setters should know nothing

**strong** and *em* and **strong** and *em*

.travis.yml

@@ -4,4 +4,5 @@ php:
   - 5.5
   - 5.4
   - 5.3
-  - 5.2
+  - 5.2
+  - hhvm

Parsedown.php

@@ -33,12 +33,24 @@ class Parsedown
 
 	private static $instances = array();
 
+	#
+	# Setters
+	#
+
+	private $breaks_enabled = false;
+
+	function set_breaks_enabled($breaks_enabled)
+	{
+		$this->breaks_enabled = $breaks_enabled;
+
+		return $this;
+	}
+
 	#
 	# Fields
 	#
 
 	private $reference_map = array();
-	private $escape_sequence_map = array();
 
 	#
 	# Public Methods
@@ -46,9 +58,6 @@ class Parsedown
 
 	function parse($text)
 	{
-		# removes UTF-8 BOM and marker characters
-		$text = preg_replace('{^\xEF\xBB\xBF|\x1A}', '', $text);
-
 		# removes \r characters
 		$text = str_replace("\r\n", "\n", $text);
 		$text = str_replace("\r", "\n", $text);
@@ -56,41 +65,14 @@ class Parsedown
 		# replaces tabs with spaces
 		$text = str_replace("\t", '    ', $text);
 
-		# encodes escape sequences
-
-		if (strpos($text, '\\') !== FALSE)
-		{
-			$escape_sequences = array('\\\\', '\`', '\*', '\_', '\{', '\}', '\[', '\]', '\(', '\)', '\>', '\#', '\+', '\-', '\.', '\!');
-
-			foreach ($escape_sequences as $index => $escape_sequence)
-			{
-				if (strpos($text, $escape_sequence) !== FALSE)
-				{
-					$code = "\x1A".'\\'.$index.';';
-
-					$text = str_replace($escape_sequence, $code, $text);
-
-					$this->escape_sequence_map[$code] = $escape_sequence;
-				}
-			}
-		}
-
 		# ~
 
-		$text = preg_replace('/\n\s*\n/', "\n\n", $text);
 		$text = trim($text, "\n");
 
 		$lines = explode("\n", $text);
 
 		$text = $this->parse_block_elements($lines);
 
-		# decodes escape sequences
-
-		foreach ($this->escape_sequence_map as $code => $escape_sequence)
-		{
-			$text = str_replace($code, $escape_sequence[1], $text);
-		}
-
 		$text = rtrim($text, "\n");
 
 		return $text;
@@ -110,12 +92,11 @@ class Parsedown
 
 		foreach ($lines as $line)
 		{
-			#
 			# fenced elements
 
 			switch ($element['type'])
 			{
-				case 'fenced_code_block':
+				case 'fenced block':
 
 					if ( ! isset($element['closed']))
 					{
@@ -135,16 +116,16 @@ class Parsedown
 
 					break;
 
-				case 'markup':
+				case 'block-level markup':
 
 					if ( ! isset($element['closed']))
 					{
-						if (preg_match('{<'.$element['subtype'].'>$}', $line)) # opening tag
+						if (strpos($line, $element['start']) !== false) # opening tag
 						{
 							$element['depth']++;
 						}
 
-						if (preg_match('{</'.$element['subtype'].'>$}', $line)) # closing tag
+						if (strpos($line, $element['end']) !== false) # closing tag
 						{
 							$element['depth'] > 0
 								? $element['depth']--
@@ -161,14 +142,15 @@ class Parsedown
 
 			# *
 
-			if ($line === '')
+			$deindented_line = ltrim($line);
+
+			if ($deindented_line === '')
 			{
 				$element['interrupted'] = true;
 
 				continue;
 			}
 
-			#
 			# composite elements
 
 			switch ($element['type'])
@@ -223,6 +205,8 @@ class Parsedown
 
 							$element['lines'] []= $line;
 
+							unset($element['interrupted']);
+
 							continue 2;
 						}
 					}
@@ -238,29 +222,19 @@ class Parsedown
 					break;
 			}
 
-			#
 			# indentation sensitive types
 
-			$deindented_line = $line;
-
 			switch ($line[0])
 			{
 				case ' ':
 
-					# ~
-
-					$deindented_line = ltrim($line);
-
-					if ($deindented_line === '')
-					{
-						continue 2;
-					}
-
 					# code block
 
-					if (preg_match('/^[ ]{4}(.*)/', $line, $matches))
+					if (isset($line[3]) and $line[3] === ' ' and $line[2] === ' ' and $line[1] === ' ')
 					{
-						if ($element['type'] === 'code_block')
+						$code_line = substr($line, 4);
+
+						if ($element['type'] === 'code block')
 						{
 							if (isset($element['interrupted']))
 							{
@@ -269,15 +243,15 @@ class Parsedown
 								unset ($element['interrupted']);
 							}
 
-							$element['text'] .= "\n".$matches[1];
+							$element['text'] .= "\n".$code_line;
 						}
 						else
 						{
 							$elements []= $element;
 
 							$element = array(
-								'type' => 'code_block',
-								'text' => $matches[1],
+								'type' => 'code block',
+								'text' => $code_line,
 							);
 						}
 
@@ -290,15 +264,20 @@ class Parsedown
 
 					# atx heading (#)
 
-					if (preg_match('/^(#{1,6})[ ]*(.+?)[ ]*#*$/', $line, $matches))
+					if (isset($line[1]))
 					{
 						$elements []= $element;
 
