HTML API: Prevent bookmarks from being set on virtual tokens

gziolo · gziolo · commit 146775070985 · 2024-12-10T09:11:35.000Z
Fixes the issue when an HTML_Processor bookmark was set at a virtual token (a node in the resulting document that does not correspond to an HTML token present in the input string), seek behavior was unreliable. Props jonsurrell, gziolo. Fixes #62521. git-svn-id: https://develop.svn.wordpress.org/trunk@59502 602fd350-edb4-49c9-b593-d223f7449a82
diff --git a/src/wp-includes/html-api/class-wp-html-processor.php b/src/wp-includes/html-api/class-wp-html-processor.php
@@ -303,7 +303,9 @@ public static function create_fragment( $html, $context = '<body>', $encoding =
 		}
 
 		while ( $context_processor->next_tag() ) {
-			$context_processor->set_bookmark( 'final_node' );
+			if ( ! $context_processor->is_virtual() ) {
+				$context_processor->set_bookmark( 'final_node' );
+			}
 		}
 
 		if (
@@ -5673,12 +5675,25 @@ public function seek( $bookmark_name ): bool {
 	 * reaching for it, as inappropriate use could lead to broken
 	 * HTML structure or unwanted processing overhead.
 	 *
+	 * Bookmarks cannot be set on tokens that do no appear in the original
+	 * HTML text. For example, the HTML `<table><td>` stops at tags `TABLE`,
+	 * `TBODY`, `TR`, and `TD`. The `TBODY` and `TR` tags do not appear in
+	 * the original HTML and cannot be used as bookmarks.
+	 *
 	 * @since 6.4.0
 	 *
 	 * @param string $bookmark_name Identifies this particular bookmark.
 	 * @return bool Whether the bookmark was successfully created.
 	 */
 	public function set_bookmark( $bookmark_name ): bool {
+		if ( $this->is_virtual() ) {
+			_doing_it_wrong(
+				__METHOD__,
+				__( 'Cannot set bookmarks on tokens that do no appear in the original HTML text.' ),
+				'6.8.0'
+			);
+			return false;
+		}
 		return parent::set_bookmark( "_{$bookmark_name}" );
 	}
 
diff --git a/tests/phpunit/tests/html-api/wpHtmlProcessor-bookmark.php b/tests/phpunit/tests/html-api/wpHtmlProcessor-bookmark.php
@@ -157,4 +157,17 @@ public static function data_processor_constructors(): array {
 			'Fragment parser' => array( array( WP_HTML_Processor::class, 'create_fragment' ) ),
 		);
 	}
+
+	/**
+	 * @ticket 62521
+	 *
+	 * @expectedIncorrectUsage WP_HTML_Processor::set_bookmark
+	 */
+	public function test_bookmarks_not_allowed_on_virtual_nodes() {
+		$processor = WP_HTML_Processor::create_full_parser( 'text' );
+		$this->assertTrue( $processor->next_tag( 'BODY' ) );
+		$this->assertFalse( $processor->set_bookmark( 'mark' ) );
+		$this->assertTrue( $processor->next_token() );
+		$this->assertTrue( $processor->set_bookmark( 'mark' ) );
+	}
 }
