HTML API: Ensure slash in attribute value is not a self-closing flag.

A trailing slash in an unquoted attribute value was incorrectly treated as a self-closing flag. For example, `<div id=test/>` is a tag with the `id` attribute value `test/`, not a self-closing tag.

Developed in WordPress#12319.

Props jonsurrell, dmsnell.
See #65372.


git-svn-id: https://develop.svn.wordpress.org/trunk@62595 602fd350-edb4-49c9-b593-d223f7449a82

Author: sirreal

src/wp-includes/html-api/class-wp-html-tag-processor.php

@@ -627,6 +627,15 @@ class WP_HTML_Tag_Processor {
 	 */
 	private $token_length;
 
+	/**
+	 * Whether the current tag token has the self-closing flag.
+	 *
+	 * @since 7.1.0
+	 *
+	 * @var bool
+	 */
+	private $has_self_closing_flag = false;
+
 	/**
 	 * Byte offset in input document where current tag name starts.
 	 *
@@ -1074,11 +1083,12 @@ private function base_class_next_token(): bool {
 		 * the closing tag to point to the opening of the special atomic
 		 * tag sequence.
 		 */
-		$tag_name_starts_at   = $this->tag_name_starts_at;
-		$tag_name_length      = $this->tag_name_length;
-		$tag_ends_at          = $this->token_starts_at + $this->token_length;
-		$attributes           = $this->attributes;
-		$duplicate_attributes = $this->duplicate_attributes;
+		$tag_name_starts_at    = $this->tag_name_starts_at;
+		$tag_name_length       = $this->tag_name_length;
+		$tag_ends_at           = $this->token_starts_at + $this->token_length;
+		$has_self_closing_flag = $this->has_self_closing_flag;
+		$attributes            = $this->attributes;
+		$duplicate_attributes  = $this->duplicate_attributes;
 
 		// Find the closing tag if necessary.
 		switch ( $tag_name ) {
@@ -1128,14 +1138,15 @@ private function base_class_next_token(): bool {
 		 * functions that skip the contents have moved all the internal cursors past
 		 * the inner content of the tag.
 		 */
-		$this->token_starts_at      = $was_at;
-		$this->token_length         = $this->bytes_already_parsed - $this->token_starts_at;
-		$this->text_starts_at       = $tag_ends_at;
-		$this->text_length          = $this->tag_name_starts_at - $this->text_starts_at;
-		$this->tag_name_starts_at   = $tag_name_starts_at;
-		$this->tag_name_length      = $tag_name_length;
-		$this->attributes           = $attributes;
-		$this->duplicate_attributes = $duplicate_attributes;
+		$this->token_starts_at       = $was_at;
+		$this->token_length          = $this->bytes_already_parsed - $this->token_starts_at;
+		$this->text_starts_at        = $tag_ends_at;
+		$this->text_length           = $this->tag_name_starts_at - $this->text_starts_at;
+		$this->tag_name_starts_at    = $tag_name_starts_at;
+		$this->tag_name_length       = $tag_name_length;
+		$this->has_self_closing_flag = $has_self_closing_flag;
+		$this->attributes            = $attributes;
+		$this->duplicate_attributes  = $duplicate_attributes;
 
 		return true;
 	}
@@ -2140,13 +2151,34 @@ private function parse_next_attribute(): bool {
 		$doc_length = strlen( $this->html );
 
 		// Skip whitespace and slashes.
-		$this->bytes_already_parsed += strspn( $this->html, " \t\f\r\n/", $this->bytes_already_parsed );
+		$skipped_length              = strspn( $this->html, " \t\f\r\n/", $this->bytes_already_parsed );
+		$this->bytes_already_parsed += $skipped_length;
 		if ( $this->bytes_already_parsed >= $doc_length ) {
 			$this->parser_state = self::STATE_INCOMPLETE_INPUT;
 
