feat: add ckeditor data attribute removal for list items and remove unnecessary whitespaces after opening and before closing tags

Author: Mainbird

Classes/EventListener/ContentMinifierEventListener.php

@@ -16,15 +16,12 @@ public function __invoke(AfterCacheableContentIsGeneratedEvent $event): void
     }
 
     /**
-    * remove javascript inline comments
+    * remove JavaScript inline comments
     * convert linebreaks to spaces
     * convert tabs to spaces
     * convert multiple spaces to one single space
     * remove spaces between tags, but ignore on some inline-tags
     * replace non-HTML5 conform closing tags
-    * remove type attributes for styles and javascript
-    * remove unnecessary whitespaces from class attributes
-    * remove unnecessary whitespaces from JSON-LD
     */
     private function minify(string $content): string
     {
@@ -37,7 +34,18 @@ private function minify(string $content): string
             '/" \/>/' => '">',
         ];
 
-        $content = str_replace(
+        $content = $this->removeUnnecessaryTypeAttributesForStyleAndScriptTags($content);
+        $content = $this->removeUnnecessaryWhitespacesFromClassAttributes($content);
+        $content = $this->removeUnnecessaryWhitespacesForJsonLdSchemas($content);
+        $content = $this->removeCkeditorDataAttributesFromListItems($content);
+        $content = $this->removeWhitespacesAfterTagStartAndBeforeTagClose($content);
+
+        return (string)preg_replace(array_keys($replacements), array_values($replacements), $content);
+    }
+
+    private function removeUnnecessaryTypeAttributesForStyleAndScriptTags(string $content): string
+    {
+        return str_replace(
             [
                 ' type="text/css"',
                 ' type=\'text/css\'',
@@ -47,17 +55,45 @@ private function minify(string $content): string
             '',
             $content
         );
+    }
+
+    /**
+     * @see https://forge.typo3.org/issues/109002
+     * @see https://github.com/ckeditor/ckeditor5/issues/19006
+     */
+    private function removeCkeditorDataAttributesFromListItems(string $content): string
+    {
+        return (string)preg_replace(
+            '/(<li)\s+data-list-item-id="[^"]*"/',
+            '$1',
+            $content
+        );
+    }
+
+    private function removeWhitespacesAfterTagStartAndBeforeTagClose(string $content): string
+    {
+        return (string)preg_replace_callback(
+            '/<(h[1-6]|p|li|td|th|dt|dd|button|label)[^>]*>\K\s+|\s+(?=<\/(h[1-6]|p|li|td|th|dt|dd|button|label)>)/',
+            static fn () => '',
+            $content
+        );
+    }
 
-        $content = (string)preg_replace_callback(
+    private function removeUnnecessaryWhitespacesFromClassAttributes(string $content): string
+    {
+        return (string)preg_replace_callback(
             '/class="([^"]+)"/',
             static function (array $matches) {
                 $cleanedClassList = trim((string)preg_replace('/\s+/', ' ', $matches[1]));
                 return 'class="' . $cleanedClassList . '"';
             },
             $content
         );
+    }
 
-        $content = (string)preg_replace_callback(
+    private function removeUnnecessaryWhitespacesForJsonLdSchemas(string $content): string
+    {
+        return (string)preg_replace_callback(
             '/<script\s+type="application\/ld\+json">(.*?)<\/script>/s',
             static function (array $matches) {
                 $json = trim($matches[1]);
@@ -76,7 +112,5 @@ static function (array $matches) {
             },
             $content
         );
-
-        return (string)preg_replace(array_keys($replacements), array_values($replacements), $content);
     }
 }