HTML API: Implement set_inner_html

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

@@ -355,6 +355,115 @@ public static function create_full_parser( $html, $known_definite_encoding = 'UT
 		return $processor;
 	}
 
+	public function set_inner_html( ?string $html ) {
+		if ( $this->is_virtual() ) {
+			return false;
+		}
+
+		if ( $this->get_token_type() !== '#tag' ) {
+			return false;
+		}
+
+		if ( $this->is_tag_closer() ) {
+			return false;
+		}
+
+		if ( ! $this->expects_closer() ) {
+			return false;
+		}
+
+		if (
+			'html' !== $this->state->current_token->namespace &&
+			$this->state->current_token->has_self_closing_flag
+		) {
+			return false;
+		}
+
+		if ( null === $html ) {
+			$html = '';
+		}
+		if ( '' !== $html ) {
+			$fragment_parser = $this->spawn_fragment_parser( $html );
+			if (
+				null === $fragment_parser
+			) {
+				return false;
+			}
+
+			try {
+				$html = $fragment_parser->serialize();
+			} catch ( Exception $e ) {
+				return false;
+			}
+		}
+
+		// @todo apply modifications if there are any???
+
+		if ( ! parent::set_bookmark( 'SET_INNER_HTML: opener' ) ) {
+			return false;
+		}
+
+		if ( ! $this->seek_to_matching_closer() ) {
+			parent::seek( 'SET_INNER_HTML: opener' );
+			return false;
+		}
+
+		if ( ! parent::set_bookmark( 'SET_INNER_HTML: closer' ) ) {
+			return false;
+		}
+
+		$inner_html_start  = $this->bookmarks['SET_INNER_HTML: opener']->start + $this->bookmarks['SET_INNER_HTML: opener']->length;
+		$inner_html_length = $this->bookmarks['SET_INNER_HTML: closer']->start - $inner_html_start;
+
+		$this->lexical_updates['innerHTML'] = new WP_HTML_Text_Replacement(
+			$inner_html_start,
+			$inner_html_length,
+			$html
+		);
+
+		parent::seek( 'SET_INNER_HTML: opener' );
+		parent::release_bookmark( 'SET_INNER_HTML: opener' );
+		parent::release_bookmark( 'SET_INNER_HTML: closer' );
+
+		// @todo check for whether that html will make a mess!
+		// Will it break out of tags?
+
+		return true;
+	}
+
+	public function seek_to_matching_closer(): bool {
+		$tag_name = $this->get_tag();
+
+		if ( null === $tag_name ) {
+			return false;
+		}
+
+		if ( $this->is_tag_closer() ) {
+			return false;
+		}
+
+		if ( ! $this->expects_closer() ) {
+			return false;
+		}
+
+		$breadcrumbs = $this->breadcrumbs;
+		array_pop( $breadcrumbs );
+
+		// @todo Can't use these queries together
+		while ( $this->next_tag(
+			array(
+				'tag_name'    => $this->get_tag(),
+				'tag_closers' => 'visit',
+			)
+		) ) {
+			if ( $this->get_breadcrumbs() === $breadcrumbs ) {
+				return true;
+			}
+		}
+		return false;
+	}
+
+
 	/**
 	 * Constructor.
 	 *
@@ -424,6 +533,61 @@ function ( WP_HTML_Token $token ): void {
 		};
 	}
 