-						$level = strlen($matches[1]);
+						$level = 1;
+
+						while (isset($line[$level]) and $line[$level] === '#')
+						{
+							$level++;
+						}
 
 						$element = array(
-							'type' => 'h.',
-							'text' => $matches[2],
+							'type' => 'heading',
+							'text' => trim($line, '# '),
 							'level' => $level,
 						);
 
@@ -308,27 +287,28 @@ class Parsedown
 					break;
 
 				case '-':
-
-					# setext heading (---)
-
-					if ($line[0] === '-' and $element['type'] === 'p' and ! isset($element['interrupted']) and preg_match('/^[-]+[ ]*$/', $line))
-					{
-						$element['type'] = 'h.';
-						$element['level'] = 2;
-
-						continue 2;
-					}
-
-					break;
-
 				case '=':
 
-					# setext heading (===)
+					# setext heading
 
-					if ($line[0] === '=' and $element['type'] === 'p' and ! isset($element['interrupted']) and preg_match('/^[=]+[ ]*$/', $line))
+					if ($element['type'] === 'paragraph' and isset($element['interrupted']) === false)
 					{
-						$element['type'] = 'h.';
-						$element['level'] = 1;
+						$chopped_line = rtrim($line);
+
+						$i = 1;
+
+						while (isset($chopped_line[$i]))
+						{
+							if ($chopped_line[$i] !== $line[0])
+							{
+								break 2;
+							}
+
+							$i++;
+						}
+
+						$element['type'] = 'heading';
+						$element['level'] = $line[0] === '-' ? 2 : 1;
 
 						continue 2;
 					}
@@ -336,41 +316,69 @@ class Parsedown
 					break;
 			}
 
-			#
 			# indentation insensitive types
 
 			switch ($deindented_line[0])
 			{
 				case '<':
 
-					# self-closing tag
+					$position = strpos($deindented_line, '>');
 
-					if (preg_match('{^<.+?/>$}', $deindented_line))
+					if ($position > 1) # tag
 					{
+						$name = substr($deindented_line, 1, $position - 1);
+						$name = rtrim($name);
+
+						if (substr($name, -1) === '/')
+						{
+							$self_closing = true;
+
+							$name = substr($name, 0, -1);
+						}
+
+						$position = strpos($name, ' ');
+
+						if ($position)
+						{
+							$name = substr($name, 0, $position);
+						}
+
+						if ( ! ctype_alpha($name))
+						{
+							break;
+						}
+
+						if (in_array($name, $this->inline_tags))
+						{
+							break;
+						}
+
 						$elements []= $element;
 
-						$element = array(
-							'type' => '',
-							'text' => $deindented_line,
-						);
+						if (isset($self_closing))
+						{
+							$element = array(
+								'type' => 'self-closing tag',
+								'text' => $deindented_line,
+							);
 
-						continue 2;
-					}
+							unset($self_closing);
 
-					# opening tag
-
-					if (preg_match('{^<(\w+)(?:[ ].*?)?>}', $deindented_line, $matches))
-					{
-						$elements []= $element;
+							continue 2;
+						}
 
 						$element = array(
-							'type' => 'markup',
-							'subtype' => strtolower($matches[1]),
+							'type' => 'block-level markup',
 							'text' => $deindented_line,
+							'start' => '<'.$name.'>',
+							'end' => '</'.$name.'>',
 							'depth' => 0,
 						);
 
-						preg_match('{</'.$matches[1].'>\s*$}', $deindented_line) and $element['closed'] = true;
+						if (strpos($deindented_line, $element['end']))
+						{
+							$element['closed'] = true;
+						}
 
 						continue 2;
 					}
@@ -401,11 +409,18 @@ class Parsedown
 
 					# reference
 
-					if (preg_match('/^\[(.+?)\]:[ ]*([^ ]+)/', $deindented_line, $matches))
+					if (preg_match('/^\[(.+?)\]:[ ]*(.+?)(?:[ ]+[\'"](.+?)[\'"])?[ ]*$/', $deindented_line, $matches))
 					{
 						$label = strtolower($matches[1]);
 
-						$this->reference_map[$label] = trim($matches[2], '<>');;
+						$this->reference_map[$label] = array(
+							'»' => trim($matches[2], '<>'),
+						);
+
+						if (isset($matches[3]))
+						{
+							$this->reference_map[$label]['#'] = $matches[3];
+						}
 
 						continue 2;
 					}
@@ -422,7 +437,7 @@ class Parsedown
 						$elements []= $element;
 
 						$element = array(
-							'type' => 'fenced_code_block',
+							'type' => 'fenced block',
 							'text' => '',
 							'fence' => $matches[1],
 						);
@@ -474,7 +489,7 @@ class Parsedown
 
 			# li
 
-			if ($deindented_line[0] <= '9' and $deindented_line >= '0' and preg_match('/^([ ]*)\d+[.][ ](.*)/', $line, $matches))
+			if ($deindented_line[0] <= '9' and $deindented_line[0] >= '0' and preg_match('/^([ ]*)\d+[.][ ](.*)/', $line, $matches))
 			{
 				$elements []= $element;
 
@@ -493,7 +508,7 @@ class Parsedown
 
 			# paragraph
 
-			if ($element['type'] === 'p')
+			if ($element['type'] === 'paragraph')
 			{
 				if (isset($element['interrupted']))
 				{
@@ -505,6 +520,8 @@ class Parsedown
 				}
 				else
 				{
+					$this->breaks_enabled and $element['text'] .= '  ';
+
 					$element['text'] .= "\n".$line;
 				}
 			}
@@ -513,7 +530,7 @@ class Parsedown
 				$elements []= $element;
 
