escape sequences with double digit codes do not get decoded properly

Allow multi-line emphasis. Fix #28.

.gitignore

@@ -1,2 +1,3 @@
 .DS_Store
+.idea
 nbproject

.travis.yml

@@ -4,3 +4,8 @@ php:
   - 5.5
   - 5.4
   - 5.3
+  - 5.2
+
+matrix:
+  allow_failures:
+    - php: 5.2

Parsedown.php

@@ -50,7 +50,8 @@ class Parsedown
 		$text = preg_replace('{^\xEF\xBB\xBF|\x1A}', '', $text);
 
 		# Removes \r characters.
-		$text = str_replace("\r", '', $text);
+		$text = str_replace("\r\n", "\n", $text);
+		$text = str_replace("\r", "\n", $text);
 
 		# Replaces tabs with spaces.
 		$text = str_replace("\t", '    ', $text);
@@ -65,7 +66,7 @@ class Parsedown
 			{
 				if (strpos($text, $escape_sequence) !== FALSE)
 				{
-					$code = "\x1A".'\\'.$index;
+					$code = "\x1A".'\\'.$index.';';
 
 					$text = str_replace($escape_sequence, $code, $text);
 
@@ -74,21 +75,14 @@ class Parsedown
 			}
 		}
 
-		# Extracts link references.
-		
-		if (preg_match_all('/^[ ]{0,3}\[(.+)\][ ]?:[ ]*\n?[ ]*(.+)$/m', $text, $matches, PREG_SET_ORDER))
-		{
-			foreach ($matches as $matches)
-			{
-				$this->reference_map[$matches[1]] = $matches[2];
-				
-				$text = str_replace($matches[0], '', $text);
-			}
-		}
-		
 		# ~
 
-		$text = $this->parse_blocks($text);
+		$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 (leaves out backslashes).
 
@@ -106,380 +100,431 @@ class Parsedown
 	# Private Methods
 	#
 
-	private function parse_blocks($text)
+	private function parse_block_elements(array $lines, $context = '')
 	{
-		# Divides text into blocks.
-		$blocks = preg_split('/\n\s*\n/', $text, -1, PREG_SPLIT_NO_EMPTY);
+		$elements = array();
 
-		# Makes sure compound blocks get rendered.
-		$blocks []= NULL;
+		$element = array(
+			'type' => '',
+		);
 
-		$markup = '';
-		
-		# Parses blocks.
-		
-		foreach ($blocks as $block)
+		foreach ($lines as $line)
 		{
-			if (isset($block) and $block[0] >= 'A')
-			{
-				$quick_block = $block;
+			# Block-Level HTML
 
-				unset($block);
+			if ($element['type'] === 'block' and ! isset($element['closed']))
+			{
+				if (preg_match('{<'.$element['subtype'].'>$}', $line)) # <open>
+				{
+					$element['depth']++;
 				}
 
-			# List 
-			
-			if (isset($block) and preg_match('/^([ ]{0,3})(\d+[.]|[*+-])[ ]/', $block, $matches)) # list item
+				if (preg_match('{</'.$element['subtype'].'>$}', $line)) # </close>
 				{
-				if (isset($list)) # subsequent 
-				{
-					$list .= "\n\n".$block;
-				}
-				else # first 
-				{
-					$list = $block;
-					$list_indentation = strlen($matches[1]);
-					
-					list($list_type, $list_marker_pattern) = ($matches[2] === '-' or $matches[2] === '+' or $matches[2] === '*')
-						? array('ul', '[*+-]')
-						: array('ol', '\d+[.]');
+					$element['depth'] > 0
+						? $element['depth']--
+						: $element['closed'] = true;
 				}
 
-				unset($block);
+				$element['text'] .= "\n".$line;
+
+				continue;
 			}
-			elseif (isset($block) and isset($list) and $block[0] === ' ') # list item block 
-			{
-				$list .= "\n\n".$block;
 
-				unset($block);
+			# Empty
+
+			if ($line === '')
+			{
+				$element['interrupted'] = true;
+
+				continue;
 			}
-			elseif (isset($list))
-			{
-				$markup .= '<'.$list_type.'>'."\n";
 
-				$list_items = preg_split('/^([ ]{'.$list_indentation.'})'.$list_marker_pattern.'[ ]/m', $list, -1, PREG_SPLIT_NO_EMPTY);
+			# Lazy Blockquote
 
-				foreach ($list_items as $list_item)
+			if ($element['type'] === 'blockquote' and ! isset($element['interrupted']))
 			{
-					$markup .= '<li>';
+				$line = preg_replace('/^[ ]*>[ ]?/', '', $line);
 
-					if (strpos($list_item, "\n\n")) # sparse 
-					{
-						$list_item = trim($list_item, "\n");
+				$element['lines'] []= $line;
 
-						if (strpos($list_item, "\n\n"))
+				continue;
+			}
+
+			# Lazy List Item
+
+			if ($element['type'] === 'li')
 			{
-							$list_item = preg_replace('/^[ ]{0,4}/m', '', $list_item);
-							$list_item = $this->parse_blocks($list_item);
+				if (preg_match('/^([ ]{0,3})(\d+[.]|[*+-])[ ](.*)/', $line, $matches))
+				{
+					if ($element['indentation'] !== $matches[1])
+					{
+						$element['lines'] []= $line;
 					}
 					else
 					{
-							$list_item = $this->parse_lines($list_item, TRUE);
+						unset($element['last']);
+
+						$elements []= $element;
+
+						$element = array(
+							'type' => 'li',
+							'indentation' => $matches[1],
+							'last' => true,
+							'lines' => array(
+								preg_replace('/^[ ]{0,4}/', '', $matches[3]),
+							),
+						);
 					}
 
-						$markup .= "\n".$list_item;
+					continue;
 				}
-					else # dense 
+
+				if (isset($element['interrupted']))
 				{
-						$list_item = trim($list_item, "\n");
+					if ($line[0] === ' ')
+					{
+						$element['lines'] []= '';
 
