HTML API: Implement reset_insertion_mode

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

@@ -59,6 +59,114 @@ class WP_HTML_Processor_State {
 	 */
 	const INSERTION_MODE_IN_BODY = 'insertion-mode-in-body';
 
+	/**
+	 * In select insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-inselect
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_SELECT = 'insertion-mode-in-select';
+
+	/**
+	 * In select in table insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-inselectintable
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_SELECT_IN_TABLE = 'insertion-mode-in-select-in-table';
+
+	/**
+	 * In table insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-intable
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_TABLE = 'insertion-mode-in-table';
+
+	/**
+	 * In caption insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-incaption
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_CAPTION = 'insertion-mode-in-caption';
+
+	/**
+	 * In table body insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-intablebody
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_TABLE_BODY = 'insertion-mode-in-table-body';
+
+	/**
+	 * In row insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-inrow
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_ROW = 'insertion-mode-in-row';
+
+	/**
+	 * In cell insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-incell
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_CELL = 'insertion-mode-in-cell';
+
+	/**
+	 * In column group insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-incolumngroup
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_COLUMN_GROUP = 'insertion-mode-in-column-group';
+
+	/**
+	 * In frameset insertion mode for full HTML parser.
+	 *
+	 * @since 6.5.0
+	 *
+	 * @see https://html.spec.whatwg.org/#parsing-main-inframeset
+	 * @see WP_HTML_Processor_State::$insertion_mode
+	 *
+	 * @var string
+	 */
+	const INSERTION_MODE_IN_FRAMESET = 'insertion-mode-in-frameset';
+
 	/**
 	 * Tracks open elements while scanning HTML.
 	 *
@@ -140,4 +248,169 @@ public function __construct() {
 		$this->stack_of_open_elements     = new WP_HTML_Open_Elements();
 		$this->active_formatting_elements = new WP_HTML_Active_Formatting_Elements();
 	}
+
+	/**
+	 * Runs the reset the insertion mode appropriately algorithm.
+	 *
+	 * @since 6.7.0
+	 *
+	 * @throws WP_HTML_Unsupported_Exception When encoutering unsupported nodes.
+	 *
+	 * @see https://html.spec.whatwg.org/multipage/parsing.html#reset-the-insertion-mode-appropriately
+	 */
+	public function reset_insertion_mode(): void {
+		/*
+		 * > 1. Let _last_ be false.
+		 * > 2. Let _node_ be the last node in the stack of open elements.
+		 * > 3. _Loop_: If _node_ is the first node in the stack of open elements, then set _last_
+		 * >            to true, and, if the parser was created as part of the HTML fragment parsing
+		 * >            algorithm (fragment case), set node to the context element passed to
+		 * >            that algorithm.
+		 * > …
+		 */
+		$last       = false;
+		$last_index = $this->stack_of_open_elements->count() - 1;
+		foreach ( $this->stack_of_open_elements->walk_up() as $i => $node ) {
+			if ( $i === $last_index ) {
+				$last = true;
+			}
+			switch ( $node->node_name ) {
+				/*
+				 * > 4. If node is a `select` element, run these substeps:
+				 * >   1. If _last_ is true, jump to the step below labeled done.
+				 * >   2. Let _ancestor_ be _node_.
+				 * >   3. _Loop_: If _ancestor_ is the first node in the stack of open elements,
+				 * >      jump to the step below labeled done.
+				 * >   4. Let ancestor be the node before ancestor in the stack of open elements.
+				 * >   …
+				 * >   7. Jump back to the step labeled _loop_.
+				 * >   8. _Done_: Switch the insertion mode to "in select" and return.
+				 */
+				case 'SELECT':
+					if ( ! $last ) {
+						foreach ( $this->stack_of_open_elements->walk_up( $node ) as $ancestor ) {
