Merge pull request #4181 from ampproject/fix/4179-duplicate-meta-tag

westonruter · web-flow · commit c431687f6da9 · 2020-01-28T06:21:54.000-08:00
Parse malformed [meta charset] tag
diff --git a/includes/class-amp-theme-support.php b/includes/class-amp-theme-support.php
@@ -2221,7 +2221,7 @@ public static function prepare_response( $response, $args = [] ) {
 		 * Note that the meta charset is supposed to appear within the first 1024 bytes.
 		 * See <https://www.w3.org/International/questions/qa-html-encoding-declarations>.
 		 */
-		if ( ! preg_match( '#<meta[^>]+charset=#i', substr( $response, 0, 1024 ) ) ) {
+		if ( ! preg_match( '#<meta[^>]+charset\s*=#i', substr( $response, 0, 1024 ) ) ) {
 			$meta_charset = sprintf( '<meta charset="%s">', esc_attr( get_bloginfo( 'charset' ) ) );
 
 			$response = preg_replace(
diff --git a/src/Dom/Document.php b/src/Dom/Document.php
@@ -118,7 +118,7 @@ final class Document extends DOMDocument {
 
 	// Regex patterns used for finding tags or extracting attribute values in an HTML string.
 	const HTML_FIND_TAG_WITHOUT_ATTRIBUTE_PATTERN = '/<%1$s[^>]*?>[^<]*(?:<\/%1$s>)?/i';
-	const HTML_FIND_TAG_WITH_ATTRIBUTE_PATTERN    = '/<%1$s [^>]*?\s*%2$s=[^>]*?>[^<]*(?:<\/%1$s>)?/i';
+	const HTML_FIND_TAG_WITH_ATTRIBUTE_PATTERN    = '/<%1$s [^>]*?\s*%2$s\s*=[^>]*?>[^<]*(?:<\/%1$s>)?/i';
 	const HTML_EXTRACT_ATTRIBUTE_VALUE_PATTERN    = '/%s=(?:([\'"])(?<full>.*)?\1|(?<partial>[^ \'";]+))/';
 
 	// Tags constants used throughout.
diff --git a/tests/php/test-tag-and-attribute-sanitizer.php b/tests/php/test-tag-and-attribute-sanitizer.php
@@ -2239,6 +2239,12 @@ public function get_html_data() {
 				[],
 				[ AMP_Tag_And_Attribute_Sanitizer::DISALLOWED_PROPERTY_IN_ATTR_VALUE, AMP_Tag_And_Attribute_Sanitizer::ATTR_REQUIRED_BUT_MISSING ],
 			],
+			'parse_malformed_meta_charset_tag'        => [
+				'<html amp><head><meta charset = "utf-8"></head><body></body></html>',
+				'<html amp><head><meta charset="utf-8"></head><body></body></html>',
+				[],
+				[],
+			],
 			'edge_meta_ua_compatible'                 => [
 				'<html amp><head><meta charset="utf-8"><meta http-equiv="X-UA-Compatible" content="IE=edge"></head><body></body></html>',
 				null, // No change.
