Remove support for Processing Instructions

Attempting to parse processing instructions conflicts with parsing bogus
comments when a document may be incomplete, which might create a
divergence in the HTML API from browser behavior.

Author: dmsnell

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

@@ -321,7 +321,6 @@
  *
  * And there are non-elements which are atomic in nature but have no modifiable text.
  *  - `DOCTYPE` nodes like `<DOCTYPE html>` which have no closing tag.
- *  - XML Processing instruction nodes like `<?xml charset="utf8"?>`.
  *  - The empty end tag `</>` which is ignored in the browser and DOM but exposed
  *    to the HTML API.
  *
@@ -483,7 +482,6 @@ class WP_HTML_Tag_Processor {
 	 * | *Matched tag*   | Found an HTML tag; it's possible to modify its attributes.           |
 	 * | *Text node*     | Found a #text node; this is plaintext and modifiable.                |
 	 * | *CDATA node*    | Found a CDATA section; this is modifiable.                           |
-	 * | *PI node*       | Found a Processing Instruction; this is modifiable.                  |
 	 * | *Comment*       | Found a comment or bogus comment; this is modifiable.                |
 	 * | *Presumptuous*  | Found an empty tag closer: `</>`.                                    |
 	 * | *Funky comment* | Found a tag closer with an invalid tag name; this is modifiable.     |
@@ -496,7 +494,6 @@ class WP_HTML_Tag_Processor {
 	 * @see WP_HTML_Tag_Processor::STATE_MATCHED_TAG
 	 * @see WP_HTML_Tag_Processor::STATE_TEXT_NODE
 	 * @see WP_HTML_Tag_Processor::STATE_CDATA_NODE
-	 * @see WP_HTML_Tag_Processor::STATE_PI_NODE
 	 * @see WP_HTML_Tag_Processor::STATE_COMMENT
 	 * @see WP_HTML_Tag_Processor::STATE_DOCTYPE
 	 * @see WP_HTML_Tag_Processor::STATE_PRESUMPTUOUS_TAG
@@ -806,7 +803,6 @@ public function next_tag( $query = null ) {
 	 *  - a text node - the plaintext inside tags.
 	 *  - an HTML comment.
 	 *  - a DOCTYPE declaration.
-	 *  - a processing instruction, e.g. `<?xml version="1.0" ?>`.
 	 *
 	 * The Tag Processor currently only supports the tag token.
 	 *
@@ -1723,9 +1719,6 @@ private function parse_next_tag() {
 			/*
 			 * <? transitions to a bogus comment state – skip to the nearest >
 			 * See https://html.spec.whatwg.org/multipage/parsing.html#tag-open-state
-			 *
-			 * Although this becomes a DOM comment, the Tag Processor is going to treat
-			 * it as a processing instruction to be able to treat it as the raw syntax.
 			 */
 			if ( '?' === $html[ $at + 1 ] ) {
 				$closer_at = strpos( $html, '>', $at + 2 );
@@ -1735,7 +1728,7 @@ private function parse_next_tag() {
 					return false;
 				}
 
-				$this->parser_state         = self::STATE_PI_NODE;
+				$this->parser_state         = self::STATE_COMMENT;
 				$this->token_length         = $closer_at + 1 - $this->token_starts_at;
 				$this->text_starts_at       = $this->token_starts_at + 2;
 				$this->text_length          = $closer_at - $this->text_starts_at;
@@ -2532,7 +2525,6 @@ public function is_tag_closer() {
 	 *  - `#tag` when matched on a tag.
 	 *  - `#text` when matched on a text node.
 	 *  - `#cdata-section` when matched on a CDATA node.
-	 *  - `#processing-instruction` when matched on a processing instruction.
 	 *  - `#comment` when matched on a comment.
 	 *  - `#presumptuous-tag` when matched on an empty tag closer.
 	 *  - `#funky-comment` when matched on a funky comment.
@@ -2549,9 +2541,6 @@ public function get_token_type() {
 			case self::STATE_DOCTYPE:
 				return '#doctype';
 
-			case self::STATE_PI_NODE:
-				return '#processing-instruction';
-
 			default:
 				return $this->get_token_name();
 		}
@@ -2566,7 +2555,6 @@ public function get_token_type() {
 	 *
 	 * Dynamic names:
 	 *  - Uppercase tag name for tag matches.
-	 *  - Tag name for processing instructions.
 	 *  - `html` for DOCTYPE declarations.
 	 *
 	 * Note that if the Tag Processor is not matched on a token
@@ -2589,10 +2577,6 @@ public function get_token_name() {
 			case self::STATE_CDATA_NODE:
 				return '#cdata-section';
 
-			case self::STATE_PI_NODE:
-				// @todo add the PI tag.
-				return '?';
-
 			case self::STATE_COMMENT:
 				return '#comment';
 
@@ -3197,22 +3181,6 @@ private function matches() {
 	 */
 	const STATE_CDATA_NODE = 'STATE_CDATA_NODE';
 
-	/**
-	 * Parser Processing Instruction State.
-	 *
-	 * Indicates that the parser has found a Processing Instruction and
-	 * it's possible to read and modify its modifiable text. Note that in
-	 * HTML there are no Processing Instruction nodes and they are treated
-	 * as HTML comments. Nonetheless, the Tag Processor still recognizes
-	 * them as they appear in the HTML stream and exposes them for
-	 * inspection and modification.
-	 *
-	 * @since 6.5.0
-	 *
-	 * @access private
-	 */
-	const STATE_PI_NODE = 'STATE_PI_NODE';
-
 	/**
 	 * Indicates that the parser has found an HTML comment and it's
 	 * possible to read and modify its modifiable text.