+							switch ( $ancestor->node_name ) {
+								/*
+								 * > 5. If _ancestor_ is a `template` node, jump to the step below
+								 * >    labeled _done_.
+								 */
+								case 'TEMPLATE':
+									break 2;
+
+								/*
+								 * > 6. If _ancestor_ is a `table` node, switch the insertion mode to
+								 * >    "in select in table" and return.
+								 */
+								case 'TABLE':
+									$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT_IN_TABLE;
+									return;
+							}
+						}
+					}
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT;
+					return;
+
+				/*
+				 * > 5. If _node_ is a `td` or `th` element and _last_ is false, then switch the
+				 * >    insertion mode to "in cell" and return.
+				 */
+				case 'TD':
+				case 'TH':
+					if ( ! $last ) {
+						$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_CELL;
+						return;
+					}
+					break;
+
+					/*
+					* > 6. If _node_ is a `tr` element, then switch the insertion mode to "in row"
+					* >    and return.
+					*/
+				case 'TR':
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_ROW;
+					return;
+
+				/*
+				 * > 7. If _node_ is a `tbody`, `thead`, or `tfoot` element, then switch the
+				 * >    insertion mode to "in table body" and return.
+				 */
+				case 'TBODY':
+				case 'THEAD':
+				case 'TFOOT':
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_TABLE_BODY;
+					return;
+
+				/*
+				 * > 8. If _node_ is a `caption` element, then switch the insertion mode to
+				 * >    "in caption" and return.
+				 */
+				case 'CAPTION':
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_CAPTION;
+					return;
+
+				/*
+				 * > 9. If _node_ is a `colgroup` element, then switch the insertion mode to
+				 * >    "in column group" and return.
+				 */
+				case 'COLGROUP':
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_COLUMN_GROUP;
+					return;
+
+				/*
+				 * > 10. If _node_ is a `table` element, then switch the insertion mode to
+				 * >     "in table" and return.
+				 */
+				case 'TABLE':
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_TABLE;
+					return;
+
+				/*
+				 * > 11. If _node_ is a `template` element, then switch the insertion mode to the
+				 * >     current template insertion mode and return.
+				 */
+				case 'TEMPLATE':
+					throw new WP_HTML_Unsupported_Exception( 'Cannot reset insertion mode at TEMPLATE node.' );
+
+				/*
+				 * > 12. If _node_ is a `head` element and _last_ is false, then switch the
+				 * >     insertion mode to "in head" and return.
+				 */
+				case 'HEAD':
+					throw new WP_HTML_Unsupported_Exception( 'Cannot reset insertion mode at HEAD node.' );
+
+				/*
+				 * > 13. If _node_ is a `body` element, then switch the insertion mode to "in body"
+				 * >     and return.
+				 */
+				case 'BODY':
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_BODY;
+					return;
+
+				/*
+				 * > 14. If _node_ is a `frameset` element, then switch the insertion mode to
+				 * >     "in frameset" and return. (fragment case)
+				 */
+				case 'FRAMESET':
+					$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_FRAMESET;
+					return;
+
+				/*
+				 * > 15. If _node_ is an `html` element, run these substeps:
+				 * >     1. If the head element pointer is null, switch the insertion mode to
+				 * >        "before head" and return. (fragment case)
+				 * >     2. Otherwise, the head element pointer is not null, switch the insertion mode to "after head" and return.
+				 */
+				case 'HTML':
+					throw new WP_HTML_Unsupported_Exception( 'Cannot reset insertion mode at HTML node.' );
+			}
+		}
+
+		/*
+		 * > 16. If _last_ is true, then switch the insertion mode to "in body"
+		 * >     and return. (fragment case)
+		 *
+		 * `$last` will always be true here, we've reached the end of the stack.
+		 */
+		$this->insertion_mode = WP_HTML_Processor_State::INSERTION_MODE_IN_BODY;
+	}
 }