-						$list_item = strpos($list_item, "\n")
-							? $this->parse_lines($list_item)
-							: $this->parse_inline_elements($list_item);
+						$line = preg_replace('/^[ ]{0,4}/', '', $line);;
 
-						$markup .= $list_item;
+						$element['lines'] []= $line;
+
+						continue;
+					}
+				}
+				else
+				{
+					$line = preg_replace('/^[ ]{0,4}/', '', $line);;
+
+					$element['lines'] []= $line;
+
+					continue;
+				}
 			}
 
-					$markup .= '</li>'."\n";
-				}
+			# Quick Paragraph
 
-				$markup .= '</'.$list_type.'>'."\n";
-				
-				unset($list);
+			if ($line[0] >= 'a' or $line[0] >= 'A' and $line[0] <= 'Z')
+			{
+				goto paragraph;
 			}
 
 			# Code Block
 
-			if (isset($block) and strlen($block) > 4 and $block[0] === ' ' and $block[1] === ' ' and $block[2] === ' ' and $block[3] === ' ')
+			if ($line[0] === ' ' and preg_match('/^[ ]{4}(.*)/', $line, $matches))
 			{
-				if (isset($code_block))
+				if (trim($line) === '')
 				{
-					$code_block .= "\n\n".$block;
+					continue;
+				}
+
+				if ($element['type'] === 'code')
+				{
+					if (isset($element['interrupted']))
+					{
+						$element['text'] .= "\n";
+
+						unset ($element['interrupted']);
+					}
+
+					$element['text'] .= "\n".$matches[1];
 				}
 				else
 				{
-					$code_block = $block;
+					$elements []= $element;
+
+					$element = array(
+						'type' => 'code',
+						'text' => $matches[1],
+					);
 				}
 
-				unset($block);
+				continue;
 			}
-			elseif (isset($code_block))
+
+			# Setext Header (---)
+
+			if ($line[0] === '-' and $element['type'] === 'p' and ! isset($element['interrupted']) and preg_match('/^[-]+[ ]*$/', $line))
 			{
-				$code_block_text = preg_replace('/^[ ]{4}/m', '', $code_block);
-				$code_block_text = htmlentities($code_block_text, ENT_NOQUOTES);
+				$element['type'] = 'h.';
+				$element['level'] = 2;
 
-				# Decodes encoded escape sequences if present.
-				strpos($code_block_text, "\x1A\\") !== FALSE and $code_block_text = strtr($code_block_text, $this->escape_sequence_map);
-				
-				$markup .= '<pre><code>'.$code_block_text.'</code></pre>'."\n";
-				
-				unset($code_block);
+				continue;
 			}
 
-			# Atx Heading 
+			# Atx Header (#)
 
-			if (isset($block) and $block[0] === '#' and preg_match('/^(#{1,6})[ ]*(.+?)[ ]*#*$/', $block, $matches))
+			if ($line[0] === '#' and preg_match('/^(#{1,6})[ ]*(.+?)[ ]*#*$/', $line, $matches))
 			{
+				$elements []= $element;
+
 				$level = strlen($matches[1]);
 
-				$heading = $this->parse_inline_elements($matches[2]);
-				
-				$markup .= '<h'.$level.'>'.$heading.'</h'.$level.'>'."\n";
+				$element = array(
+					'type' => 'h.',
+					'text' => $matches[2],
+					'level' => $level,
+				);
 
 				continue;
 			}
 
-			# Quote Block 
+			# Setext Header (===)
 
-			if (isset($block) and preg_match('/^[ ]{0,3}>/', $block))
+			if ($line[0] === '=' and $element['type'] === 'p' and ! isset($element['interrupted']) and preg_match('/^[=]+[ ]*$/', $line))
 			{
-				$block = preg_replace('/^[ ]{0,3}>[ ]?/m', '', $block);
-				$block = $this->parse_blocks($block);
-				
-				$markup .= '<blockquote>'."\n".$block.'</blockquote>'."\n";
-				
-				continue;
-			}
-			
-			# Horizontal Line 
-			
-			if (isset($block) and preg_match('/^[ ]{0,3}([-*_])([ ]{0,2}\1){2,}$/', $block))
-			{
-				$markup .= '<hr />'."\n";
+				$element['type'] = 'h.';
+				$element['level'] = 1;
 
 				continue;
 			}
 
 			# ~
 
-			if (isset($quick_block))
-			{
-				$block = $quick_block;
+			$pure_line = $line[0] !== ' ' ? $line : ltrim($line);
 
-				unset ($quick_block);
+			if ($pure_line === '')
+			{
+				continue;
 			}
 
-			# 
-			# Paragraph 
+			# Link Reference
 
-			if (isset($block))
+			if ($pure_line[0] === '[' and preg_match('/^\[(.+?)\]:[ ]*([^ ]+)/', $pure_line, $matches))
 			{
-				if (strpos($block, "\n"))
+				$label = strtolower($matches[1]);
+				$url = trim($matches[2], '<>');
+
+				$this->reference_map[$label] = $url;
+
+				continue;
+			}
+
+			# Blockquote
+
+			if ($pure_line[0] === '>' and preg_match('/^>[ ]?(.*)/', $pure_line, $matches))
 			{
-					$markup .= $this->parse_lines($block, TRUE);
+				if ($element['type'] === 'blockquote')
+				{
+					if (isset($element['interrupted']))
+					{
+						$element['lines'] []= '';
+
+						unset($element['interrupted']);
+					}
+
+					$element['lines'] []= $matches[1];
 				}
 				else
 				{
-					$element_text = $this->parse_inline_elements($block);
-					$element = '<p>'.$element_text.'</p>'."\n";
+					$elements []= $element;
 
-					$markup .= $element;
-				}
-			}
+					$element = array(
+						'type' => 'blockquote',
+						'lines' => array(
+							$matches[1],
+						),
+					);
 				}
 