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

@@ -87,7 +87,7 @@ public function test_basic_assertion_element() {
 		);
 
 		$attributes     = $processor->get_attribute_names_with_prefix( '' );
-		$attribute_list = array_map( array( 'self', 'quoted' ), $attributes );
+		$attribute_list = array_map( 'Tests_HtmlApi_WpHtmlProcessor_Token_Scanning::quoted', $attributes );
 		$this->assertSame(
 			array( 'id', 'inert' ),
 			$attributes,
@@ -127,7 +127,7 @@ public function test_basic_assertion_script_element() {
 		);
 
 		$attributes     = $processor->get_attribute_names_with_prefix( '' );
-		$attribute_list = array_map( array( 'self', 'quoted' ), $attributes );
+		$attribute_list = array_map( 'Tests_HtmlApi_WpHtmlProcessor_Token_Scanning::quoted', $attributes );
 		$this->assertSame(
 			array( 'type' ),
 			$attributes,
@@ -178,7 +178,7 @@ public function test_basic_assertion_textarea_element() {
 		);
 
 		$attributes     = $processor->get_attribute_names_with_prefix( '' );
-		$attribute_list = array_map( array( 'self', 'quoted' ), $attributes );
+		$attribute_list = array_map( 'Tests_HtmlApi_WpHtmlProcessor_Token_Scanning::quoted', $attributes );
 		$this->assertSame(
 			array( 'rows', 'cols' ),
 			$attributes,
@@ -224,7 +224,7 @@ public function test_basic_assertion_title_element() {
 		);
 
 		$attributes     = $processor->get_attribute_names_with_prefix( '' );
-		$attribute_list = array_map( array( 'self', 'quoted' ), $attributes );
+		$attribute_list = array_map( 'Tests_HtmlApi_WpHtmlProcessor_Token_Scanning::quoted', $attributes );
 		$this->assertSame(
 			array( 'class' ),
 			$attributes,
@@ -273,7 +273,7 @@ public function test_basic_assertion_rawtext_elements( $tag_name ) {
 		);
 
 		$attributes     = $processor->get_attribute_names_with_prefix( '' );
-		$attribute_list = array_map( array( 'self', 'quoted' ), $attributes );
+		$attribute_list = array_map( 'Tests_HtmlApi_WpHtmlProcessor_Token_Scanning::quoted', $attributes );
 		$this->assertSame(
 			array( 'class' ),
 			$attributes,
@@ -338,48 +338,6 @@ public function test_basic_assertion_cdata_section() {
 		);
 	}
 
-	/**
-	 * Ensures that normative Processing Instruction nodes are properly parsed.
-	 *
-	 * @ticket 60170
-	 *
-	 * @since 6.5.0
-	 *
-	 * @covers WP_HTML_Tag_Processor::next_token
-	 */
-	public function test_basic_assertion_processing_instruction() {
-		$processor = WP_HTML_Processor::create_fragment( '<?xml charset="utf-8">' );
-		$processor->next_token();
-
-		$this->assertSame(
-			'#processing-instruction',
-			$processor->get_token_type(),
-			"Should have found PI node but found {$processor->get_token_type()} instead."
-		);
-
-		$this->assertSame(
-			'xml',
-			$processor->get_token_name(),
-			"Should have found PI tag as name but found {$processor->get_token_name()} instead."
-		);
-
-		$this->assertNull(
-			$processor->get_tag(),
-			'Should not have been able to query tag name on non-element token.'
-		);
-
-		$this->assertNull(
-			$processor->get_attribute( 'type' ),
-			'Should not have been able to query attributes on non-element token.'
-		);
-
-		$this->assertSame(
-			' charset="utf-8"',
-			$processor->get_modifiable_text(),
-			'Found incorrect modifiable text.'
-		);
-	}
-
 	/**
 	 * Ensures that common comments are properly parsed.
 	 *
@@ -436,7 +394,7 @@ public function data_common_comments() {
 		return array(
 			'Shortest comment'    => array( '<!-->', '' ),
 			'Short comment'       => array( '<!--->', '' ),
-			'Invalid PI node'     => array( '<? missing>', ' missing' ),
+			'Invalid PI node'     => array( '<?/missing/>', '/missing/' ),
 			'Invalid ! directive' => array( '<!something else>', 'something else' ),
 		);
 	}