HTML API: Ensure that full processor can seek to earlier bookmarks

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

@@ -5324,8 +5324,7 @@ public function seek( $bookmark_name ): bool {
 		 */
 		if ( 'backward' === $direction ) {
 			/*
-			 * Instead of clearing the parser state and starting fresh, calling the stack methods
-			 * maintains the proper flags in the parser.
+			 * When moving backward, stateful stacks should be cleared.
 			 */
 			foreach ( $this->state->stack_of_open_elements->walk_up() as $item ) {
 				if ( 'context-node' === $item->bookmark_name ) {
@@ -5343,32 +5342,65 @@ public function seek( $bookmark_name ): bool {
 				$this->state->active_formatting_elements->remove_node( $item );
 			}
 
-			parent::seek( 'context-node' );
-			$this->state->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_BODY;
-			$this->state->frameset_ok    = true;
-			$this->element_queue         = array();
-			$this->current_element       = null;
+			/*
+			 * **After** clearing stacks, more processor state can be reset.
+			 * This must be done after clearing the stack because those stacks generate events that
+			 * would appear on a subsequent call to `next_token()`.
+			 */
+			$this->state->frameset_ok                       = true;
+			$this->state->stack_of_template_insertion_modes = array();
+			$this->state->head_element                      = null;
+			$this->state->form_element                      = null;
+			$this->state->current_token                     = null;
+			$this->current_element                          = null;
+			$this->element_queue                            = array();
 
-			if ( isset( $this->context_node ) ) {
-				$this->breadcrumbs = array_slice( $this->breadcrumbs, 0, 2 );
+			/*
+			 * The absence of a context node indicates a full parse.
+			 * The presence of a context node indicates a fragment parser.
+			 */
+			if ( null === $this->context_node ) {
+				$this->change_parsing_namespace( 'html' );
+				$this->state->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_INITIAL;
+				$this->breadcrumbs           = array();
+
+				$this->bookmarks['initial'] = new WP_HTML_Span( 0, 0 );
+				parent::seek( 'initial' );
+				unset( $this->bookmarks['initial'] );
 			} else {
-				$this->breadcrumbs = array();
-			}
-		}
+				$this->change_parsing_namespace(
+					$this->context_node->integration_node_type
+						? 'html'
+						: $this->context_node->namespace
+				);
+
+				if ( 'TEMPLATE' === $this->context_node->node_name ) {
+					$this->state->stack_of_template_insertion_modes[] = WP_HTML_Processor_State::INSERTION_MODE_IN_TEMPLATE;
+				}
 
-		// When moving forwards, reparse the document until reaching the same location as the original bookmark.
-		if ( $bookmark_starts_at === $this->bookmarks[ $this->state->current_token->bookmark_name ]->start ) {
-			return true;
+				$this->reset_insertion_mode_appropriately();
+				$this->breadcrumbs = array_slice( $this->breadcrumbs, 0, 2 );
+				parent::seek( $this->context_node->bookmark_name );
+			}
 		}
 
-		while ( $this->next_token() ) {
+		/*
+		 * Here, the processor moves forward through the document until it matches the bookmark.
+		 * do-while is used here because the processor is expected to already be stopped on
+		 * a token than may match the bookmarked location.
+		 */
+		do {
+			/*
+			 * The processor will stop on virtual tokens, but bookmarks may not be set on them.
+			 * They should not be matched when seeking a bookmark, skip them.
+			 */
+			if ( $this->is_virtual() ) {
+				continue;
+			}
 			if ( $bookmark_starts_at === $this->bookmarks[ $this->state->current_token->bookmark_name ]->start ) {
-				while ( isset( $this->current_element ) && WP_HTML_Stack_Event::POP === $this->current_element->operation ) {
-					$this->current_element = array_shift( $this->element_queue );
-				}
 				return true;
 			}
-		}
+		} while ( $this->next_token() );
 