 				$element = array(
-					'type' => 'p',
+					'type' => 'paragraph',
 					'text' => $line,
 				);
 			}
@@ -533,12 +550,10 @@ class Parsedown
 		{
 			switch ($element['type'])
 			{
-				case 'p':
+				case 'paragraph':
 
 					$text = $this->parse_span_elements($element['text']);
 
-					$text = preg_replace('/[ ]{2}\n/', '<br />'."\n", $text);
-
 					if ($context === 'li' and $markup === '')
 					{
 						if (isset($element['interrupted']))
@@ -565,18 +580,27 @@ class Parsedown
 
 					break;
 
-				case 'code_block':
-				case 'fenced_code_block':
+				case 'code block':
 
-					$text = htmlentities($element['text'], ENT_NOQUOTES);
-
-					strpos($text, "\x1A\\") !== FALSE and $text = strtr($text, $this->escape_sequence_map);
+					$text = htmlspecialchars($element['text'], ENT_NOQUOTES, 'UTF-8');
 
 					$markup .= '<pre><code>'.$text.'</code></pre>'."\n";
 
 					break;
 
-				case 'h.':
+				case 'fenced block':
+
+					$text = htmlspecialchars($element['text'], ENT_NOQUOTES, 'UTF-8');
+
+					$markup .= '<pre><code';
+
+					isset($element['language']) and $markup .= ' class="language-'.$element['language'].'"';
+
+					$markup .= '>'.$text.'</code></pre>'."\n";
+
+					break;
+
+				case 'heading':
 
 					$text = $this->parse_span_elements($element['text']);
 
@@ -612,6 +636,12 @@ class Parsedown
 
 					break;
 
+				case 'block-level markup':
+
+					$markup .= $element['text']."\n";
+
+					break;
+
 				default:
 
 					$markup .= $element['text']."\n";
@@ -621,165 +651,383 @@ class Parsedown
 		return $markup;
 	}
 
-	private function parse_span_elements($text)
+	private function parse_span_elements($text, $markers = array("  \n", '![', '&', '*', '<', '[', '\\', '_', '`', 'http', '~~'))
 	{
-		$map = array();
-
-		$index = 0;
-
-		# code span
-
-		if (strpos($text, '`') !== FALSE and preg_match_all('/`(.+?)`/', $text, $matches, PREG_SET_ORDER))
+		if (isset($text[1]) === false or $markers === array())
 		{
-			foreach ($matches as $matches)
-			{
-				$element_text = $matches[1];
-				$element_text = htmlentities($element_text, ENT_NOQUOTES);
-
-				# decodes escape sequences
-
-				$this->escape_sequence_map
-					and strpos($element_text, "\x1A") !== FALSE
-					and $element_text = strtr($element_text, $this->escape_sequence_map);
-
-				# composes element
-
-				$element = '<code>'.$element_text.'</code>';
-
-				# encodes element
-
-				$code = "\x1A".'$'.$index;
-
-				$text = str_replace($matches[0], $code, $text);
-
-				$map[$code] = $element;
-
-				$index ++;
-			}
-		}
-
-		# inline link or image
-
-		if (strpos($text, '](') !== FALSE and preg_match_all('/(!?)(\[((?:[^\[\]]|(?2))*)\])\((.*?)\)/', $text, $matches, PREG_SET_ORDER)) # inline
-		{
-			foreach ($matches as $matches)
-			{
-				$url = $matches[4];
-
-				strpos($url, '&') !== FALSE and $url = preg_replace('/&(?!#?\w+;)/', '&amp;', $url);
-
-				if ($matches[1]) # image
-				{
-					$element = '<img alt="'.$matches[3].'" src="'.$url.'">';
-				}
-				else
-				{
-					$element_text = $this->parse_span_elements($matches[3]);
-
-					$element = '<a href="'.$url.'">'.$element_text.'</a>';
-				}
-
-				# ~
-
-				$code = "\x1A".'$'.$index;
-
-				$text = str_replace($matches[0], $code, $text);
-
-				$map[$code] = $element;
-
-				$index ++;
-			}
-		}
-
-		# reference link or image
-
-		if ($this->reference_map and strpos($text, '[') !== FALSE and preg_match_all('/(!?)\[(.+?)\](?:\n?[ ]?\[(.*?)\])?/ms', $text, $matches, PREG_SET_ORDER))
-		{
-			foreach ($matches as $matches)
-			{
-				$link_definition = isset($matches[3]) && $matches[3]
-					? $matches[3]
-					: $matches[2]; # implicit
-
-				$link_definition = strtolower($link_definition);
-
-				if (isset($this->reference_map[$link_definition]))
-				{
-					$url = $this->reference_map[$link_definition];
-
-					strpos($url, '&') !== FALSE and $url = preg_replace('/&(?!#?\w+;)/', '&amp;', $url);
-
-					if ($matches[1]) # image
-					{
-						$element = '<img alt="'.$matches[2].'" src="'.$url.'">';
-					}
-					else # anchor
-					{
-						$element_text = $this->parse_span_elements($matches[2]);
-
-						$element = '<a href="'.$url.'">'.$element_text.'</a>';
-					}
-
-					# ~
-
-					$code = "\x1A".'$'.$index;
-
-					$text = str_replace($matches[0], $code, $text);
-
-					$map[$code] = $element;
-
-					$index ++;
-				}
-			}
-		}
-
-		# automatic link
-
-		if (strpos($text, '<') !== FALSE and preg_match_all('/<((https?|ftp|dict):[^\^\s]+?)>/i', $text, $matches, PREG_SET_ORDER))
-		{
-			foreach ($matches as $matches)
-			{
-				$url = $matches[1];
-
-				strpos($url, '&') !== FALSE and $url = preg_replace('/&(?!#?\w+;)/', '&amp;', $url);
-
-				$element = '<a href=":href">:text</a>';
-				$element = str_replace(':text', $url, $element);
-				$element = str_replace(':href', $url, $element);
-
-				# ~
-
-				$code = "\x1A".'$'.$index;
-
-				$text = str_replace($matches[0], $code, $text);
-
-				$map[$code] = $element;
-
-				$index ++;
-			}
+			return $text;
 		}
 