 			return false;
 		}
 
+		/**
+		 * This block serves two purposes:
+		 *
+		 * - A fast path for common tag-ending `>`.
+		 * - A check for the self-closing flag which must appear as `/>`.
+		 *
+		 * In a tag like `<g attr=/>`, `/` is the attribute value, not a self-closing
+		 * flag. When it appears in this form, the parser has already consumed the
+		 * attribute value, `$skipped_length` is 0, and this checks below correctly
+		 * identify whether there is a self-closing flag.
+		 *
+		 * Note: Both start and end tags may have the self-closing flag.
+		 */
+		if ( '>' === $this->html[ $this->bytes_already_parsed ] ) {
+			if ( $skipped_length > 0 && '/' === $this->html[ $this->bytes_already_parsed - 1 ] ) {
+				$this->has_self_closing_flag = true;
+			}
+			return false;
+		}
+
 		/*
 		 * Treat the equal sign as a part of the attribute
 		 * name if it is the first encountered byte.
@@ -2324,6 +2356,7 @@ private function after_tag(): void {
 
 		$this->token_starts_at          = null;
 		$this->token_length             = null;
+		$this->has_self_closing_flag    = false;
 		$this->tag_name_starts_at       = null;
 		$this->tag_name_length          = null;
 		$this->text_starts_at           = 0;
@@ -3332,15 +3365,7 @@ public function has_self_closing_flag(): bool {
 			return false;
 		}
 
-		/*
-		 * The self-closing flag is the solidus at the _end_ of the tag, not the beginning.
-		 *
-		 * Example:
-		 *
-		 *     <figure />
-		 *             ^ this appears one character before the end of the closing ">".
-		 */
-		return '/' === $this->html[ $this->token_starts_at + $this->token_length - 2 ];
+		return $this->has_self_closing_flag;
 	}
 
 	/**

tests/phpunit/tests/html-api/wpHtmlProcessor.php

@@ -583,6 +583,29 @@ public function test_expects_closer_foreign_content_self_closing() {
 		$this->assertTrue( $processor->expects_closer() );
 	}
 
+	/**
+	 * Ensures a slash-only unquoted attribute value does not close foreign content.
+	 *
+	 * @ticket 65372
+	 */
+	public function test_unquoted_slash_attribute_does_not_self_close_foreign_content(): void {
+		$processor = WP_HTML_Processor::create_fragment( '<math><mi a=/>math:mi is not self-closing, it has [a="/"] attribute.' );
+
+		$this->assertTrue( $processor->next_tag( 'MI' ), 'Failed to find the MI tag: check test setup.' );
+		$this->assertSame( '/', $processor->get_attribute( 'a' ), 'Failed to treat the slash as the unquoted attribute value.' );
+		$this->assertFalse(
+			$processor->has_self_closing_flag(),
+			'Failed to avoid interpreting the slash-only unquoted attribute value as a self-closing flag.'
+		);
+
+		$this->assertTrue( $processor->next_token(), 'Failed to find text following the MI tag: check test setup.' );
+		$this->assertSame(
+			array( 'HTML', 'BODY', 'MATH', 'MI', '#text' ),
+			$processor->get_breadcrumbs(),
+			'Failed to keep text following the MI tag inside the MI element.'
+		);
+	}
+
 	/**
 	 * Ensures that expects_closer works for void-like elements in foreign content.
 	 *

tests/phpunit/tests/html-api/wpHtmlTagProcessor.php

@@ -111,10 +111,13 @@ public static function data_has_self_closing_flag() {
 			'No self-closing flag on a foreign element'  => array( '<circle>', false ),
 			// These involve syntax peculiarities.
 			'Self-closing flag after extra spaces'       => array( '<div      />', true ),
-			'Self-closing flag after attribute'          => array( '<div id=test/>', true ),
+			'Self-closing flag after attribute'          => array( '<div id=test />', true ),
+			'Slash inside unquoted attribute value'      => array( '<div id=test/>', false ),
+			'Slash only unquoted attribute value'        => array( '<div attr=/>', false ),
+			'Attribute "=" with value ""'                => array( '<div =/>', false ),
 			'Self-closing flag after quoted attribute'   => array( '<div id="test"/>', true ),
 			'Self-closing flag after boolean attribute'  => array( '<div enabled/>', true ),
-			'Boolean attribute that looks like a self-closer' => array( '<div / >', false ),
+			'Ignored "/" and whitespace'                 => array( '<div / >', false ),
 		);
 	}