-		return $markup;
+				continue;
 			}
 
-	private function parse_lines($text, $paragraph_based = FALSE)
+			# HTML
+
+			if ($pure_line[0] === '<')
 			{
-		$text = trim($text, "\n");
+				# Block-Level HTML <self-closing/>
 
-		$lines = explode("\n", $text);
+				if (preg_match('{^<.+?/>$}', $pure_line))
+				{
+					$elements []= $element;
 
-		$lines []= NULL;
+					$element = array(
+						'type' => '',
+						'text' => $pure_line,
+					);
+
+					continue;
+				}
+
+				# Block-Level HTML <open>
+
+				if (preg_match('{^<(\w+)(?:[ ].*?)?>}', $pure_line, $matches))
+				{
+					$elements []= $element;
+
+					$element = array(
+						'type' => 'block',
+						'subtype' => strtolower($matches[1]),
+						'text' => $pure_line,
+						'depth' => 0,
+					);
+
+					preg_match('{</'.$matches[1].'>\s*$}', $pure_line) and $element['closed'] = true;
+
+					continue;
+				}
+			}
+
+			# Horizontal Rule
+
+			if (preg_match('/^([-*_])([ ]{0,2}\1){2,}[ ]*$/', $pure_line))
+			{
+				$elements []= $element;
+
+				$element = array(
+					'type' => 'hr',
+				);
+
+				continue;
+			}
+
+			# List Item
+
+			if (preg_match('/^([ ]*)(\d+[.]|[*+-])[ ](.*)/', $line, $matches))
+			{
+				$elements []= $element;
+
+				$element = array(
+					'type' => 'li',
+					'ordered' => isset($matches[2][1]),
+					'indentation' => $matches[1],
+					'last' => true,
+					'lines' => array(
+						preg_replace('/^[ ]{0,4}/', '', $matches[3]),
+					),
+				);
+
+				continue;
+			}
+
+			# ~
+
+			paragraph:
+
+			if ($element['type'] === 'p')
+			{
+				if (isset($element['interrupted']))
+				{
+					$elements []= $element;
+
+					$element['text'] = $line;
+
+					unset($element['interrupted']);
+				}
+				else
+				{
+					$element['text'] .= "\n".$line;
+				}
+			}
+			else
+			{
+				$elements []= $element;
+
+				$element = array(
+					'type' => 'p',
+					'text' => $line,
+				);
+			}
+		}
+
+		$elements []= $element;
+
+		array_shift($elements);
+
+		#
+		# ~
+		#
 
 		$markup = '';
 
-		foreach ($lines as $line)
+		foreach ($elements as $index => $element)
 		{
-			if (isset($line) and $line === '')
+			switch ($element['type'])
 			{
-				unset($line);
-			}
+				case 'li':
 
-			# Paragraph 
-			
-			if (isset($line) and $line[0] >= 'A')
+					if (isset($element['ordered'])) # first
 					{
-				$quick_line = $line;
-				
-				unset($line);
-			}
-			
-			# List 
-			
-			if (isset($line) and preg_match('/^([ ]*)(\d+[.]|[*+-])[ ](.*)/', $line, $matches)) # list item 
-			{
-				$list_item_indentation = strlen($matches[1]);
-				$list_item_type = ($matches[2] === '-' or $matches[2] === '+' or $matches[2] === '*')
-					? 'ul'
-					: 'ol';
-				
-				if (isset($list)) # subsequent 
-				{
-					if ($list_item_indentation === $list_indentation and $list_item_type === $list_type)
-					{
-						# Adds last list item to the list.
-						$list []= $list_item;
-						
-						# Creates a separate list item.
-						$list_item = $matches[3];
-					}
-					else 
-					{
-						# Adds line to the current list item.
-						$list_item .= "\n".$line;
-					}
-				}
-				else # first 
-				{
-					$list = array();
-					$list_indentation = $list_item_indentation;
-					$list_type = $list_item_type;
-					
-					$list_item = $matches[3];
-				}
-				
-				unset($line);
-			}
-			else 
-			{
-				if (isset($list))
-				{
-					$list []= $list_item;
+						$list_type = $element['ordered'] ? 'ol' : 'ul';
 
 						$markup .= '<'.$list_type.'>'."\n";
-					
-					foreach ($list as $list_item)
-					{
-						$list_item_text = strpos($list_item, "\n") 
-							? $this->parse_lines($list_item)
-							: $this->parse_inline_elements($list_item);
-						
-						$markup .= '<li>'.$list_item_text.'</li>'."\n";
 					}
 
-					$markup .= '</'.$list_type.'>'."\n";
-					
-					unset($list);
-				}
+					if (isset($element['interrupted']) and ! isset($element['last']))
+					{
+						$element['lines'] []= '';
 					}
 
-			# Quote Block 
+					$text = $this->parse_block_elements($element['lines'], 'li');
 
