add simple html dom implementation

composer.json

@@ -43,7 +43,8 @@
 	"require": {
 		"php": ">=7.3",
 		"cloudflare/cf-ip-rewrite": "^1.0",
-		"composer/installers": "^1.0 || ^2.0"
+		"composer/installers": "^1.0 || ^2.0",
+		"voku/simple_html_dom": "^4.8"
 	},
 	"require-dev": {
 		"php": "^7 || ^8",

inc/Engine/Media/Lazyload/Content/Processor/Processor.php

@@ -11,6 +11,8 @@ public function set_processor( $processor ) {
 			$this->processor = new Dom();
 		} elseif ( 'regex' === $processor ) {
 			$this->processor = new Regex();
+		} elseif ( 'simple_html_dom' === $processor ) {
+			$this->processor = new SimpleHtmlDom();
 		}
 	}
 

inc/Engine/Media/Lazyload/Content/Processor/SimpleHtmlDom.php

@@ -0,0 +1,56 @@
+<?php
+declare(strict_types=1);
+
+namespace WP_Rocket\Engine\Media\Lazyload\Content\Processor;
+
+use voku\helper\HtmlDomParser;
+use voku\helper\SimpleHtmlDomBlank;
+
+class SimpleHtmlDom {
+	public function add_locations_hash_to_html( $html ) {
+		$dom = HtmlDomParser::str_get_html( $html );
+
+		$body = $dom->getElementByTagName( 'body' );
+
+		if ( $body instanceof SimpleHtmlDomBlank ) {
+			return $html;
+		}
+
+		$this->add_hash_to_element( $body, 2 );
+
+		return $dom->save();
+	}
+
+	private function add_hash_to_element( $element, $depth ) {
+		if ( $depth < 0 ) {
+			return;
+		}
+
+		$skip_tags = [
+			'DIV',
+			'MAIN',
+			'FOOTER',
+			'SECTION',
+			'ARTICLE',
+			'HEADER',
+		];
+
+		foreach ( $element->childNodes() as $child ) {
+			if ( ! in_array( strtoupper( $child->getTag() ), $skip_tags, true ) ) {
+				continue;
+			}
+
+			// Calculate the hash of the opening tag.
+			$child_html       = $child->html();
+			$opening_tag_html = strstr( $child_html, '>', true ) . '>';
+
+			$hash = md5( $opening_tag_html );
+
+			// Add the data-rocket-location-hash attribute.
+			$child->setAttribute( 'data-rocket-location-hash', $hash );
+
+			// Recursively process child elements.
+			$this->add_hash_to_element( $child, $depth - 1 );
+		}
+	}
+}

inc/Engine/Media/Lazyload/Content/Subscriber.php

@@ -34,7 +34,7 @@ public static function get_subscribed_events(): array {
 	public function lazyload_content( string $buffer ): string {
 		$start = hrtime( true );
 
-		$this->processor->set_processor( 'regex' );
+		$this->processor->set_processor( 'simple_html_dom' );
 
 		$buffer = $this->processor->get_processor()->add_locations_hash_to_html( $buffer );