 		# ~
 
-		strpos($text, '&') !== FALSE and $text = preg_replace('/&(?!#?\w+;)/', '&amp;', $text);
-		strpos($text, '<') !== FALSE and $text = preg_replace('/<(?!\/?\w.*?>)/', '&lt;', $text);
+		$markup = '';
 
-		# ~
-
-		if (strpos($text, '_') !== FALSE)
+		while ($markers)
 		{
-			$text = preg_replace('/__(?=\S)(.+?)(?<=\S)__(?!_)/s', '<strong>$1</strong>', $text);
-			$text = preg_replace('/_(?=\S)(.+?)(?<=\S)_/s', '<em>$1</em>', $text);
+			$closest_marker = null;
+			$closest_marker_index = 0;
+			$closest_marker_position = null;
+
+			foreach ($markers as $index => $marker)
+			{
+				$marker_position = strpos($text, $marker);
+
+				if ($marker_position === false)
+				{
+					unset($markers[$index]);
+
+					continue;
+				}
+
+				if ($closest_marker === null or $marker_position < $closest_marker_position)
+				{
+					$closest_marker = $marker;
+					$closest_marker_index = $index;
+					$closest_marker_position = $marker_position;
+				}
+			}
+
+			# ~
+
+			if ($closest_marker === null or isset($text[$closest_marker_position + 1]) === false)
+			{
+				$markup .= $text;
+
+				break;
+			}
+			else
+			{
+				$markup .= substr($text, 0, $closest_marker_position);
+			}
+
+			$text = substr($text, $closest_marker_position);
+
+			# ~
+
+			unset($markers[$closest_marker_index]);
+
+			# ~
+
+			switch ($closest_marker)
+			{
+				case "  \n":
+
+					$markup .= '<br />'."\n";
+
+					$offset = 3;
+
+					break;
+
+				case '![':
+				case '[':
+
+					if (strpos($text, ']') and preg_match('/\[((?:[^][]|(?R))*)\]/', $text, $matches))
+					{
+						$element = array(
+							'!' => $text[0] === '!',
+							'a' => $matches[1],
+						);
+
+						$offset = strlen($matches[0]);
+
+						$element['!'] and $offset++;
+
+						$remaining_text = substr($text, $offset);
+
+						if ($remaining_text[0] === '(' and preg_match('/\([ ]*(.*?)(?:[ ]+[\'"](.+?)[\'"])?[ ]*\)/', $remaining_text, $matches))
+						{
+							$element['»'] = $matches[1];
+
+							if (isset($matches[2]))
+							{
+								$element['#'] = $matches[2];
+							}
+
+							$offset += strlen($matches[0]);
+						}
+						elseif ($this->reference_map)
+						{
+							$reference = $element['a'];
+
+							if (preg_match('/^\s*\[(.*?)\]/', $remaining_text, $matches))
+							{
+								$reference = $matches[1] ? $matches[1] : $element['a'];
+
+								$offset += strlen($matches[0]);
+							}
+
+							$reference = strtolower($reference);
+
+							if (isset($this->reference_map[$reference]))
+							{
+								$element['»'] = $this->reference_map[$reference]['»'];
+
+								if (isset($this->reference_map[$reference]['#']))
+								{
+									$element['#'] = $this->reference_map[$reference]['#'];
+								}
+							}
+							else
+							{
+								unset($element);
+							}
+						}
+						else
+						{
+							unset($element);
+						}
+					}
+
+					if (isset($element))
+					{
+						$element['»'] = str_replace('&', '&amp;', $element['»']);
+						$element['»'] = str_replace('<', '&lt;', $element['»']);
+
+						if ($element['!'])
+						{
+							$markup .= '<img alt="'.$element['a'].'" src="'.$element['»'].'" />';
+						}
+						else
+						{
+							$element['a'] = $this->parse_span_elements($element['a'], $markers);
+
+							$markup .= isset($element['#'])
+								? '<a href="'.$element['»'].'" title="'.$element['#'].'">'.$element['a'].'</a>'
+								: '<a href="'.$element['»'].'">'.$element['a'].'</a>';
+						}
+
+						unset($element);
+					}
+					else
+					{
+						$markup .= $closest_marker;
+
+						$offset = $closest_marker === '![' ? 2 : 1;
+					}
+
+					break;
+
+				case '&':
+
+					if (preg_match('/^&#?\w+;/', $text, $matches))
+					{
+						$markup .= $matches[0];
+
+						$offset = strlen($matches[0]);
+					}
+					else
+					{
+						$markup .= '&amp;';
+
+						$offset = 1;
+					}
+
+					break;
+
+				case '*':
+				case '_':
+