-			if (isset($line) and preg_match('/^[ ]*>[ ]?(.*)/', $line, $matches))
+					$markup .= '<li>'.$text.'</li>'."\n";
+
+					isset($element['last']) and $markup .= '</'.$list_type.'>'."\n";
+
+					break;
+
+				case 'p':
+
+					$text = $this->parse_inline_elements($element['text']);
+
+					$text = preg_replace('/[ ]{2}\n/', '<br />'."\n", $text);
+
+					if ($context === 'li' and $index === 0)
 					{
-				if (isset($quote))
+						if (isset($element['interrupted']))
 						{
-					$quote .= "\n".$matches[1];
+							$markup .= "\n".'<p>'.$text.'</p>'."\n";
 						}
 						else
 						{
-					$quote = $matches[1];
-				}
-				
-				unset($line);
-			}
-			else 
-			{
-				if (isset($quote))
-				{
-					$quote = $this->parse_blocks($quote);
-					
-					$markup .= '<blockquote>'."\n".$quote.'</blockquote>'."\n";
-					
-					unset($quote);
-				}
-			}
-			
-			# Atx Heading 
-			
-			if (isset($atx_heading))
-			{
-				$markup .= '<h'.$atx_heading_level.'>'.$atx_heading.'</h'.$atx_heading_level.'>'."\n";
-				
-				unset($atx_heading);
-			}
-			
-			if (isset($line) and $line[0] === '#' and preg_match('/^(#{1,6})[ ]*(.+?)[ ]*#*$/', $line, $matches))
-			{
-				$atx_heading_level = strlen($matches[1]);
-				
-				$atx_heading = $this->parse_inline_elements($matches[2]);
-				
-				unset($line);
-			}
-			
-			# Setext Heading 
-			
-			if (isset($line) and isset($paragraph))
-			{
-				$setext_characters = array('=', '-');
-				
-				foreach ($setext_characters as $index => $setext_character)
-				{
-					if ($line[0] === $setext_character and preg_match('/^['.$setext_character.']+[ ]*$/', $line))
-					{
-						$setext_heading_level = $index + 1;
-						
-						$setext_heading_text = $this->parse_inline_elements($paragraph);
-						
-						$markup .= '<h'.$setext_heading_level.'>'.$setext_heading_text.'</h'.$setext_heading_level.'>'."\n";
-						
-						unset($paragraph, $line);
-						
-						continue 2;
-					}
-				}
-			}
-			
-			# Paragraph 
-			
-			if (isset($quick_line))
-			{
-				$line = $quick_line;
-				
-				unset($quick_line);
-			}
-			
-			if (isset($line))
-			{
-				substr($line, -2) === '  '
-					and $line = substr($line, 0, -2)
-					and $line .= '<br/>';
-				
-				if (isset($paragraph))
-				{
-					$paragraph .= "\n".$line;
-				}
-				else 
-				{
-					$paragraph = $line;
+							$markup .= $text;
 						}
 					}
 					else
 					{
-				if (isset($paragraph))
-				{
-					$paragraph_text = $this->parse_inline_elements($paragraph);
-					
-					$markup .= $markup === '' && $paragraph_based === FALSE
-						? $paragraph_text
-						: '<p>'.$paragraph_text.'</p>'."\n";
-					
-					unset($paragraph);
+						$markup .= '<p>'.$text.'</p>'."\n";
 					}
+
+					break;
+
+				case 'code':
+
+					$text = htmlentities($element['text'], ENT_NOQUOTES);
+
+					strpos($text, "\x1A\\") !== FALSE and $text = strtr($text, $this->escape_sequence_map);
+
+					$markup .= '<pre><code>'.$text.'</code></pre>'."\n";
+
+					break;
+
+				case 'blockquote':
+
+					$text = $this->parse_block_elements($element['lines']);
+
+					$markup .= '<blockquote>'."\n".$text.'</blockquote>'."\n";
+
+					break;
+
+				case 'h.':
+
+					$text = $this->parse_inline_elements($element['text']);
+
+					$markup .= '<h'.$element['level'].'>'.$text.'</h'.$element['level'].'>'."\n";
+
+					break;
+
+				case 'hr':
+
+					$markup .= '<hr />'."\n";
+
+					break;
+
+				default:
+
+					$markup .= $element['text']."\n";
 			}
 		}
 
@@ -523,15 +568,56 @@ class Parsedown
 			}
 		}
 
-		# Reference(d) Link / Image 
+		# Inline Link / Image
 
-		if ($this->reference_map and strpos($text, '[') !== FALSE and preg_match_all('/(!?)\[(.+?)\][ ]?\[(.+?)\]/', $text, $matches, PREG_SET_ORDER))
+		if (strpos($text, '](') !== FALSE and preg_match_all('/(!?)(\[((?:[^\[\]]|(?2))*)\])\((.*?)\)/', $text, $matches, PREG_SET_ORDER)) # inline
 		{
 			foreach ($matches as $matches)
 			{
-				if (array_key_exists($matches[3], $this->reference_map))
+				$url = $matches[4];
+
+				strpos($url, '&') !== FALSE and $url = preg_replace('/&(?!#?\w+;)/', '&amp;', $url);
+
+				if ($matches[1]) # image
 				{
-					$url = $this->reference_map[$matches[3]];
+					$element = '<img alt="'.$matches[3].'" src="'.$url.'">';
+				}
+				else
+				{
+					$element_text = $this->parse_inline_elements($matches[3]);
+
+					$element = '<a href="'.$url.'">'.$element_text.'</a>';
+				}
+
+				# ~
+
+				$code = "\x1A".'$'.$index;
+
+				$text = str_replace($matches[0], $code, $text);
+
+				$map[$code] = $element;
+
+				$index ++;
+			}
+		}
+
+		# Reference(d) Link / Image
+
+		if ($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
 					{
@@ -557,44 +643,19 @@ class Parsedown
 			}
 		}
 
-		# Inline Link / Image 
-		
-		if (strpos($text, '](') !== FALSE and preg_match_all('/(!?)(\[((?:[^][]+|(?2))*)\])\((.*?)\)/', $text, $matches, PREG_SET_ORDER)) # inline 
-		{
-			foreach ($matches as $matches)
-			{
-				if ($matches[1]) # image 
-				{
-					$element = '<img alt="'.$matches[3].'" src="'.$matches[4].'">';
-				}
-				else 
-				{
-					$element_text = $this->parse_inline_elements($matches[3]);
-					
-					$element = '<a href="'.$matches[4].'">'.$element_text.'</a>';
-				}
-				
-				$element_text = $this->parse_inline_elements($matches[1]);
-				
-				# ~ 
-
-				$code = "\x1A".'$'.$index;
-
-				$text = str_replace($matches[0], $code, $text);
-
-				$map[$code] = $element;
-				
-				$index ++;
-			}
-		}
+		# Automatic Links
 
 		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', $matches[1], $element);
-				$element = str_replace(':href', $matches[1], $element);
+				$element = str_replace(':text', $url, $element);
+				$element = str_replace(':href', $url, $element);
 