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

@@ -0,0 +1,107 @@
+<?php
+/**
+ * Unit tests covering WP_HTML_Processor_State functionality.
+ *
+ * @package WordPress
+ * @subpackage HTML-API
+ *
+ * @since 6.7.0
+ *
+ * @group html-api
+ *
+ * @coversDefaultClass WP_HTML_Processor_State
+ */
+class Tests_HtmlApi_WpHtmlProcessorState extends WP_UnitTestCase {
+	/**
+	 * @dataProvider data_insertion_mode_cases
+	 *
+	 * @ticket 61549
+	 *
+	 * @param array $stack_of_open_elements   Stack of open elements.
+	 * @param string $expected_insertion_mode Expected insertion mode after running the algorithm.
+	 */
+	public function test_reset_insertion_mode(
+		array $stack_of_open_elements,
+		string $expected_insertion_mode
+	): void {
+		$state               = new WP_HTML_Processor_State();
+		$state->context_node = array( 'BODY', array() );
+
+		foreach ( $stack_of_open_elements as $i => $tag_name ) {
+			if ( ! ctype_upper( $tag_name ) ) {
+				throw new Error( 'Expected upper case tag names.' );
+			}
+			$state->stack_of_open_elements->push( new WP_HTML_Token( $i, $tag_name, false ) );
+		}
+		$state->reset_insertion_mode();
+
+		$this->assertSame( $expected_insertion_mode, $state->insertion_mode );
+	}
+
+	/**
+	 * Data provider.
+	 *
+	 * @return array[]
+	 */
+	public static function data_insertion_mode_cases(): array {
+		return array(
+			'SELECT last element'         => array( array( 'SELECT' ), WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT ),
+			'SELECT'                      => array( array( 'HTML', 'BODY', 'SELECT' ), WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT ),
+			'SELECT in table'             => array( array( 'HTML', 'BODY', 'TABLE', 'TBODY', 'TR', 'TD', 'SELECT' ), WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT_IN_TABLE ),
+			'SELECT in template in table' => array( array( 'HTML', 'BODY', 'TABLE', 'TBODY', 'TR', 'TD', 'TEMPLATE', 'SELECT' ), WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT ),
+			'SELECT > OPTION'             => array( array( 'HTML', 'BODY', 'SELECT', 'OPTION' ), WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT ),
+			'SELECT > OPTGROUP > OPTION'  => array( array( 'HTML', 'BODY', 'SELECT', 'OPTGROUP', 'OPTION' ), WP_HTML_Processor_State::INSERTION_MODE_IN_SELECT ),
+			'TD'                          => array( array( 'HTML', 'BODY', 'TABLE', 'TBODY', 'TR', 'TD' ), WP_HTML_Processor_State::INSERTION_MODE_IN_CELL ),
+			'TD (last element)'           => array( array( 'TD' ), WP_HTML_Processor_State::INSERTION_MODE_IN_BODY ),
+			'TH'                          => array( array( 'HTML', 'BODY', 'TABLE', 'TBODY', 'TR', 'TH' ), WP_HTML_Processor_State::INSERTION_MODE_IN_CELL ),
+			'TH (last element)'           => array( array( 'TH' ), WP_HTML_Processor_State::INSERTION_MODE_IN_BODY ),
+			'TR'                          => array( array( 'HTML', 'BODY', 'TABLE', 'TBODY', 'TR' ), WP_HTML_Processor_State::INSERTION_MODE_IN_ROW ),
+			'TBODY'                       => array( array( 'HTML', 'BODY', 'TABLE', 'TBODY' ), WP_HTML_Processor_State::INSERTION_MODE_IN_TABLE_BODY ),
+			'THEAD'                       => array( array( 'HTML', 'BODY', 'TABLE', 'THEAD' ), WP_HTML_Processor_State::INSERTION_MODE_IN_TABLE_BODY ),
+			'TFOOT'                       => array( array( 'HTML', 'BODY', 'TABLE', 'TFOOT' ), WP_HTML_Processor_State::INSERTION_MODE_IN_TABLE_BODY ),
+			'CAPTION'                     => array( array( 'HTML', 'BODY', 'TABLE', 'CAPTION' ), WP_HTML_Processor_State::INSERTION_MODE_IN_CAPTION ),
+			'COLGROUP'                    => array( array( 'HTML', 'BODY', 'TABLE', 'COLGROUP' ), WP_HTML_Processor_State::INSERTION_MODE_IN_COLUMN_GROUP ),
+			'TABLE'                       => array( array( 'HTML', 'BODY', 'TABLE' ), WP_HTML_Processor_State::INSERTION_MODE_IN_TABLE ),
+			'BODY'                        => array( array( 'HTML', 'BODY' ), WP_HTML_Processor_State::INSERTION_MODE_IN_BODY ),
+			'FRAMESET'                    => array( array( 'HTML', 'BODY', 'FRAMESET' ), WP_HTML_Processor_State::INSERTION_MODE_IN_FRAMESET ),
+			'Last element (DIV)'          => array( array( 'DIV' ), WP_HTML_Processor_State::INSERTION_MODE_IN_BODY ),
+		);
+	}
+
+	/**
+	 * @dataProvider data_insertion_mode_unsupported
+	 *
+	 * @ticket 61549
+	 *
+	 * @param array $stack_of_open_elements Stack of open elements.
+	 */
+	public function test_reset_insertion_mode_unsupported( array $stack_of_open_elements ): void {
+		$this->expectException( WP_HTML_Unsupported_Exception::class );
+
+		$state               = new WP_HTML_Processor_State();
+		$state->context_node = array( 'BODY', array() );
+
+		foreach ( $stack_of_open_elements as $i => $tag_name ) {
+			if ( ! ctype_upper( $tag_name ) ) {
+				throw new Error( 'Expected upper case tag names.' );
+			}
+			$state->stack_of_open_elements->push( new WP_HTML_Token( $i, $tag_name, false ) );
+		}
+		$state->reset_insertion_mode();
+	}
+
+	/**
+	 * Data provider.
+	 *
+	 * These tests should be migrated to the supported tests as support for more elements is added.
+	 *
+	 * @return array[]
+	 */
+	public static function data_insertion_mode_unsupported(): array {
+		return array(
+			'TEMPLATE requires template insertion mode stack' => array( array( 'HTML', 'BODY', 'TEMPLATE' ) ),
+			'HEAD requires more insertion modes' => array( array( 'HTML', 'HEAD' ) ),
+			'HTML requires head pointer and insertion modes' => array( array( 'HTML' ) ),
+		);
+	}
+}