+					if ($text[1] === $closest_marker and preg_match($this->strong_regex[$closest_marker], $text, $matches))
+					{
+						$matches[1] = $this->parse_span_elements($matches[1], $markers);
+
+						$markup .= '<strong>'.$matches[1].'</strong>';
+					}
+					elseif (preg_match($this->em_regex[$closest_marker], $text, $matches))
+					{
+						$matches[1] = $this->parse_span_elements($matches[1], $markers);
+
+						$markup .= '<em>'.$matches[1].'</em>';
+					}
+					elseif ($text[1] === $closest_marker and preg_match($this->strong_em_regex[$closest_marker], $text, $matches))
+					{
+						$matches[2] = $this->parse_span_elements($matches[2], $markers);
+
+						$matches[1] and $matches[1] = $this->parse_span_elements($matches[1], $markers);
+						$matches[3] and $matches[3] = $this->parse_span_elements($matches[3], $markers);
+
+						$markup .= '<strong>'.$matches[1].'<em>'.$matches[2].'</em>'.$matches[3].'</strong>';
+					}
+					elseif (preg_match($this->em_strong_regex[$closest_marker], $text, $matches))
+					{
+						$matches[2] = $this->parse_span_elements($matches[2], $markers);
+
+						$matches[1] and $matches[1] = $this->parse_span_elements($matches[1], $markers);
+						$matches[3] and $matches[3] = $this->parse_span_elements($matches[3], $markers);
+
+						$markup .= '<em>'.$matches[1].'<strong>'.$matches[2].'</strong>'.$matches[3].'</em>';
+					}
+
+					if (isset($matches) and $matches)
+					{
+						$offset = strlen($matches[0]);
+					}
+					else
+					{
+						$markup .= $closest_marker;
+
+						$offset = 1;
+					}
+
+					break;
+
+				case '<':
+
+					if (strpos($text, '>') !== false)
+					{
+						if ($text[1] === 'h' and preg_match('/^<(https?:[\/]{2}[^\s]+?)>/i', $text, $matches))
+						{
+							$element_url = $matches[1];
+							$element_url = str_replace('&', '&amp;', $element_url);
+							$element_url = str_replace('<', '&lt;', $element_url);
+
+							$markup .= '<a href="'.$element_url.'">'.$element_url.'</a>';
+
+							$offset = strlen($matches[0]);
+						}
+						elseif (preg_match('/^<\/?\w.*?>/', $text, $matches))
+						{
+							$markup .= $matches[0];
+
+							$offset = strlen($matches[0]);
+						}
+						else
+						{
+							$markup .= '&lt;';
+
+							$offset = 1;
+						}
+					}
+					else
+					{
+						$markup .= '&lt;';
+
+						$offset = 1;
+					}
+
+					break;
+
+				case '\\':
+
+					if (in_array($text[1], $this->special_characters))
+					{
+						$markup .= $text[1];
+
+						$offset = 2;
+					}
+					else
+					{
+						$markup .= '\\';
+
+						$offset = 1;
+					}
+
+					break;
+
+				case '`':
+
+					if (preg_match('/^`(.+?)`/', $text, $matches))
+					{
+						$element_text = $matches[1];
+						$element_text = htmlspecialchars($element_text, ENT_NOQUOTES, 'UTF-8');
+
+						$markup .= '<code>'.$element_text.'</code>';
+
+						$offset = strlen($matches[0]);
+					}
+					else
+					{
+						$markup .= '`';
+
+						$offset = 1;
+					}
+
+					break;
+
+				case 'http':
+
+					if (preg_match('/^https?:[\/]{2}[^\s]+\b/i', $text, $matches))
+					{
+						$element_url = $matches[0];
+						$element_url = str_replace('&', '&amp;', $element_url);
+						$element_url = str_replace('<', '&lt;', $element_url);
+
+						$markup .= '<a href="'.$element_url.'">'.$element_url.'</a>';
+
+						$offset = strlen($matches[0]);
+					}
+					else
+					{
+						$markup .= 'http';
+
+						$offset = 4;
+					}
+
+					break;
+
+				case '~~':
+
+					if (preg_match('/^~~(?=\S)(.+?)(?<=\S)~~/', $text, $matches))
+					{
+						$matches[1] = $this->parse_span_elements($matches[1], $markers);
+
+						$markup .= '<del>'.$matches[1].'</del>';
+
+						$offset = strlen($matches[0]);
+					}
+					else
+					{
+						$markup .= '~~';
+
+						$offset = 2;
+					}
+
+					break;
+			}
+
+			if (isset($offset))
+			{
+				$text = substr($text, $offset);
+			}
+
+			$markers[$closest_marker_index] = $closest_marker;
 		}
 