 				# ~
 
@@ -608,10 +669,23 @@ class Parsedown
 			}
 		}
 
-		if (strpos($text, '*') !== FALSE or strpos($text, '_') !== FALSE)
+		# ~
+
+		strpos($text, '&') !== FALSE and $text = preg_replace('/&(?!#?\w+;)/', '&amp;', $text);
+		strpos($text, '<') !== FALSE and $text = preg_replace('/<(?!\/?\w.*?>)/', '&lt;', $text);
+
+		# ~
+
+		if (strpos($text, '_') !== FALSE)
 		{
-			$text = preg_replace('/(\*\*|__)(.+?[*_]*)(?<=\S)\1/', '<strong>$2</strong>', $text);
-			$text = preg_replace('/(\*|_)(.+?)(?<=\S)\1/', '<em>$2</em>', $text);
+			$text = preg_replace('/__(?=\S)(.+?)(?<=\S)__(?!_)/s', '<strong>$1</strong>', $text);
+			$text = preg_replace('/_(?=\S)(.+?)(?<=\S)_/s', '<em>$1</em>', $text);
+		}
+
+		if (strpos($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);
@@ -619,4 +693,3 @@ class Parsedown
 		return $text;
 	}
 }
-

README.md

@@ -1,9 +1,8 @@
 ## Parsedown PHP
 
-Parsedown is a parser for Markdown. It parses Markdown text the way people do. First, it divides texts into blocks. Then it looks at how these blocks start and how they relate to each other. Finally, it looks for special characters to identify inline elements. As a result, Parsedown is (super) fast, consistent and clean.
+Parsedown PHP is a parser for Markdown. It reads Markdown the way people do. First, it breaks texts into lines. Then, it looks at how these lines start and relate to each other. Finally, it looks for special characters to identify inline elements. As a result, Parsedown PHP is (very) fast and consistent.
 
-[Explorer (demo)](http://parsedown.org/explorer/)  
-[Tests](http://parsedown.org/tests/)
+[Home](http://parsedown.org) · [Demo](http://parsedown.org/explorer/) · [Tests](http://parsedown.org/tests/)
 
 ### Installation
 

tests/Test.php

@@ -34,6 +34,8 @@ class Test extends PHPUnit_Framework_TestCase
 					continue;
 				
 				$expected_markup = file_get_contents(__DIR__ . '/' . self::provider_dir . $basename . '.html');
+				$expected_markup = str_replace("\r\n", "\n", $expected_markup);
+				$expected_markup = str_replace("\r", "\n", $expected_markup);
 				
 				$provider [] = array($markdown, $expected_markup);
 			}

tests/data/atx_heading.html

@@ -1,14 +1,7 @@
-<h1>This is an h1</h1>
-<h2>This is an h2</h2>
-<h3>This is an h3</h3>
-<h4>This is an h4</h4>
-<h5>This is an h5</h5>
-<h6>This is an h6</h6>
-<h1>This is a closed h1</h1>
-<h2>This is a closed h2</h2>
-<h3>This is a closed h3</h3>
-<h4>This is a closed h4</h4>
-<h5>This is a closed h5</h5>
-<h6>This is a closed h6</h6>
-<h1>This is an irregularly closed h1</h1>
-<h4>This is an irregularly closed h4</h4>
+<h1>h1</h1>
+<h2>h2</h2>
+<h3>h3</h3>
+<h4>h4</h4>
+<h5>h5</h5>
+<h6>h6</h6>
+<h1>closed h1</h1>

tests/data/atx_heading.md

@@ -1,27 +1,13 @@
-# This is an h1
+# h1
 
-## This is an h2
+## h2
 
-### This is an h3
+### h3
 
-#### This is an h4
+#### h4
 
-##### This is an h5
+##### h5
 
-###### This is an h6
+###### h6
 
-# This is a closed h1 #
-
-## This is a closed h2 ##
-
-### This is a closed h3 ###
-
-#### This is a closed h4 ####
-
-##### This is a closed h5 #####
-
-###### This is a closed h6 ######
-
-# This is an irregularly closed h1 ###
-
-#### This is an irregularly closed h4 ##
+# closed h1 #

tests/data/block-level_html.html

@@ -0,0 +1,9 @@
+<div>content</div>
+<hr style="background: #eaa;" />
+<p>nested elements:</p>
+<div>
+parent
+<div>
+child
+</div>
+</div>

tests/data/block-level_html.md

@@ -0,0 +1,12 @@
+<div>content</div>
+
+<hr style="background: #eaa;" />
+
+nested elements:
+
+<div>
+parent
+<div>
+child
+</div>
+</div>

tests/data/blockquote.html

@@ -1,25 +0,0 @@
-<p>Here's a regular blockquote:</p>
-<blockquote>
-<p>This is a blockquote.</p>
-</blockquote>
-<p>Here's one with no space after the ">":</p>
-<blockquote>
-<p>This is a blockquote.</p>
-</blockquote>
-<p>Here's one with multiple paragraphs:</p>
-<blockquote>
-<p>This is line one.</p>
-<p>This is line two.</p>
-</blockquote>
-<p>Here's one with multiple types of blocks:</p>
-<blockquote>
-<p>This is a quoted paragraph.</p>
-<ul>
-<li>This is a list item of a quoted list.</li>
-<li>This is another list item.</li>
-</ul>
-<blockquote>
-<p>This is a nested quote block.</p>
-</blockquote>
-<p>This is another paragraph.</p>
-</blockquote>

tests/data/blockquote.md

@@ -1,24 +0,0 @@
-Here's a regular blockquote:
-
-> This is a blockquote.
-
-Here's one with no space after the ">":
-
->This is a blockquote.
-
-Here's one with multiple paragraphs:
-
-> This is line one.
->
-> This is line two.
-
-Here's one with multiple types of blocks:
-
-> This is a quoted paragraph.
-> 
-> - This is a list item of a quoted list.
-> - This is another list item.
->
-> > This is a nested quote block.
-> 
-> This is another paragraph.