+	/**
+	 * Creates a fragment processor with the current node as its context element.
+	 *
+	 * @see https://html.spec.whatwg.org/multipage/parsing.html#html-fragment-parsing-algorithm
+	 *
+	 * @param string $html     Input HTML fragment to process.
+	 * @return static|null     The created processor if successful, otherwise null.
+	 */
+	public function spawn_fragment_parser( string $html ): ?self {
+		if ( $this->get_token_type() !== '#tag' ) {
+			return null;
+		}
+
+		$namespace = $this->get_namespace();
+
+		/*
+		 * Prevent creating fragments at "self-contained" nodes.
+		 *
+		 * @see https://github.com/WordPress/wordpress-develop/pull/7141
+		 * @see https://github.com/WordPress/wordpress-develop/pull/7198
+		 */
+		if (
+			'html' === $namespace &&
+			in_array( $this->get_tag(), array( 'IFRAME', 'NOEMBED', 'NOFRAMES', 'SCRIPT', 'STYLE', 'TEXTAREA', 'TITLE', 'XMP' ), true )
+		) {
+			return null;
+		}
+
+		$fragment_processor              = self::create_fragment( $html );
+		$fragment_processor->compat_mode = $this->compat_mode;
+
+		$fragment_processor->context_node                = clone $this->state->current_token;
+		$fragment_processor->context_node->bookmark_name = 'context-node';
+		$fragment_processor->context_node->on_destroy    = null;
+
+		$context_element = array( $fragment_processor->context_node->node_name, array() );
+		foreach ( $this->get_attribute_names_with_prefix( '' ) as $name => $value ) {
+			$context_element[1][ $name ] = $value;
+		}
+
+		$fragment_processor->breadcrumbs = array();
+
+		if ( 'TEMPLATE' === $context_element[0] ) {
+			$fragment_processor->state->stack_of_template_insertion_modes[] = WP_HTML_Processor_State::INSERTION_MODE_IN_TEMPLATE;
+		}
+
+		$fragment_processor->reset_insertion_mode_appropriately();
+
+		// @todo Set the parser's form element pointer.
+
+		$fragment_processor->state->encoding_confidence = 'irrelevant';
+
+		return $fragment_processor;
+	}
+
 	/**
 	 * Stops the parser and terminates its execution when encountering unsupported markup.
 	 *
@@ -522,6 +686,7 @@ public function get_unsupported_exception() {
 	 *                                     1 for "first" tag, 3 for "third," etc.
 	 *                                     Defaults to first tag.
 	 *     @type string|null $class_name   Tag must contain this whole class name to match.
+	 *     @type string      $tag_name     Tag name to match.
 	 *     @type string[]    $breadcrumbs  DOM sub-path at which element is found, e.g. `array( 'FIGURE', 'IMG' )`.
 	 *                                     May also contain the wildcard `*` which matches a single element, e.g. `array( 'SECTION', '*' )`.
 	 * }
@@ -545,7 +710,7 @@ public function next_tag( $query = null ): bool {
 		}
 
 		if ( is_string( $query ) ) {
-			$query = array( 'breadcrumbs' => array( $query ) );
+			$query = array( 'tag_name' => $query );
 		}
 
 		if ( ! is_array( $query ) ) {

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

@@ -863,4 +863,101 @@ public function test_adjusts_for_mathml_integration_points() {
 			'Found the wrong namespace for the transformed "IMAGE"/"IMG" element.'
 		);
 	}
+
+	/**
+	 * @ticket TBD
+	 *
+	 * @dataProvider data_set_inner_html
+	 */
+	public function test_set_inner_html( string $html, ?string $replacement, string $expected ) {
+		$processor = WP_HTML_Processor::create_fragment( $html );
+		while ( $processor->next_tag() ) {
+			if ( $processor->get_attribute( 'target' ) ) {
+				break;
+			}
+		}
+
+		$this->assertTrue( $processor->set_inner_html( $replacement ) );
+		$this->assertSame( $expected, $processor->get_updated_html() );
+	}
+
+	public static function data_set_inner_html() {
+		return array(
+			array(
+				'<div target>replace me</div>',
+				'with me!',
+				'<div target>with me!</div>',
+			),
+			array(
+				'<div target><div><p><a>replace me</div></div>',
+				'with me!',
+				'<div target>with me!</div>',
+			),
+			array(
+				'<table target><td>replace me</table>',
+				'<td>with me!',
+				'<table target><tbody><tr><td>with me!</td></tr></tbody></table>',
+			),
+		);
+	}
+
+	/**
+	 * @ticket TBD
+	 *
+	 * @dataProvider data_set_inner_html_not_allowed
+	 */
+	public function test_set_inner_html_not_allowed( string $html, string $replacement ) {
+		$processor = WP_HTML_Processor::create_fragment( $html );
+		while ( $processor->next_tag() ) {
+			if ( $processor->get_attribute( 'target' ) ) {
+				break;
+			}
+		}
+		$this->assertFalse( $processor->set_inner_html( $replacement ), "Should have failed but produced: {$processor->get_updated_html()}" );
+		$this->assertSame( $html, $processor->get_updated_html() );
+	}
+
+	/**
+	 * Data provider.
+	 *
+	 * @return array[]
+	 */
+	public static function data_set_inner_html_not_allowed(): array {
+		return array(
+			'not allowed in void tags' => array(
+				'<br target>',
+				'anything',
+			),
+			'not allowed in self-closing tags' => array(
+				'<svg><text target />',
+				'anything',
+			),
+			'must have closing tag' => array(
+				'<body><div target></body>',
+				'anything',
+			),
+
+			'a in a' => array(
+				'<a target></a>',
+				'<a>',
+			),
+			'a nested in a' => array(
+				'<a><i><em><strong target></a>',
+				'<a>A cannot nest inside a',
+			),
+
+			'text in table' => array(
+				'<table target><td>hello</table>',
+				'text triggers forstering - not allowed',
+			),
+			'text in thead' => array(
+				'<table><thead target><td>hello</thead>',
+				'text triggers forstering - not allowed',
+			),
+			'text in tr' => array(
+				'<table><tr target>hello</tr>',
+				'text triggers forstering - not allowed',
+			),
+		);
+	}
 }

tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php

@@ -138,10 +138,6 @@ public function data_external_html5lib_tests() {
 	 * @return bool True if the test case should be skipped. False otherwise.
 	 */
 	private static function should_skip_test( ?string $test_context_element, string $test_name ): bool {
-		if ( null !== $test_context_element && 'body' !== $test_context_element ) {
-			return true;
-		}
-
 		if ( array_key_exists( $test_name, self::SKIP_TESTS ) ) {
 			return true;
 		}
@@ -157,11 +153,77 @@ private static function should_skip_test( ?string $test_context_element, string
 	 * @return string|null Tree structure of parsed HTML, if supported, else null.
 	 */
 	private static function build_tree_representation( ?string $fragment_context, string $html ) {
-		$processor = $fragment_context
-			? WP_HTML_Processor::create_fragment( $html, "<{$fragment_context}>" )
-			: WP_HTML_Processor::create_full_parser( $html );
-		if ( null === $processor ) {
-			throw new WP_HTML_Unsupported_Exception( "Could not create a parser with the given fragment context: {$fragment_context}.", '', 0, '', array(), array() );
+		$processor = null;
+		if ( $fragment_context ) {
+			if ( 'body' === $fragment_context ) {
+				$processor = WP_HTML_Processor::create_fragment( $html );
+			} else {
+
+				/*
+				 * If the string of characters starts with "svg ", the context
+				 * element is in the SVG namespace and the substring after
+				 * "svg " is the local name. If the string of characters starts
+				 * with "math ", the context element is in the MathML namespace
+				 * and the substring after "math " is the local name.
+				 * Otherwise, the context element is in the HTML namespace and
+				 * the string is the local name.
+				 */
+				if ( str_starts_with( $fragment_context, 'svg ' ) ) {
+					$tag_name = substr( $fragment_context, 4 );
+					if ( 'svg' === $tag_name ) {
+						$parent_processor = WP_HTML_Processor::create_full_parser( '<!DOCTYPE html><svg>' );
+					} else {
+						$parent_processor = WP_HTML_Processor::create_full_parser( "<!DOCTYPE html><svg><{$tag_name}>" );
+					}
+					$parent_processor->next_tag( $tag_name );
+				} elseif ( str_starts_with( $fragment_context, 'math ' ) ) {
+					$tag_name = substr( $fragment_context, 5 );
+					if ( 'math' === $tag_name ) {
+						$parent_processor = WP_HTML_Processor::create_full_parser( '<!DOCTYPE html><math>' );
+					} else {
+						$parent_processor = WP_HTML_Processor::create_full_parser( "<!DOCTYPE html><math><{$tag_name}>" );
+					}
+					$parent_processor->next_tag( $tag_name );
+				} else {
+					if ( in_array(
+						$fragment_context,
+						array(
+							'caption',
+							'col',
+							'colgroup',
+							'tbody',
+							'td',
+							'tfoot',
+							'th',
+							'thead',
+							'tr',
+						),
+						true
+					) ) {
+						$parent_processor = WP_HTML_Processor::create_full_parser( "<!DOCTYPE html><table><{$fragment_context}>" );
+						$parent_processor->next_tag();
+					} else {
+						$parent_processor = WP_HTML_Processor::create_full_parser( "<!DOCTYPE html><{$fragment_context}>" );
+					}
+					$parent_processor->next_tag( $fragment_context );
+				}
+				if ( null !== $parent_processor->get_unsupported_exception() ) {
+					throw $parent_processor->get_unsupported_exception();
+				}
+				if ( null !== $parent_processor->get_last_error() ) {
+					throw new Exception( $parent_processor->get_last_error() );
+				}
+				$processor = $parent_processor->spawn_fragment_parser( $html );
+			}
+
+			if ( null === $processor ) {
+				throw new WP_HTML_Unsupported_Exception( "Could not create a parser with the given fragment context: {$fragment_context}.", '', 0, '', array(), array() );
+			}
+		} else {
+			$processor = WP_HTML_Processor::create_full_parser( $html );
+			if ( null === $processor ) {
+				throw new Exception( 'Could not create a full parser.' );
+			}
 		}
 
 		/*