Convert HTML spans to act like void blocks; there is no closer.

Author: dmsnell

src/wp-includes/class-wp-block-processor.php

@@ -138,6 +138,11 @@
  *
  * These special block types can be passed into any method which searches for blocks.
  *
+ * There is one additional special block type which may be returned from
+ * {@see self::get_printable_block_type()}. This is the `#innerHTML` type, which
+ * indicates that the HTML span on which the processor is paused is inner HTML for
+ * a containing block.
+ *
  * ### Spans of HTML
  *
  * Non-block content plays a complicated role in processing block documents. This
@@ -258,10 +263,6 @@
  *    with a reimplementation of {@see static::get_attributes()}.
  *
  * @since 6.9.0
- * @todo Add optimized lookup for blocks of a known name
- * @todo is it necessary to have opening and closing HTML delimiters? or would one text node work?
- * @todo there will almost never be incomplete tokens, but we check on every step of tokenization.
- *       Could we run a happy-path without checking and then perform a single verification at the end?
  */
 class WP_Block_Processor {
 	/**
@@ -456,8 +457,7 @@ class WP_Block_Processor {
 	 *
 	 * @see self::READY
 	 * @see self::MATCHED
-	 * @see self::HTML_OPEN
-	 * @see self::HTML_CLOSE
+	 * @see self::HTML_SPAN
 	 * @see self::INCOMPLETE_INPUT
 	 * @see self::COMPLETE
 	 *
@@ -745,11 +745,6 @@ public function next_token() {
 			$this->was_void = false;
 		}
 
-		if ( self::HTML_OPEN === $this->state ) {
-			$this->state = self::HTML_CLOSE;
-			return true;
-		}
-
 		$text = $this->source_text;
 		$end  = strlen( $text );
 
@@ -760,7 +755,7 @@ public function next_token() {
 		 * document it implies that a real delimiter was found; otherwise this must be the
 		 * terminating HTML span and the parsing is complete.
 		 */
-		if ( self::HTML_CLOSE === $this->state ) {
+		if ( self::HTML_SPAN === $this->state ) {
 			if ( $this->matched_delimiter_at >= $end ) {
 				$this->state = self::COMPLETE;
 				return false;
@@ -829,7 +824,7 @@ public function next_token() {
 
 				// Whether or not there is a potential delimiter, there might be an HTML span.
 				if ( $after_prev_delimiter < ( $end - $backup ) ) {
-					$this->state                    = self::HTML_OPEN;
+					$this->state                    = self::HTML_SPAN;
 					$this->after_previous_delimiter = $after_prev_delimiter;
 					$this->matched_delimiter_at     = $end - $backup;
 					$this->matched_delimiter_length = $backup;
@@ -1101,7 +1096,7 @@ public function next_token() {
 
 		// HTML spans are visited before the delimiter that follows them.
 		if ( $comment_opening_at > $after_prev_delimiter ) {
-			$this->state                = self::HTML_OPEN;
+			$this->state                = self::HTML_SPAN;
 			$this->open_blocks_at[]     = $after_prev_delimiter;
 			$this->open_blocks_length[] = 0;
 			$this->was_void             = true;
@@ -1376,11 +1371,8 @@ public function get_last_json_error(): int {
 	 */
 	public function get_delimiter_type(): ?string {
 		switch ( $this->state ) {
-			case self::HTML_OPEN:
-				return self::OPENER;
-
-			case self::HTML_CLOSE:
-				return self::CLOSER;
+			case self::HTML_SPAN:
+				return self::VOID;
 
 			case self::MATCHED:
 				return $this->type;
@@ -1555,12 +1547,8 @@ public static function are_equal_block_types( $a_text, $a_at, $a_length, $b_text
 	 *              opens a block of one of the given block types, if provided.
 	 */
 	public function opens_block( string ...$block_type ): bool {
-		if ( self::HTML_CLOSE === $this->state ) {
-			return false;
-		}
-
 		// HTML spans only open implicit freeform content at the top level.
-		if ( self::HTML_OPEN === $this->state && 1 !== count( $this->open_blocks_at ) ) {
+		if ( self::HTML_SPAN === $this->state && 1 !== count( $this->open_blocks_at ) ) {
 			return false;
 		}
 
@@ -1597,10 +1585,7 @@ public function opens_block( string ...$block_type ): bool {
 	 * @return bool Whether the scanner is matched on an HTML span.
 	 */
 	public function is_html(): bool {
-		return (
-			self::HTML_OPEN === $this->state ||
-			self::HTML_CLOSE === $this->state
-		);
+		return self::HTML_SPAN === $this->state;
 	}
 
 	/**
@@ -1650,11 +1635,6 @@ public function get_html_content_and_advance(): ?string {
 			return null;
 		}
 
-		// Finish on the (implicit) end of the HTML span.
-		if ( self::HTML_OPEN === $this->state ) {
-			$this->state = self::HTML_CLOSE;
-		}
-
 		return substr(
 			$this->source_text,
 			$this->after_previous_delimiter,
@@ -1764,7 +1744,9 @@ public function get_printable_block_type(): ?string {
 
 		// This is a core/freeform text block, it’s special.
 		if ( $this->is_html() ) {
-			return 'core/freeform';
+			return 1 === count( $this->open_blocks_at )
+				? 'core/freeform'
+				: '#innerHTML';
 		}
 
 		$block_type = substr( $this->source_text, $this->namespace_at, $this->name_at - $this->namespace_at + $this->name_length );
@@ -1895,8 +1877,7 @@ public function allocate_and_return_parsed_attributes(): ?array {
 	 */
 	public function get_span(): ?WP_HTML_Span {
 		switch ( $this->state ) {
-			case self::HTML_OPEN:
-			case self::HTML_CLOSE:
+			case self::HTML_SPAN:
 				return new WP_HTML_Span( $this->after_previous_delimiter, $this->matched_delimiter_at - $this->after_previous_delimiter );
 
 			case self::MATCHED:
@@ -1964,16 +1945,7 @@ public function get_span(): ?WP_HTML_Span {
 	 *
 	 * @since 6.9.0
 	 */
-	const HTML_OPEN = 'scanner-opening-freeform';
-
-	/**
-	 * Indicates that the scanner is matched on the closing of an implicit freeform delimiter.
-	 *
-	 * @see self::$state
-	 *
-	 * @since 6.9.0
-	 */
-	const HTML_CLOSE = 'scanner-closing-freeform';
+	const HTML_SPAN = 'scanner-html-span';
 
 	/**
 	 * Indicates that the parser started parsing a block comment delimiter, but