus placerat a vel sapien.</blockquote>
+Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vivamus enim ante, mattis eget imperdiet nec, <a href="http://google.com/">pharetra </a>vel velit. Sed at euismod nibh. Praesent lacus tellus, posuere et convallis a, mollis et tellus.
+
+<pre><code>&lt;?php
+class DataTest extends PHPUnit_Framework_TestCase
+{
+    /**
+     * @dataProvider provider
+     */
+    public function testAdd($a, $b, $c)
+    {
+        $this-&gt;assertEquals($c, $a + $b);
+    }
+
+    public function provider()
+    {
+        return array(
+          array(0, 0, 0),
+          array(0, 1, 1),
+          array(1, 0, 1),
+          array(1, 1, 3)
+        );
+    }
+}</code></pre>
+<ul>
+	<li>Suspendisse potenti. Phasellus tincidunt dignissim est eget mattis.</li>
+	<li>Vestibulum lacinia condimentum tellus, non vestibulum erat dapibus quis.</li>
+	<li>Aliquam arcu nibh, <a href="http://google.com/">viverra</a> adipiscing eleifend quis, pretium vitae ipsum.</li>
+	<li>Curabitur turpis ante, congue ac <a href="http://google.com/">dapibus quis</a>, vehicula ac orci.</li>
+</ul>
+Nunc luctus neque non massa porta sed pharetra ante accumsan. Nam suscipit risus quis libero convallis viverra. Ut at arcu enim, vel pharetra dolor. Donec at massa ante, sagittis fermentum urna.
+
+<object width="480" height="390"><param name="movie" value="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US" /><param name="allowFullScreen" value="true" /><param name="allowscriptaccess" value="always" /><embed type="application/x-shockwave-flash" width="480" height="390" src="http://www.youtube.com/v/zW9YOMaVTFI?fs=1&amp;hl=en_US" allowfullscreen="true" allowscriptaccess="always"></embed></object>
+\ No newline at end of file
diff --git a/engine/tests/test_files/output/autop/wp-welcome.exp.html b/engine/tests/test_files/output/autop/wp-welcome.exp.html
new file mode 100644
index 000000000..2f612e3dd
--- /dev/null
+++ b/engine/tests/test_files/output/autop/wp-welcome.exp.html
@@ -0,0 +1,22 @@
+
+<p>Welcome to WordPress!  This post contains important information.  After you read it, you can make it private to hide it from visitors but still have the information handy for future reference.</p>
+
+<p>First things first:</p>
+<ul><li><a href="%1%24s" title="Subscribe to the WordPress mailing list for Release Notifications">Subscribe to the WordPress mailing list for release notifications</a></li>
+</ul>
+<p>As a subscriber, you will receive an email every time an update is available (and only then).  This will make it easier to keep your site up to date, and secure from evildoers.<br />When a new version is released, <a href="%2%24s" title="If you are already logged in, this will take you directly to the Dashboard">log in to the Dashboard</a> and follow the instructions.<br />Upgrading is a couple of clicks!</p>
+
+<p>Then you can start enjoying the WordPress experience:</p>
+<ul><li>Edit your personal information at <a href="%3%24s" title="Edit settings like your password, your display name and your contact information">Users &#8250; Your Profile</a></li>
+    <li>Start publishing at <a href="%4%24s" title="Create a new post">Posts &#8250; Add New</a> and at <a href="%5%24s" title="Create a new page">Pages &#8250; Add New</a></li>
+    <li>Browse and install plugins at <a href="%6%24s" title="Browse and install plugins at the official WordPress repository directly from your Dashboard">Plugins &#8250; Add New</a></li>
+    <li>Browse and install themes at <a href="%7%24s" title="Browse and install themes at the official WordPress repository directly from your Dashboard">Appearance &#8250; Add New Themes</a></li>
+    <li>Modify and prettify your website&#8217;s links at <a href="%8%24s" title="For example, select a link structure like: http://example.com/1999/12/post-name">Settings &#8250; Permalinks</a></li>
+    <li>Import content from another system or WordPress site at <a href="%9%24s" title="WordPress comes with importers for the most common publishing systems">Tools &#8250; Import</a></li>
+    <li>Find answers to your questions at the <a href="%10%24s" title="The official WordPress documentation, maintained by the WordPress community">WordPress Codex</a></li>
+</ul>
+<p>To keep this post for reference, <a href="%11%24s" title="Click to edit the content and settings of this post">click to edit it</a>, go to the Publish box and change its Visibility from Public to Private.</p>
+
+<p>Thank you for selecting WordPress.  We wish you happy publishing!</p>
+
+<p>PS.  Not yet subscribed for update notifications?  <a href="%1%24s" title="Subscribe to the WordPress mailing list for Release Notifications">Do it now!</a></p>
diff --git a/engine/tests/test_files/output/autop/wp-welcome.in.html b/engine/tests/test_files/output/autop/wp-welcome.in.html
new file mode 100644
index 000000000..338ede73f
--- /dev/null
+++ b/engine/tests/test_files/output/autop/wp-welcome.in.html
@@ -0,0 +1,25 @@
+Welcome to WordPress!  This post contains important information.  After you read it, you can make it private to hide it from visitors but still have the information handy for future reference.