 		return false;
 	}

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

@@ -0,0 +1,143 @@
+<?php
+/**
+ * Unit tests covering WP_HTML_Processor bookmark functionality.
+ *
+ * @package WordPress
+ * @subpackage HTML-API
+ */
+
+/**
+ * @group html-api
+ *
+ * @coversDefaultClass WP_HTML_Processor
+ */
+class Tests_HtmlApi_WpHtmlProcessor_Bookmark extends WP_UnitTestCase {
+	/**
+	 * @dataProvider data_processor_constructors
+	 *
+	 * @ticket 62290
+	 */
+	public function test_processor_seek_same_location( callable $factory ) {
+		$processor = $factory( '<div><span>' );
+		$this->assertTrue( $processor->next_tag( 'DIV' ) );
+		$this->assertTrue( $processor->set_bookmark( 'mark' ), 'Failed to set bookmark.' );
+		$this->assertTrue( $processor->has_bookmark( 'mark' ), 'Failed has_bookmark check.' );
+
+		// Confirm the bookmark works and processing continues normally.
+		$this->assertTrue( $processor->seek( 'mark' ), 'Failed to seek to bookmark.' );
+		$this->assertSame( 'DIV', $processor->get_tag() );
+		$this->assertSame( array( 'HTML', 'BODY', 'DIV' ), $processor->get_breadcrumbs() );
+		$this->assertTrue( $processor->next_tag() );
+		$this->assertSame( 'SPAN', $processor->get_tag() );
+		$this->assertSame( array( 'HTML', 'BODY', 'DIV', 'SPAN' ), $processor->get_breadcrumbs() );
+	}
+
+	/**
+	 * @dataProvider data_processor_constructors
+	 *
+	 * @ticket 62290
+	 */
+	public function test_processor_seek_backward( callable $factory ) {
+		$processor = $factory( '<div><span>' );
+		$this->assertTrue( $processor->next_tag( 'DIV' ) );
+		$this->assertTrue( $processor->set_bookmark( 'mark' ), 'Failed to set bookmark.' );
+		$this->assertTrue( $processor->has_bookmark( 'mark' ), 'Failed has_bookmark check.' );
+
+		// Move past the bookmark so it must scan backwards.
+		$this->assertTrue( $processor->next_tag( 'SPAN' ) );
+
+		// Confirm the bookmark works.
+		$this->assertTrue( $processor->seek( 'mark' ), 'Failed to seek to bookmark.' );
+		$this->assertSame( 'DIV', $processor->get_tag() );
+	}
+
+	/**
+	 * @dataProvider data_processor_constructors
+	 *
+	 * @ticket 62290
+	 */
+	public function test_processor_seek_forward( callable $factory ) {
+		$processor = $factory( '<div one></div><span two></span><a three>' );
+		$this->assertTrue( $processor->next_tag( 'DIV' ) );
+		$this->assertTrue( $processor->set_bookmark( 'one' ), 'Failed to set bookmark "one".' );
+		$this->assertTrue( $processor->has_bookmark( 'one' ), 'Failed "one" has_bookmark check.' );
+
+		// Move past the bookmark so it must scan backwards.
+		$this->assertTrue( $processor->next_tag( 'SPAN' ) );
+		$this->assertTrue( $processor->get_attribute( 'two' ) );
+		$this->assertTrue( $processor->set_bookmark( 'two' ), 'Failed to set bookmark "two".' );
+		$this->assertTrue( $processor->has_bookmark( 'two' ), 'Failed "two" has_bookmark check.' );
+
+		// Seek back.
+		$this->assertTrue( $processor->seek( 'one' ), 'Failed to seek to bookmark "one".' );
+		$this->assertSame( 'DIV', $processor->get_tag() );
+
+		// Seek forward and continue processing.
+		$this->assertTrue( $processor->seek( 'two' ), 'Failed to seek to bookmark "two".' );
+		$this->assertSame( 'SPAN', $processor->get_tag() );
+		$this->assertTrue( $processor->get_attribute( 'two' ) );
+
