[BUGFIX] Correctly consider nested tags in frontend HTML parser

Nested tags starting with the same literals, like `<s><span>...` or `<a ...><abbr>...` are not correctly nested due to a flaw in identifying proper start and end of HTML tags. Resolves: #91194 Releases: master Change-Id: I2029c8e01d66e5286790fd04a259153cd130c753 Reviewed-on: https://review.typo3.org/c/Packages/TYPO3.CMS/+/64409 Tested-by: TYPO3com <noreply@typo3.com> Tested-by: Oliver Hader <oliver.hader@typo3.org> Tested-by: Markus Klein <markus.klein@typo3.org> Tested-by: Susanne Moog <look@susi.dev> Reviewed-by: Markus Klein <markus.klein@typo3.org> Reviewed-by: Helmut Hummel <typo3@helhum.io> Reviewed-by: Daniel Goerz <daniel.goerz@posteo.de> Reviewed-by: Oliver Hader <oliver.hader@typo3.org> Reviewed-by: Susanne Moog <look@susi.dev>
TYPO3 · May 8, 2020 · 03f5310 · 03f5310
1 parent ee06649
commit 03f5310
Show file tree

Hide file tree

Showing 2 changed files with 116 additions and 161 deletions.
diff --git a/typo3/sysext/frontend/Classes/ContentObject/ContentObjectRenderer.php b/typo3/sysext/frontend/Classes/ContentObject/ContentObjectRenderer.php
@@ -3932,7 +3932,8 @@ public function _parseFunc($theValue, $conf)
                     // call this method recursively if found
                     if (strpos($data, '<') !== false && isset($conf['tags.']) && is_array($conf['tags.'])) {
                         foreach ($conf['tags.'] as $tag => $tagConfig) {
-                            if (strpos($data, '<' . $tag) !== false) {
+                            // only match tag `a` in `<a href"...">` but not in `<abbr>`
+                            if (preg_match('#<' . $tag . '[\s/>]#', $data)) {
                                 $data = $this->_parseFunc($data, $conf);
                                 break;
                             }
@@ -7113,7 +7114,13 @@ protected function getContentLengthOfCurrentTag(string $theValue, int $pointer,
         // Take care for nested tags
         do {
             $nextMatchingEndTagPosition = strpos($tempContent, $endTag);
-            $nextSameTypeTagPosition = strpos($tempContent, $startTag);
+            // only match tag `a` in `<a href"...">` but not in `<abbr>`
+            $nextSameTypeTagPosition = preg_match(
+                '#' . $startTag . '[\s/>]#',
+                $tempContent,
+                $nextSameStartTagMatches,
+                PREG_OFFSET_CAPTURE
+            ) ? $nextSameStartTagMatches[0][1] : false;
 
             // filter out nested tag contents to help getting the correct closing tag
             if ($nextSameTypeTagPosition !== false && $nextSameTypeTagPosition < $nextMatchingEndTagPosition) {