Handle all the different document context in html5lib tests

sirreal · sirreal · commit 25b18fa88d86 · 2024-10-11T10:21:34.000+02:00
diff --git a/tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php b/tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php
@@ -153,11 +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.' );
+			}
 		}
 
 		/*