+		$this->assertTrue( $processor->next_tag() );
+		$this->assertSame( 'A', $processor->get_tag() );
+		$this->assertTrue( $processor->get_attribute( 'three' ) );
+	}
+
+	/**
+	 * Ensure the parsing namespace is handled when seeking from foreign content.
+	 *
+	 * @dataProvider data_processor_constructors
+	 *
+	 * @ticket 62290
+	 */
+	public function test_seek_back_from_foreign_content( callable $factory ) {
+		$processor = $factory( '<custom-element /><svg><rect />' );
+		$this->assertTrue( $processor->next_tag( 'CUSTOM-ELEMENT' ) );
+		$this->assertTrue( $processor->set_bookmark( 'mark' ), 'Failed to set bookmark "mark".' );
+		$this->assertTrue( $processor->has_bookmark( 'mark' ), 'Failed "mark" has_bookmark check.' );
+
+		/*
+		 * <custom-element /> has self-closing flag, but HTML elements (that are not void elements) cannot self-close,
+		 * they must be closed by some means, usually a closing tag.
+		 *
+		 * If the div were interpreted as foreign content, it would self-close.
+		 */
+		$this->assertTrue( $processor->has_self_closing_flag() );
+		$this->assertTrue( $processor->expects_closer(), 'Incorrectly interpreted HTML custom-element with self-closing flag as self-closing element.' );
+
+		// Proceed into foreign content.
+		$this->assertTrue( $processor->next_tag( 'RECT' ) );
+		$this->assertSame( 'svg', $processor->get_namespace() );
+		$this->assertTrue( $processor->has_self_closing_flag() );
+		$this->assertFalse( $processor->expects_closer() );
+		$this->assertSame( array( 'HTML', 'BODY', 'CUSTOM-ELEMENT', 'SVG', 'RECT' ), $processor->get_breadcrumbs() );
+
+		// Seek back.
+		$this->assertTrue( $processor->seek( 'mark' ), 'Failed to seek to bookmark "mark".' );
+		$this->assertSame( 'CUSTOM-ELEMENT', $processor->get_tag() );
+		// If the parsing namespace were not correct here (html),
+		// then the self-closing flag would be misinterpreted.
+		$this->assertTrue( $processor->has_self_closing_flag() );
+		$this->assertTrue( $processor->expects_closer(), 'Incorrectly interpreted HTML custom-element with self-closing flag as self-closing element.' );
+
+		// Proceed into foreign content again.
+		$this->assertTrue( $processor->next_tag( 'RECT' ) );
+		$this->assertSame( 'svg', $processor->get_namespace() );
+		$this->assertTrue( $processor->has_self_closing_flag() );
+		$this->assertFalse( $processor->expects_closer() );
+
+		// The RECT should still descend from the CUSTOM-ELEMENT despite its self-closing flag.
+		$this->assertSame( array( 'HTML', 'BODY', 'CUSTOM-ELEMENT', 'SVG', 'RECT' ), $processor->get_breadcrumbs() );
+	}
+
+	/**
+	 * Data provider.
+	 *
+	 * @return array
+	 */
+	public static function data_processor_constructors(): array {
+		return array(
+			'Full parser'     => array( array( WP_HTML_Processor::class, 'create_full_parser' ) ),
+			'Fragment parser' => array( array( WP_HTML_Processor::class, 'create_fragment' ) ),
+		);
+	}
+}

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

@@ -133,7 +133,7 @@ public function test_clear_to_navigate_after_seeking() {
 
 			// Create a bookmark inside of that stack.
 			if ( null !== $processor->get_attribute( 'two' ) ) {
-				$processor->set_bookmark( 'two' );
+				$this->assertTrue( $processor->set_bookmark( 'two' ) );
 				break;
 			}
 		}