tests/data/code_block.html

@@ -1,9 +1,8 @@
-<p>Here's a regular code block:</p>
 <pre><code>&lt;?php
 
-echo 'Hello World!'; 
-
-?&gt;</code></pre>
-<p>Here's one that holds a list:</p>
-<pre><code>- list item
-- another list item</code></pre>
+$message = 'Hello World!'; 
+echo $message;</code></pre>
+<hr />
+<pre><code>&gt; not a quote 
+- not a list item
+[not a reference]: http://foo.com</code></pre>

tests/data/code_block.md

@@ -1,13 +1,11 @@
-Here's a regular code block:
-
     <?php
     
-    echo 'Hello World!'; 
+    $message = 'Hello World!'; 
+    echo $message;
 
-	?>
+---
 
-Here's one that holds a list:
-
-    - list item
-    - another list item
+    > not a quote 
+    - not a list item
+    [not a reference]: http://foo.com
     

tests/data/code_span.html

@@ -1 +1 @@
-<p>This is a <code>code span</code>.</p>
+<p>a <code>code span</code></p>

tests/data/code_span.md

@@ -1 +1 @@
-This is a `code span`.
+a `code span`

tests/data/compound_blockquote.html

@@ -0,0 +1,9 @@
+<blockquote>
+<h2>header</h2>
+<p>paragraph</p>
+<ul>
+<li>li</li>
+</ul>
+<hr />
+<p>paragraph</p>
+</blockquote>

tests/data/compound_blockquote.md

@@ -0,0 +1,10 @@
+> header
+> ------
+>
+> paragraph
+>
+> - li
+>
+> ---
+>
+> paragraph

tests/data/compound_list.html

@@ -1,13 +1,12 @@
-<p>Here's a compound list:</p>
 <ul>
 <li>
-<p>This is the first paragraph of the list item.</p>
-<p>This is the second one.</p>
+<p>paragraph</p>
+<p>paragraph</p>
 </li>
 <li>
-<p>This is another list item.</p>
+<p>paragraph</p>
 <blockquote>
-<p>This is a quote block that belongs to it.</p>
+<p>quote</p>
 </blockquote>
 </li>
 </ul>

tests/data/compound_list.md

@@ -1,10 +1,7 @@
-Here's a compound list:
+- paragraph
 
-- This is the first paragraph of the list item.
+  paragraph
 
-  This is the second one.
-
-- This is another list item.
-
-  > This is a quote block that belongs to it.
+- paragraph
 
+  > quote

tests/data/dense_list.html

@@ -1,24 +0,0 @@
-<p>Here's a regular list:</p>
-<ul>
-<li>list item</li>
-<li>another list item</li>
-<li>3rd list item</li>
-</ul>
-<p>Here's one with white space around items:</p>
-<ul>
-<li>list item  </li>
-<li>another list item  </li>
-</ul>
-<p>Here's one with too much space before items:</p>
-<pre><code>- list item
-- another list item</code></pre>
-<p>Here's one with no space after markers:</p>
-<p>-list item
--another list item</p>
-<p>Here's one where items contain line breaks:</p>
-<ul>
-<li>list
-item</li>
-<li>another
-list item</li>
-</ul>

tests/data/dense_list.md

@@ -1,27 +0,0 @@
-Here's a regular list:
-
-- list item
-- another list item
-- 3rd list item
-
-Here's one with white space around items:
-
-  - list item  
-  - another list item  
-
-Here's one with too much space before items:
-
-    - list item
-    - another list item
-
-Here's one with no space after markers:
-
--list item
--another list item
-
-Here's one where items contain line breaks:
-
-- list
-item
-- another
-list item

tests/data/em_strong.html

@@ -0,0 +1,5 @@
+<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>

tests/data/em_strong.md

@@ -0,0 +1,9 @@
+___em strong___
+
+___one_ at the start__
+
+__one at the _end___
+
+__one _in the_ middle__
+
+**one with *asterisks***

tests/data/emphasis.html

@@ -1,8 +1,5 @@
-<p>Here's <em>an emphasis</em>.</p>
-<p>A short emphasis <em>a</em> <em>b</em> .</p>
-<p>Here's <strong>a strong one</strong>. </p>
-<p>Here's <em>an emphasis that uses underscores</em>. </p>
-<p>Here's <strong>a strong emphasis that uses underscores</strong>.</p>
-<p>This is _ not an emphasis _ neither is * that * .</p>
-<p>Empty emphasis ** is not __ an emphasis.</p>
-<p>Three asterisks are an emphasized asterisk <em>*</em> .</p>
+<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>

tests/data/emphasis.md

@@ -1,15 +1,8 @@
-Here's *an emphasis*.
+_underscore_, *asterisk*, _one two_, *three four*, _a_, *b*
 
-A short emphasis _a_ *b* .
+_multiline
+emphasis_
 
-Here's **a strong one**. 
+_ this _ is not an emphasis, neither is _ this_, _this _, or _this*
 
-Here's _an emphasis that uses underscores_. 
-
-Here's __a strong emphasis that uses underscores__.
-
-This is _ not an emphasis _ neither is * that * .
-
-Empty emphasis ** is not __ an emphasis.
-
-Three asterisks are an emphasized asterisk *** .
+an empty emphasis __ ** is not an emphasis

tests/data/escaping.html

@@ -1,6 +1,4 @@
-<p>Here's an <em>emphasis</em> and here's an escaped *emphasis*. Here are also an escaped `code span`, escaped [inline link](http://example.com).</p>
-<p>Here's <code>an escaped \*emphasis\* inside of a code span</code>.</p>
-<p>Here's one inside of a code block:</p>
-<pre><code>An escaped \*emphasis\*.</code></pre>
-<p>Finally, an escaped reference:</p>
-<p>[1]: http://example.com</p>
+<p>escaped *emphasis*.</p>
+<p><code>escaped \*emphasis\* in a code span</code></p>
+<pre><code>escaped \*emphasis\* in a code block</code></pre>
+<p>\ ` * _ { } [ ] ( ) > # + - . !</p>

tests/data/escaping.md

@@ -1,11 +1,7 @@
-Here's an *emphasis* and here's an escaped \*emphasis\*. Here are also an escaped \`code span\`, escaped \[inline link](http://example.com).
+escaped \*emphasis\*.
 