-		if (strpos($text, '*') !== FALSE)
-		{
-			$text = preg_replace('/\*\*(?=\S)(.+?)(?<=\S)\*\*(?!\*)/s', '<strong>$1</strong>', $text);
-			$text = preg_replace('/\*(?=\S)(.+?)(?<=\S)\*/s', '<em>$1</em>', $text);
-		}
-
-		$text = strtr($text, $map);
-
-		return $text;
+		return $markup;
 	}
+
+	#
+	# Read-only
+	#
+
+	private $inline_tags = array(
+		'a', 'abbr', 'acronym', 'b', 'bdo', 'big', 'br', 'button',
+		'cite', 'code', 'dfn', 'em', 'i', 'img', 'input', 'kbd',
+		'label', 'map', 'object', 'q', 'samp', 'script', 'select', 'small',
+		'span', 'strong', 'sub', 'sup', 'textarea', 'tt', 'var',
+	);
+
+	private $special_characters = array('\\', '`', '*', '_', '{', '}', '[', ']', '(', ')', '>', '#', '+', '-', '.', '!');
+
+	# ~
+
+	private $strong_regex = array(
+		'*' => '/^[*]{2}([^*]+?)[*]{2}(?![*])/s',
+		'_' => '/^__([^_]+?)__(?!_)/s',
+	);
+
+	private $em_regex = array(
+		'*' => '/^[*]([^*]+?)[*](?![*])/s',
+		'_' => '/^_([^_]+?)[_](?![_])\b/s',
+	);
+
+	private $strong_em_regex = array(
+		'*' => '/^[*]{2}(.*?)[*](.+?)[*](.*?)[*]{2}/s',
+		'_' => '/^__(.*?)_(.+?)_(.*?)__/s',
+	);
+
+	private $em_strong_regex = array(
+		'*' => '/^[*](.*?)[*]{2}(.+?)[*]{2}(.*?)[*]/s',
+		'_' => '/^_(.*?)__(.+?)__(.*?)_/s',
+	);
 }

README.md

@@ -1,8 +1,20 @@
 ## Parsedown
 
-Parsedown is a Markdown parser for PHP. It is fast, consistent and easy to use.
+Better [Markdown](http://en.wikipedia.org/wiki/Markdown) parser for PHP.
 
-[Home](http://parsedown.org)  · [Demo](http://parsedown.org/explorer/) · [Tests](http://parsedown.org/tests/)
+***
+
+[ [demo](http://parsedown.org/demo) ] [ [tests](http://parsedown.org/tests/) ]
+
+***
+
+### Features
+
+* [fast](http://parsedown.org/speed)
+* [consistent](http://parsedown.org/consistency)
+* [GitHub Flavored](https://help.github.com/articles/github-flavored-markdown)
+* friendly to international input
+* [tested](https://travis-ci.org/erusev/parsedown) in PHP 5.2, 5.3, 5.4, 5.5 and [hhvm](http://www.hhvm.com/)
 
 ### Installation
 
@@ -11,9 +23,9 @@ Include `Parsedown.php` or install [the composer package](https://packagist.org/
 ### Example
 
 ```php
-$text = 'Hello **Parsedown**!';
+$text = 'Hello _Parsedown_!';
 
 $result = Parsedown::instance()->parse($text);
 
-echo $result; # prints: <p>Hello <strong>Parsedown</strong>!</p>
+echo $result; # prints: <p>Hello <em>Parsedown</em>!</p>
 ```

tests/data/atx_heading.html

@@ -4,4 +4,5 @@
 <h4>h4</h4>
 <h5>h5</h5>
 <h6>h6</h6>
-<h1>closed h1</h1>
+<h1>closed h1</h1>
+<p>#</p>

tests/data/atx_heading.md

@@ -10,4 +10,6 @@
 
 ###### h6
 
-# closed h1 #
+# closed h1 #
+
+#

tests/data/block-level_html.html

@@ -1,9 +1,5 @@
-<div>content</div>
-<hr style="background: #eaa;" />
-<p>nested elements:</p>
+<div>_content_</div>
+<p>sparse:</p>
 <div>
-parent
-<div>
-child
-</div>
+_content_
 </div>

tests/data/block-level_html.md

@@ -1,12 +1,7 @@
-<div>content</div>
+<div>_content_</div>
 
-<hr style="background: #eaa;" />
-
-nested elements:
+sparse:
 
 <div>
-parent
-<div>
-child
-</div>
+_content_
 </div>

tests/data/compound_emphasis.html

@@ -0,0 +1,2 @@
+<p><em><code>code</code></em> <strong><code>code</code></strong></p>
+<p><em><code>code</code><strong><code>code</code></strong><code>code</code></em></p>

tests/data/compound_emphasis.md

@@ -0,0 +1,4 @@
+_`code`_ __`code`__
+
+*`code`**`code`**`code`*
+

tests/data/em_strong.html

@@ -1,5 +1,8 @@
 <p><strong><em>em strong</em></strong></p>
-<p><strong><em>one</em> at the start</strong></p>
-<p><strong>one at the <em>end</em></strong></p>
-<p><strong>one <em>in the</em> middle</strong></p>
-<p><strong>one with <em>asterisks</em></strong></p>
+<p><strong><em>em strong</em> strong</strong></p>
+<p><strong>strong <em>em strong</em></strong></p>
+<p><strong>strong <em>em strong</em> strong</strong></p>
+<p><strong><em>em strong</em></strong></p>
+<p><strong><em>em strong</em> strong</strong></p>
+<p><strong>strong <em>em strong</em></strong></p>
+<p><strong>strong <em>em strong</em> strong</strong></p>

tests/data/em_strong.md

@@ -1,9 +1,15 @@
 ___em strong___
 
-___one_ at the start__
+___em strong_ strong__
 
-__one at the _end___
+__strong _em strong___
 
-__one _in the_ middle__
+__strong _em strong_ strong__
 
-**one with *asterisks***
+***em strong***
+
+***em strong* strong**
+
+**strong *em strong***
+
+**strong *em strong* strong**