+
+First things first:
+<ul>
+    <li><a href="%1$s" title="Subscribe to the WordPress mailing list for Release Notifications">Subscribe to the WordPress mailing list for release notifications</a></li>
+</ul>
+As a subscriber, you will receive an email every time an update is available (and only then).  This will make it easier to keep your site up to date, and secure from evildoers.
+When a new version is released, <a href="%2$s" title="If you are already logged in, this will take you directly to the Dashboard">log in to the Dashboard</a> and follow the instructions.
+Upgrading is a couple of clicks!
+
+Then you can start enjoying the WordPress experience:
+<ul>
+    <li>Edit your personal information at <a href="%3$s" title="Edit settings like your password, your display name and your contact information">Users &#8250; Your Profile</a></li>
+    <li>Start publishing at <a href="%4$s" title="Create a new post">Posts &#8250; Add New</a> and at <a href="%5$s" title="Create a new page">Pages &#8250; Add New</a></li>
+    <li>Browse and install plugins at <a href="%6$s" title="Browse and install plugins at the official WordPress repository directly from your Dashboard">Plugins &#8250; Add New</a></li>
+    <li>Browse and install themes at <a href="%7$s" title="Browse and install themes at the official WordPress repository directly from your Dashboard">Appearance &#8250; Add New Themes</a></li>
+    <li>Modify and prettify your website&#8217;s links at <a href="%8$s" title="For example, select a link structure like: http://example.com/1999/12/post-name">Settings &#8250; Permalinks</a></li>
+    <li>Import content from another system or WordPress site at <a href="%9$s" title="WordPress comes with importers for the most common publishing systems">Tools &#8250; Import</a></li>
+    <li>Find answers to your questions at the <a href="%10$s" title="The official WordPress documentation, maintained by the WordPress community">WordPress Codex</a></li>
+</ul>
+To keep this post for reference, <a href="%11$s" title="Click to edit the content and settings of this post">click to edit it</a>, go to the Publish box and change its Visibility from Public to Private.
+
+Thank you for selecting WordPress.  We wish you happy publishing!
+
+PS.  Not yet subscribed for update notifications?  <a href="%1$s" title="Subscribe to the WordPress mailing list for Release Notifications">Do it now!</a>
diff --git a/engine/tests/test_files/output/autop/wpautop-fails.exp.html b/engine/tests/test_files/output/autop/wpautop-fails.exp.html
new file mode 100644
index 000000000..d018db4ff
--- /dev/null
+++ b/engine/tests/test_files/output/autop/wpautop-fails.exp.html
@@ -0,0 +1,31 @@
+
+<p>paragraph</p>
+
+<p>paragraph</p>
+<div class="whatever"><blockquote>
+<p>paragraph</p>
+</blockquote>
+<p>line</p>
+</div>
+<p>paragraph</p>
+<ul><li>line</li>
+<li>paragraph
+
+paragraph</li>
+</ul>
+<p>paragraph<br />line<br />line</p>
+<pre>Honor
+this whitespace
+</pre>
+<p>paragraph</p>
+<style><!--
+Do not alter!
+--></style>
+<p>paragraph <!-- do not alter --></p>
+<dl><dt>term</dt> <dd>paragraph
+
+<a href="xx"> <img src="yy"></a>
+
+paragraph</dd> </dl><div><a href="xx"> <img src="yy"></a></div>
+<p>Hello <a href="link"><br /><br />World</a></p>
+<p id="abc">Paragraph</p><div>Line</div>
+\ No newline at end of file
diff --git a/engine/tests/test_files/output/autop/wpautop-fails.in.html b/engine/tests/test_files/output/autop/wpautop-fails.in.html
new file mode 100644
index 000000000..9aa24be59
--- /dev/null
+++ b/engine/tests/test_files/output/autop/wpautop-fails.in.html
@@ -0,0 +1,41 @@
+
+paragraph
+
+paragraph    <div class="whatever"><blockquote>
+    paragraph
+  </blockquote>
+  line
+</div>
+
+paragraph
+<ul>
+<li>line</li>
+<li>paragraph
+
+paragraph</li>
+</ul>
+paragraph
+line<br>    
+  line
+<pre>Honor
+this whitespace
+</pre>
+paragraph
+<style><!--
+Do not alter!