-Here's `an escaped \*emphasis\* inside of a code span`.
+`escaped \*emphasis\* in a code span`
 
-Here's one inside of a code block:
+    escaped \*emphasis\* in a code block
 
-    An escaped \*emphasis\*.
-
-Finally, an escaped reference:
-
-\[1]: http://example.com
+\\ \` \* \_ \{ \} \[ \] \( \) \> \# \+ \- \. \!

tests/data/horizontal_rule.html

@@ -0,0 +1,5 @@
+<hr />
+<hr />
+<hr />
+<hr />
+<hr />

tests/data/horizontal_rule.md

@@ -0,0 +1,9 @@
+---
+
+- - -
+
+   - - -
+
+***
+
+___

tests/data/image_reference.html

@@ -0,0 +1 @@
+<p><img alt="Markdown Logo" src="/md.png"></p>

tests/data/image_reference.md

@@ -0,0 +1,3 @@
+![Markdown Logo][image]
+
+[image]: /md.png

tests/data/implicit_reference.html

@@ -0,0 +1,2 @@
+<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>

tests/data/implicit_reference.md

@@ -0,0 +1,5 @@
+an [implicit] reference link
+
+[implicit]: http://example.com
+
+an [implicit][] reference link with an empty link definition

tests/data/inline_link.html

@@ -1,2 +1,2 @@
-<p>Here's a <a href="http://parsedown.org">link</a>.</p>
-<p>Here's an image link: <a href="http://daringfireball.net/projects/markdown/"><img alt="MD Logo" src="http://parsedown.org/md.png"></a>.</p>
+<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>

tests/data/inline_link.md

@@ -1,3 +1,3 @@
-Here's a [link](http://parsedown.org).
+[link](http://example.com)
 
-Here's an image link: [![MD Logo](http://parsedown.org/md.png)](http://daringfireball.net/projects/markdown/).
+[![MD Logo](http://parsedown.org/md.png)](http://example.com)

tests/data/lazy_blockquote.html

@@ -0,0 +1,4 @@
+<blockquote>
+<p>quote
+the rest of it</p>
+</blockquote>

tests/data/lazy_blockquote.md

@@ -0,0 +1,2 @@
+> quote
+the rest of it

tests/data/lazy_list.html

@@ -0,0 +1,4 @@
+<ul>
+<li>li
+the rest of it</li>
+</ul>

tests/data/lazy_list.md

@@ -0,0 +1,2 @@
+- li
+the rest of it

tests/data/line_break.html

@@ -0,0 +1,2 @@
+<p>line<br />
+line</p>

tests/data/line_break.md

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

tests/data/ordered_list.html

@@ -1,16 +1,13 @@
-<p>Here's a regular ordered list:</p>
-<ol>
-<li>one</li>
-<li>two</li>
-<li>three</li>
-</ol>
-<p>Here's one with repeating numbers:</p>
 <ol>
 <li>one</li>
 <li>two</li>
 </ol>
-<p>Here's one with large numbers:</p>
+<p>repeating numbers:</p>
 <ol>
 <li>one</li>
 <li>two</li>
 </ol>
+<p>large numbers:</p>
+<ol>
+<li>one</li>
+</ol>

tests/data/ordered_list.md

@@ -1,16 +1,11 @@
-Here's a regular ordered list:
-
 1. one
 2. two
-3. three
 
-Here's one with repeating numbers:
+repeating numbers:
 
 1. one
 1. two
 
-Here's one with large numbers:
+large numbers:
 
 123. one
-123. two
-

tests/data/paragraph_list.html

@@ -1,5 +0,0 @@
-<p>Here's a list that's "inside" a paragraph:</p>
-<ul>
-<li>list item</li>
-<li>another list item</li>
-</ul>

tests/data/paragraph_list.md

@@ -1,4 +0,0 @@
-Here's a list that's "inside" a paragraph:
-- list item
-- another list item
-

tests/data/quote_block.html

@@ -1,20 +0,0 @@
-<p>Here's a regular quote block:</p>
-<blockquote>
-<p>Some quoted text.
-Here goes some more.</p>
-</blockquote>
-<p>Here's one with space before lines:</p>
-<blockquote>
-<p>Some quoted text.
-Here goes some more.</p>
-</blockquote>
-<p>Here's one with no space after >:</p>
-<blockquote>
-<p>Some quoted text.
-Here goes some more.</p>
-</blockquote>
-<p>Here's one with no > on the second line:</p>
-<blockquote>
-<p>Some quoted text.
-Here goes some more.</p>
-</blockquote>

tests/data/quote_block.md

@@ -1,19 +0,0 @@
-Here's a regular quote block:
-
-> Some quoted text.
-> Here goes some more.
-
-Here's one with space before lines:
-
-   > Some quoted text.
-   > Here goes some more.
-
-Here's one with no space after >:
-
->Some quoted text.
->Here goes some more.
-
-Here's one with no > on the second line:
-
-> Some quoted text.
-Here goes some more.