tests/data/emphasis.html

@@ -1,5 +1,8 @@
 <p><em>underscore</em>, <em>asterisk</em>, <em>one two</em>, <em>three four</em>, <em>a</em>, <em>b</em></p>
-<p><em>multiline
-emphasis</em></p>
-<p>_ this _ is not an emphasis, neither is _ this_, _this _, or _this*</p>
-<p>an empty emphasis __ ** is not an emphasis</p>
+<p><strong>strong</strong> and <em>em</em> and <strong>strong</strong> and <em>em</em></p>
+<p><em>line
+line
+line</em></p>
+<p>this_is_not_an_emphasis</p>
+<p>an empty emphasis __ ** is not an emphasis</p>
+<p>*mixed *<em>double and</em> single asterisk** spans</p>

tests/data/emphasis.md

@@ -1,8 +1,13 @@
 _underscore_, *asterisk*, _one two_, *three four*, _a_, *b*
 
-_multiline
-emphasis_
+**strong** and *em* and **strong** and *em*
 
-_ this _ is not an emphasis, neither is _ this_, _this _, or _this*
+_line
+line
+line_
 
-an empty emphasis __ ** is not an emphasis
+this_is_not_an_emphasis
+
+an empty emphasis __ ** is not an emphasis
+
+*mixed **double and* single asterisk** spans

tests/data/fenced_code_block.html

@@ -2,4 +2,5 @@
 
 $message = 'fenced code block';
 echo $message;</code></pre>
-<pre><code>tilde</code></pre>
+<pre><code>tilde</code></pre>
+<pre><code class="language-php">echo 'language identifier';</code></pre>

