[BUGFIX] Ignore HTML comments between TYPO3SEARCH markers

The extractor was not working correctly, if there were HTML comments in between
the markers. This change should also make extraction process more performant.

Change-Id: I3761951a594bad138776549a5bbd6fb8c894c100

Author: botweb

Classes/Typo3PageContentExtractor.php

@@ -69,27 +69,11 @@ public function getIndexableContent() {
 	 * @return	string	HTML markup found between TYPO3SEARCH markers
 	 */
 	protected function extractContentMarkedForIndexing($html) {
-		$explodedContent  = preg_split('/\<\!\-\-[\s]?TYPO3SEARCH_/', $html);
-		$indexableContent = '';
-
-		if(count($explodedContent) > 1) {
-
-			foreach($explodedContent as $explodedContentPart) {
-				$contentPart = explode('-->', $explodedContentPart, 2);
-
-				if (trim($contentPart[0]) == 'begin') {
-					$indexableContent .= $contentPart[1];
-					$previousExplodedContentPart = '';
-				} elseif (trim($contentPart[0]) == 'end') {
-					$indexableContent .= $previousExplodedContentPart;
-				} else {
-					$previousExplodedContentPart = $explodedContentPart;
-				}
-			}
-		} else {
-			if ($GLOBALS['TSFE']->tmpl->setup['plugin.']['tx_solr.']['logging.']['indexing.']['missingTypo3SearchMarkers']) {
-				t3lib_div::devLog('No TYPO3SEARCH markers found.', 'solr', 2);
-			}
+		preg_match_all('/\<\!\-\-[\s]?TYPO3SEARCH_begin[\s]?\-\-\>(.|\s)*?\<\!\-\-[\s]?TYPO3SEARCH_end[\s]?\-\-\>/mi', $html, $indexableContents);
+		$indexableContent = implode($indexableContents[0], '');
+		
+		if(empty($indexableContent) && $GLOBALS['TSFE']->tmpl->setup['plugin.']['tx_solr.']['logging.']['indexing.']['missingTypo3SearchMarkers']) {
+			t3lib_div::devLog('No TYPO3SEARCH markers found.', 'solr', 2);
 		}
 
 		return $indexableContent;