+--></style>
+paragraph <!-- do not alter -->
+<dl> <dt>term</dt> <dd>paragraph
+
+<a href="xx"> <img src="yy" /> </a>
+
+paragraph</dd> </dl>
+<div><a href="xx"> <img src="yy" /> </a></div>
+
+Hello <a href="link">
+
+World</a>
+
+<p id="abc">Paragraph</p>
+
+<div>Line</div>
+\ No newline at end of file
diff --git a/engine/tests/test_files/output/autop/wysiwyg-test.exp.html b/engine/tests/test_files/output/autop/wysiwyg-test.exp.html
new file mode 100644
index 000000000..1f23d6154
--- /dev/null
+++ b/engine/tests/test_files/output/autop/wysiwyg-test.exp.html
@@ -0,0 +1,51 @@
+
+<p>&nbps;<br />&#8820;</p>
+<h1>h1</h1>
+<p>Paragraph <a href="http://google.com/">link</a> <strong>Bold</strong> <em>italic</em> <em><strong>bolditalic</strong></em>&nbsp;<span style="background-color: #ffff00; "></span></p>
+<h2>h2</h2>
+<p>Paragraph <span style="font-size: xx-small;">size1</span> <span style="font-size: x-small;">size2</span> <span style="font-size: medium;">size4</span></p>
+<h3>h3</h3>
+<p>Paragraph <span style="text-decoration: underline;">underline</span> <span style="text-decoration: line-through;">strikethrough</span> <span style="color: #ff0000;">color</span> <span style="background-color: #ffff00; ">background</span></p>
+<blockquote>
+<p>Blockquoted paragraph</p>
+</blockquote>
+<p>Paragraph following blockquote</p>
+<ul><li>Unordered</li>
+    <li>List</li>
+</ul>
+<p>Paragraph between lists</p>
+<ol><li>Ordered</li>
+    <li>List</li>
+</ol>
+<p>Paragraph between lists</p>
+<ul><li>OL list</li>
+    <li>nested
+        <ol><li>inside a</li>
+            <li>UL list</li>
+        </ol></li>
+</ul>
+<p>Paragraph between lists</p>
+<table border="0"><tbody><tr></tr><tr><td>Table with</td>
+            <td></td>
+        </tr><tr><td></td>
+            <td>border=0</td>
+        </tr></tbody></table>
+<p>Paragraph</p>
+<ol><li>UL list</li>
+    <li>nested
+        <ul><li>inside a</li>
+            <li>OL list</li>
+        </ul></li>
+</ol>
+<p>Paragraph between tables</p>
+<table border="1" cellpadding="5"><tbody><tr><td>Table with border=1</td>
+            <td></td>
+        </tr><tr><td></td>
+            <td>cellpadding = 5</td>
+        </tr></tbody></table>
+<p>Paragraph between tables</p>
+<table border="2"><tbody><tr><td>Table with</td>
+            <td></td>
+        </tr><tr><td></td>
+            <td>border=2</td>
+        </tr></tbody></table>
+\ No newline at end of file
diff --git a/engine/tests/test_files/output/autop/wysiwyg-test.in.html b/engine/tests/test_files/output/autop/wysiwyg-test.in.html
new file mode 100644
index 000000000..733b0e2ec
--- /dev/null
+++ b/engine/tests/test_files/output/autop/wysiwyg-test.in.html
@@ -0,0 +1,79 @@
+&nbps;
+&#8820;
+<h1>h1</h1>
+Paragraph <a href="http://google.com/">link</a> <strong>Bold</strong> <em>italic</em> <em><strong>bolditalic</strong></em>&nbsp;<span style="background-color: #ffff00; "></span>
+<h2>h2</h2>
+Paragraph <span style="font-size: xx-small;">size1</span> <span style="font-size: x-small;">size2</span> <span style="font-size: medium;">size4</span>
+<h3>h3</h3>
+Paragraph <span style="text-decoration: underline;">underline</span> <span style="text-decoration: line-through;">strikethrough</span> <span style="color: #ff0000;">color</span> <span style="background-color: #ffff00; ">background</span>
+<blockquote>Blockquoted paragraph</blockquote>
+Paragraph following blockquote
+<ul>
+    <li>Unordered</li>
+    <li>List</li>
+</ul>
+Paragraph between lists
+<ol>
+    <li>Ordered</li>
+    <li>List</li>
+</ol>
+Paragraph between lists
+<ul>
+    <li>OL list</li>
+    <li>nested
+        <ol>
+            <li>inside a</li>
+            <li>UL list</li>
+        </ol></li>
+</ul>
+Paragraph between lists
+<table border="0">
+    <tbody>
+        <tr>
+        </tr>
+        <tr>
+            <td>Table with</td>
+            <td></td>
+        </tr>
+        <tr>
+            <td></td>
+            <td>border=0</td>
+        </tr>
+    </tbody>
+</table>
+Paragraph
+<ol>
+    <li>UL list</li>
+    <li>nested
+        <ul>
+            <li>inside a</li>
+            <li>OL list</li>
+        </ul>
+    </li>
+</ol>
+Paragraph between tables
+<table border="1" cellpadding="5">
+    <tbody>
+        <tr>
+            <td>Table with border=1</td>
+            <td></td>
+        </tr>
+        <tr>
+            <td></td>
+            <td>cellpadding = 5</td>
+        </tr>
+    </tbody>
+</table>
+Paragraph between tables
+<table border="2">
+    <tbody>
+        <tr>
+            <td>Table with</td>
+            <td></td>
+        </tr>
+        <tr>
+            <td></td>
+            <td>border=2</td>
+        </tr>
+    </tbody>
+</table>
+\ No newline at end of file