tests/data/fenced_code_block.md

@@ -7,4 +7,8 @@ echo $message;
 
 ~~~
 tilde
-~~~
+~~~
+
+```php
+echo 'language identifier';
+```

tests/data/html_entity.html

@@ -0,0 +1 @@
+<p>&amp; &copy; &#123;</p>

tests/data/html_entity.md

@@ -0,0 +1 @@
+& © {

tests/data/image_reference.html

@@ -1 +1 @@
-<p><img alt="Markdown Logo" src="/md.png"></p>
+<p><img alt="Markdown Logo" src="/md.png" /></p>

tests/data/image_reference.md

@@ -1,3 +1,3 @@
 ![Markdown Logo][image]
 
-[image]: /md.png
+[image]: /md.png

tests/data/implicit_reference.html

@@ -1,2 +1,3 @@
 <p>an <a href="http://example.com">implicit</a> reference link</p>
-<p>an <a href="http://example.com">implicit</a> reference link with an empty link definition</p>
+<p>an <a href="http://example.com">implicit</a> reference link with an empty link definition</p>
+<p>an <a href="http://example.com" title="Example">explicit</a> reference link with a title</p>

tests/data/implicit_reference.md

@@ -2,4 +2,8 @@ an [implicit] reference link
 
 [implicit]: http://example.com
 
-an [implicit][] reference link with an empty link definition
+an [implicit][] reference link with an empty link definition
+
+an [explicit][example] reference link with a title
+
+[example]: http://example.com "Example"

tests/data/inline_link.html

@@ -1,2 +1,4 @@
-<p><a href="http://example.com">link</a></p>
-<p><a href="http://example.com"><img alt="MD Logo" src="http://parsedown.org/md.png"></a></p>
+<p><a href="http://example.com">link</a> and <a href="/tests/">another link</a></p>
+<p><a href="http://example.com"><code>link</code></a></p>
+<p><a href="http://example.com"><img alt="MD Logo" src="http://parsedown.org/md.png" /></a></p>
+<p><a href="http://example.com"><img alt="MD Logo" src="http://parsedown.org/md.png" /> and text</a></p>

tests/data/inline_link.md

@@ -1,3 +1,7 @@
-[link](http://example.com)
+[link](http://example.com) and [another link](/tests/)
 
-[![MD Logo](http://parsedown.org/md.png)](http://example.com)
+[`link`](http://example.com)
+
+[![MD Logo](http://parsedown.org/md.png)](http://example.com)
+
+[![MD Logo](http://parsedown.org/md.png) and text](http://example.com)

tests/data/inline_title.html

@@ -0,0 +1 @@
+<p><a href="http://example.com" title="Example">single quotes</a> and <a href="http://example.com" title="Example">double quotes</a></p>

tests/data/inline_title.md

@@ -0,0 +1 @@
+[single quotes](http://example.com 'Example') and [double quotes](http://example.com "Example")

tests/data/multiline_list_paragraph.html

@@ -0,0 +1,7 @@
+<ul>
+<li>
+<p>li</p>
+<p>line
+line</p>
+</li>
+</ul>

tests/data/multiline_list_paragraph.md

@@ -0,0 +1,4 @@
+- li
+
+  line
+  line

tests/data/nested_block-level_html.html

@@ -0,0 +1,10 @@
+<div>
+_parent_
+<div>
+_child_
+</div>
+<pre>
+_adopted child_
+</pre>
+</div>
+<p><em>outside</em></p>

tests/data/nested_block-level_html.md

@@ -0,0 +1,11 @@
+<div>
+_parent_
+<div>
+_child_
+</div>
+<pre>
+_adopted child_
+</pre>
+</div>
+
+_outside_

tests/data/paragraph_list.html

@@ -0,0 +1,12 @@
+<p>paragraph</p>
+<ul>
+<li>li</li>
+<li>li</li>
+</ul>
+<p>paragraph</p>
+<ul>
+<li>
+<p>li</p>
+</li>
+<li>li</li>
+</ul>

tests/data/paragraph_list.md

@@ -0,0 +1,9 @@
+paragraph
+- li
+- li
+
+paragraph
+
+   * li
+   
+   * li

tests/data/reference_title.html

@@ -0,0 +1 @@
+<p><a href="http://example.com" title="Title">single quotes</a> and <a href="http://example.com" title="Title">double quotes</a></p>

tests/data/reference_title.md

@@ -0,0 +1 @@
+[single quotes](http://example.com 'Title') and [double quotes](http://example.com "Title")

tests/data/self-closing_block-level_html.html

@@ -0,0 +1,4 @@
+<hr />
+<p>attributes:</p>
+<hr style="background: #9bd;" />
+<p>...</p>

tests/data/self-closing_block-level_html.md

@@ -0,0 +1,7 @@
+<hr />
+
+attributes:
+
+<hr style="background: #9bd;" />
+
+...

tests/data/span-level_html.html

@@ -1,3 +1,4 @@
 <p>an <b>important</b> <a href=''>link</a></p>
 <p>broken<br/>
-line</p>
+line</p>
+<p><b>inline tag</b> at the beginning</p>

tests/data/span-level_html.md

@@ -1,4 +1,6 @@
 an <b>important</b> <a href=''>link</a>
 
 broken<br/>
-line
+line
+
+<b>inline tag</b> at the beginning

tests/data/special_characters.html

@@ -1,7 +1,6 @@
 <p>AT&amp;T has an ampersand in their name</p>
-<p>AT&amp;T is another way to write it</p>
 <p>this &amp; that</p>
 <p>4 &lt; 5 and 6 > 5</p>
-<p><a href="http://example.com/autolink?a=1&amp;b=2">http://example.com/autolink?a=1&amp;b=2</a> </p>
+<p><a href="http://example.com/autolink?a=1&amp;b=2">http://example.com/autolink?a=1&amp;b=2</a></p>
 <p><a href="/script?a=1&amp;b=2">inline link</a></p>
 <p><a href="http://example.com/?a=1&amp;b=2">reference link</a></p>

tests/data/special_characters.md

@@ -1,12 +1,10 @@
 AT&T has an ampersand in their name
 
-AT&T is another way to write it
-
 this & that
 
 4 < 5 and 6 > 5
 
-<http://example.com/autolink?a=1&b=2> 
+<http://example.com/autolink?a=1&b=2>
 
 [inline link](/script?a=1&b=2)
 

tests/data/strikethrough.html

@@ -0,0 +1,3 @@
+<p><del>strikethrough</del></p>
+<p>here's <del>one</del> followed by <del>another one</del></p>
+<p>~~ this ~~ is not one neither is ~this~</p>

tests/data/strikethrough.md

@@ -0,0 +1,5 @@
+~~strikethrough~~
+
+here's ~~one~~ followed by ~~another one~~
+
+~~ this ~~ is not one neither is ~this~

tests/data/strong_em.html

@@ -0,0 +1,6 @@
+<p><em>em <strong>strong em</strong></em></p>
+<p><em><strong>strong em</strong> em</em></p>
+<p><em>em <strong>strong em</strong> em</em></p>
+<p><em>em <strong>strong em</strong></em></p>
+<p><em><strong>strong em</strong> em</em></p>
+<p><em>em <strong>strong em</strong> em</em></p>

tests/data/strong_em.md

@@ -0,0 +1,11 @@
+*em **strong em***
+
+***strong em** em*
+
+*em **strong em** em*
+
+_em __strong em___
+
+___strong em__ em_
+
+_em __strong em__ em_

tests/data/tab-indented_code_block.html

@@ -0,0 +1,6 @@
+<pre><code>&lt;?php
+
+$message = 'Hello World!'; 
+echo $message;
+
+echo "following a blank line";</code></pre>

tests/data/tab-indented_code_block.md

@@ -0,0 +1,6 @@
+	<?php
+	
+	$message = 'Hello World!'; 
+	echo $message;
+
+	echo "following a blank line";

tests/data/text_reference.html

@@ -3,4 +3,5 @@
 <p>[one][404] with no definition</p>
 <p><a href="http://example.com">multiline
 one</a> defined on 2 lines</p>
-<p><a href="http://example.com">one</a> with an upper case label</p>
+<p><a href="http://example.com">one</a> with an upper case label</p>
+<p><a href="http://example.com"><code>link</code></a></p>

tests/data/text_reference.md

@@ -13,4 +13,6 @@ one][website] defined on 2 lines
 
 [one][label] with an upper case label
 
-[LABEL]: http://example.com
+[LABEL]: http://example.com
+
+[`link`][website]

tests/data/url_autolinking.html

@@ -0,0 +1,2 @@
+<p>an autolink <a href="http://example.com">http://example.com</a></p>
+<p>inside of brackets [<a href="http://example.com">http://example.com</a>], inside of braces {<a href="http://example.com">http://example.com</a>},  inside of parentheses (<a href="http://example.com">http://example.com</a>)</p>

tests/data/url_autolinking.md

@@ -0,0 +1,3 @@
+an autolink http://example.com
+
+inside of brackets [http://example.com], inside of braces {http://example.com},  inside of parentheses (http://example.com)