tests/data/reference_link.html

@@ -1,8 +0,0 @@
-<p>Here's a <a href="http://parsedown.org">reference link</a>.</p>
-<p>Here's <a href="http://parsedown.org">one</a> with an alternative syntax.</p>
-<p>Here's <a href="http://parsedown.org">one</a> on the next line.</p>
-<p>Here's <a href="http://parsedown.org">one</a> on 2 lines.</p>
-<p>Here's <a href="http://parsedown.org/tests/">one</a> with a different URL.</p>
-<p>Here's <a href="http://parsedown.org">one</a> with a semantic name.</p>
-<p>Here's [one][404] with no definition.</p>
-<p>Here's an image: <img alt="Markdown Logo" src="https://raw.github.com/dcurtis/markdown-mark/master/png/32x20-solid.png"></p>

tests/data/reference_link.md

@@ -1,29 +0,0 @@
-Here's a [reference link][1].
-
-[1]: http://parsedown.org
-
-Here's [one] [2] with an alternative syntax.
-
-[2] :http://parsedown.org
-
-Here's [one][3] on the next line.
-[3]: http://parsedown.org
-
-Here's [one][4] on 2 lines.
-
-[4]:
-http://parsedown.org
-
-Here's [one][5] with a different URL.
-
-[5]: http://parsedown.org/tests/
-
-Here's [one][the website] with a semantic name.
-
-[the website]: http://parsedown.org
-
-Here's [one][404] with no definition.
-
-Here's an image: ![Markdown Logo][image]
-
-[image]: https://raw.github.com/dcurtis/markdown-mark/master/png/32x20-solid.png

tests/data/setext_header.html

@@ -0,0 +1,5 @@
+<h1>h1</h1>
+<h2>h2</h2>
+<h2>single character</h2>
+<p>not a header</p>
+<hr />

tests/data/setext_header.md

@@ -0,0 +1,12 @@
+h1
+==
+
+h2
+--
+
+single character
+-
+
+not a header
+
+------------

tests/data/setext_heading.html

@@ -1,6 +0,0 @@
-<h1>Heading 1</h1>
-<h2>Heading 2</h2>
-<h2>Block Heading</h2>
-<p>This is the rest of the block.</p>
-<h1>Single "="</h1>
-<h2>Single "-"</h2>

tests/data/setext_heading.md

@@ -1,16 +0,0 @@
-Heading 1
-=========
-
-Heading 2
----------
-
-Block Heading
--------------
-This is the rest of the block.
-
-Single "="
-=
-
-Single "-"
--
-

tests/data/simple_blockquote.html

@@ -0,0 +1,11 @@
+<blockquote>
+<p>quote</p>
+</blockquote>
+<p>indented:</p>
+<blockquote>
+<p>quote</p>
+</blockquote>
+<p>no space after <code>&gt;</code>:</p>
+<blockquote>
+<p>quote</p>
+</blockquote>

tests/data/simple_blockquote.md

@@ -0,0 +1,7 @@
+> quote
+
+indented:
+   > quote
+
+no space after `>`:
+>quote

tests/data/span-level_html.html

@@ -0,0 +1,3 @@
+<p>an <b>important</b> <a href=''>link</a></p>
+<p>broken<br/>
+line</p>

tests/data/span-level_html.md

@@ -0,0 +1,4 @@
+an <b>important</b> <a href=''>link</a>
+
+broken<br/>
+line

tests/data/sparse_list.html

@@ -1,14 +1,15 @@
-<p>Here's a list where items are separated by empty lines:</p>
 <ul>
 <li>
-<p>list item</p>
+<p>li</p>
 </li>
-<li>another list item</li>
+<li>li</li>
 </ul>
-<p>Here's an ordered one:</p>
-<ol>
+<hr />
+<ul>
 <li>
-<p>item one</p>
+<p>li</p>
+<ul>
+<li>indented li</li>
+</ul>
 </li>
-<li>item two</li>
-</ol>
+</ul>

tests/data/sparse_list.md

@@ -1,11 +1,9 @@
-Here's a list where items are separated by empty lines:
+- li
 
-- list item
+- li
 
-- another list item
+---
 
-Here's an ordered one:
+- li
 
-1. item one
-
-2. item two
+    - indented li

tests/data/special_characters.html

@@ -0,0 +1,7 @@
+<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="/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

@@ -0,0 +1,15 @@
+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> 
+
+[inline link](/script?a=1&b=2)
+
+[reference link][1]
+
+[1]: http://example.com/?a=1&b=2

tests/data/text_reference.html

@@ -0,0 +1,6 @@
+<p><a href="http://example.com">reference link</a></p>
+<p><a href="http://example.com">one</a> with a semantic name</p>
+<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>

tests/data/text_reference.md

@@ -0,0 +1,16 @@
+[reference link][1]
+
+[1]: http://example.com
+
+[one][website] with a semantic name
+
+[website]: http://example.com
+
+[one][404] with no definition
+
+[multiline
+one][website] defined on 2 lines
+
+[one][label] with an upper case label
+
+[LABEL]: http://example.com

tests/data/unordered_list.html

@@ -1,20 +1,10 @@
-<p>Here's a regular unordered list:</p>
 <ul>
-<li>list item</li>
-<li>another list item</li>
-<li>3rd list item</li>
+<li>li</li>
+<li>li</li>
 </ul>
-<p>Here's one with a variety of markers:</p>
+<p>mixed markers:</p>
 <ul>
-<li>hyphen</li>
-<li>plus</li>
-<li>asterisk</li>
+<li>li</li>
+<li>li</li>
+<li>li</li>
 </ul>
-<p>Here's one with white space around items:</p>
-<ul>
-<li>list item  </li>
-<li>another list item  </li>
-</ul>
-<p>Here's one with no space after markers:</p>
-<p>-list item
--another list item</p>

tests/data/unordered_list.md

@@ -1,21 +1,8 @@
-Here's a regular unordered list:
+- li
+- li
 
-- list item
-- another list item
-- 3rd list item
+mixed markers:
 
-Here's one with a variety of markers:
-
-- hyphen
-+ plus
-* asterisk
-
-Here's one with white space around items:
-
-  - list item  
-  - another list item  
-
-Here's one with no space after markers:
-
--list item
--another list item
+* li
++ li
+- li

tests/data/whitespace.html

@@ -0,0 +1 @@
+<pre><code>code</code></pre>

tests/data/whitespace.md

@@ -0,0 +1,5 @@
+    
+
+    code
+
+