HTML API: Apply HTML specification change to DOCTYPE parsing.

sirreal · sirreal · commit 8512f0f704af · 2026-07-08T11:15:31.000Z
The HTML standard was updated to remove a distinction between empty and missing SYSTEM identifiers for the purposes of determining the mode of a document. See whatwg/html#12023. Developed in WordPress#12382 Props jonsurrell, dmsnell. Fixes #64504. git-svn-id: https://develop.svn.wordpress.org/trunk@62666 602fd350-edb4-49c9-b593-d223f7449a82
diff --git a/src/wp-includes/html-api/class-wp-html-doctype-info.php b/src/wp-includes/html-api/class-wp-html-doctype-info.php
@@ -49,6 +49,8 @@
  * @see https://www.iso.org/standard/16387.html
  *
  * @since 6.7.0
+ * @since 7.1.0 Spec update: missing and empty SYSTEM identifiers are handled
+ *              the same for determining the document mode.
  *
  * @access private
  *
@@ -229,13 +231,9 @@ private function __construct(
 		 * >
 		 * > The system identifier and public identifier strings must be compared...
 		 * > in an ASCII case-insensitive manner.
-		 * >
-		 * > A system identifier whose value is the empty string is not considered missing
-		 * > for the purposes of the conditions above.
 		 */
-		$system_identifier_is_missing = null === $system_identifier;
-		$public_identifier            = null === $public_identifier ? '' : strtolower( $public_identifier );
-		$system_identifier            = null === $system_identifier ? '' : strtolower( $system_identifier );
+		$public_identifier = null === $public_identifier ? '' : strtolower( $public_identifier );
+		$system_identifier = null === $system_identifier ? '' : strtolower( $system_identifier );
 
 		/*
 		 * > The public identifier is set to…
@@ -335,10 +333,11 @@ private function __construct(
 		}
 
 		/*
-		 * > The system identifier is missing and the public identifier starts with…
+		 * > The system identifier is missing or the empty string, and the
+		 * > public identifier starts with…
 		 */
 		if (
-			$system_identifier_is_missing && (
+			'' === $system_identifier && (
 				str_starts_with( $public_identifier, '-//w3c//dtd html 4.01 frameset//' ) ||
 				str_starts_with( $public_identifier, '-//w3c//dtd html 4.01 transitional//' )
 			)
@@ -364,10 +363,11 @@ private function __construct(
 		}
 
 		/*
-		 * > The system identifier is not missing and the public identifier starts with…
+		 * > The system identifier is neither missing nor the empty string, and the
+		 * > public identifier starts with…
 		 */
 		if (
-			! $system_identifier_is_missing && (
+			'' !== $system_identifier && (
 				str_starts_with( $public_identifier, '-//w3c//dtd html 4.01 frameset//' ) ||
 				str_starts_with( $public_identifier, '-//w3c//dtd html 4.01 transitional//' )
 			)
diff --git a/tests/phpunit/tests/html-api/wpHtmlDoctypeInfo.php b/tests/phpunit/tests/html-api/wpHtmlDoctypeInfo.php
@@ -87,7 +87,8 @@ public static function data_parseable_raw_doctypes(): array {
 			'Emoji'                                     => array( '<!DOCTYPE 🏴󠁧󠁢󠁥󠁮󠁧󠁿 PUBLIC "🔥" "😈">', 'quirks', "\u{1F3F4}\u{E0067}\u{E0062}\u{E0065}\u{E006E}\u{E0067}\u{E007F}", '🔥', '😈' ),
 			'Bogus characters instead of SYSTEM quote after public' => array( "<!DOCTYPE html PUBLIC ''x''>", 'quirks', 'html', '' ),
 			'Special quirks mode if system unset'       => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//">', 'quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//' ),
-			'Special limited-quirks mode if system set' => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//" "">', 'limited-quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//', '' ),
+			'Special quirks mode if system empty'       => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//" "">', 'quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//', '' ),
+			'Special limited-quirks mode if system is non-empty' => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//" "non-empty">', 'limited-quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//', 'non-empty' ),
 		);
 	}
 

Original file line number	Diff line number	Diff line change
`@@ -87,7 +87,8 @@ public static function data_parseable_raw_doctypes(): array {`
`87`	`87`	`'Emoji' => array( '<!DOCTYPE 🏴󠁧󠁢󠁥󠁮󠁧󠁿 PUBLIC "🔥" "😈">', 'quirks', "\u{1F3F4}\u{E0067}\u{E0062}\u{E0065}\u{E006E}\u{E0067}\u{E007F}", '🔥', '😈' ),`
`88`	`88`	`'Bogus characters instead of SYSTEM quote after public' => array( "<!DOCTYPE html PUBLIC ''x''>", 'quirks', 'html', '' ),`
`89`	`89`	`'Special quirks mode if system unset' => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//">', 'quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//' ),`
`90`		`- 'Special limited-quirks mode if system set' => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//" "">', 'limited-quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//', '' ),`
	`90`	`+ 'Special quirks mode if system empty' => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//" "">', 'quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//', '' ),`
	`91`	`+ 'Special limited-quirks mode if system is non-empty' => array( '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//" "non-empty">', 'limited-quirks', 'html', '-//W3C//DTD HTML 4.01 Frameset//', 'non-empty' ),`
`91`	`92`	`);`
`92`	`93`	`}`
`93`	`94`