Merge branch 'markdown-extra'

* markdown-extra:
  finished table support for GFM and markdown Extra
  Markdown Extra tables
  added context information to the parser
  markdown extra special attributes on code blocks
  implemented markdown extra attributes
  created MarkdownExtra class
  tests for gfm tables

Conflicts:
	Parser.php
	phpunit.xml.dist

GithubMarkdown.php

@@ -7,30 +7,57 @@
 
 namespace cebe\markdown;
 
+use cebe\markdown\block\TableTrait;
+
 /**
  * Markdown parser for github flavored markdown.
  *
  * @author Carsten Brandt <mail@cebe.cc>
  */
 class GithubMarkdown extends Markdown
 {
+	use TableTrait;
+
 	/**
 	 * @var boolean whether to interpret newlines as `<br />`-tags.
 	 * This feature is useful for comments where newlines are often meant to be real new lines.
 	 */
 	public $enableNewlines = false;
 
+	/**
+	 * @inheritDoc
+	 */
+	protected $escapeCharacters = [
+		// from Markdown
+		'\\', // backslash
+		'`', // backtick
+		'*', // asterisk
+		'_', // underscore
+		'{', '}', // curly braces
+		'[', ']', // square brackets
+		'(', ')', // parentheses
+		'#', // hash mark
+		'+', // plus sign
+		'-', // minus sign (hyphen)
+		'.', // dot
+		'!', // exclamation mark
+		'<', '>',
+		// added by MarkdownExtra
+		':', // colon
+		'|', // pipe
+	];
+
 	/**
 	 * @inheritDoc
 	 */
 	protected function inlineMarkers()
 	{
-		$markers = [
+		return parent::inlineMarkers() + [
 			'http'  => 'parseUrl',
 			'ftp'   => 'parseUrl',
 			'~~'    => 'parseStrike',
+			'|'     => 'parseTd',
 		];
-		return array_merge(parent::inlineMarkers(), $markers);
 	}
 
 
@@ -45,6 +72,9 @@ protected function identifyLine($lines, $current)
 		if (isset($lines[$current]) && (strncmp($lines[$current], '```', 3) === 0 || strncmp($lines[$current], '~~~', 3) === 0)) {
 			return 'fencedCode';
 		}
+		if ($this->identifyTable($lines, $current)) {
+			return 'table';
+		}
 		return parent::identifyLine($lines, $current);
 	}
 
@@ -105,7 +135,7 @@ protected function parseUrl($markdown)
 			)/x
 REGEXP;
 
-		if (preg_match($pattern, $markdown, $matches)) {
+		if (!in_array('parseLink', $this->context) && preg_match($pattern, $markdown, $matches)) {
 			$href = htmlspecialchars($matches[0], ENT_COMPAT | ENT_HTML401, 'UTF-8');
 			$text = htmlspecialchars(urldecode($matches[0]), ENT_NOQUOTES, 'UTF-8');
 			return [

Markdown.php

@@ -225,7 +225,7 @@ protected function identifyHeadline($lines, $current)
 	/**
 	 * Consume lines for a paragraph
 	 */
-	public function consumeParagraph($lines, $current)
+	protected function consumeParagraph($lines, $current)
 	{
 		// consume until newline or intended line
 
@@ -598,19 +598,22 @@ protected function parseEntity($text)
 	protected function parseLt($text)
 	{
 		if (strpos($text, '>') !== false) {
-			if (preg_match('/^<([^\s]*?@[^\s]*?\.\w+?)>/', $text, $matches)) {
-				// email address
-				$email = htmlspecialchars($matches[1], ENT_NOQUOTES, 'UTF-8');
-				return [
-					"<a href=\"mailto:$email\">$email</a>", // TODO encode mail with entities
-					strlen($matches[0])
-				];
-			} elseif (preg_match('/^<([a-z]{3,}:\/\/[^\s]+?)>/', $text, $matches)) {
-				// URL
-				$url = htmlspecialchars($matches[1], ENT_COMPAT | ENT_HTML401, 'UTF-8');
-				$text = htmlspecialchars(urldecode($matches[1]), ENT_NOQUOTES, 'UTF-8');
-				return ["<a href=\"$url\">$text</a>", strlen($matches[0])];
-			} elseif (preg_match('~^</?(\w+\d?)( .*?)?>~', $text, $matches)) {
+			if (!in_array('parseLink', $this->context)) { // do not allow links in links
+				if (preg_match('/^<([^\s]*?@[^\s]*?\.\w+?)>/', $text, $matches)) {
+					// email address
+					$email = htmlspecialchars($matches[1], ENT_NOQUOTES, 'UTF-8');
+					return [
+						"<a href=\"mailto:$email\">$email</a>", // TODO encode mail with entities
+						strlen($matches[0])
+					];
+				} elseif (preg_match('/^<([a-z]{3,}:\/\/[^\s]+?)>/', $text, $matches)) {
+					// URL
+					$url = htmlspecialchars($matches[1], ENT_COMPAT | ENT_HTML401, 'UTF-8');
+					$text = htmlspecialchars(urldecode($matches[1]), ENT_NOQUOTES, 'UTF-8');
+					return ["<a href=\"$url\">$text</a>", strlen($matches[0])];
+				}
+			}
+			if (preg_match('~^</?(\w+\d?)( .*?)?>~', $text, $matches)) {
 				// HTML tags
 				return [$matches[0], strlen($matches[0])];
 			} elseif (preg_match('~^<!--.*?-->~', $text, $matches)) {
@@ -645,7 +648,7 @@ protected function parseEscape($text)
 	 */
 	protected function parseLink($markdown)
 	{
-		if (($parts = $this->parseLinkOrImage($markdown)) !== false) {
+		if (!in_array('parseLink', array_slice($this->context, 1)) && ($parts = $this->parseLinkOrImage($markdown)) !== false) {
 			list($text, $url, $title, $offset) = $parts;
 
 			$link = '<a href="' . htmlspecialchars($url, ENT_COMPAT | ENT_HTML401, 'UTF-8') . '"'
@@ -712,8 +715,9 @@ protected function parseLinkOrImage($markdown)
 					isset($refMatches[2]) ? $refMatches[2] : '', // url
 					empty($refMatches[5]) ? null: $refMatches[5], // title
 					$offset + strlen($refMatches[0]), // offset
+					null, // reference key
 				];
-			} elseif (preg_match('/^[ \n]?(\[(.*?)\])?/', $markdown, $refMatches)) {
+			} elseif (preg_match('/^([ \n]?\[(.*?)\])?/', $markdown, $refMatches)) {
 				// reference style link
 				if (empty($refMatches[2])) {
 					$key = strtolower($text);
@@ -726,6 +730,7 @@ protected function parseLinkOrImage($markdown)
 						$this->references[$key]['url'], // url
 						empty($this->references[$key]['title']) ? null: $this->references[$key]['title'], // title
 						$offset + strlen($refMatches[0]), // offset
+						$key,
 					];
 				}
 			}