From fd3ef27e254645cac4d4f1dd3abbfc9c431592ce Mon Sep 17 00:00:00 2001
From: Alec Geatches <alec.geatches@automattic.com>
Date: Fri, 18 Oct 2024 10:19:08 -0600
Subject: [PATCH 1/4] Add failing test for multiple-node raw sources

---
 tests/parser/sources/test-source-raw.php | 28 ++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/tests/parser/sources/test-source-raw.php b/tests/parser/sources/test-source-raw.php
index f6d52b9..813d6bc 100644
--- a/tests/parser/sources/test-source-raw.php
+++ b/tests/parser/sources/test-source-raw.php
@@ -79,4 +79,32 @@ public function test_parse_raw_source__nested() {
 		$this->assertArrayHasKey( 'blocks', $blocks, sprintf( 'Unexpected parser output: %s', wp_json_encode( $blocks ) ) );
 		$this->assertArraySubset( $expected_blocks, $blocks['blocks'], true );
 	}
+
+	public function test_parse_raw_source_multiple_top_level_nodes() {
+		$this->register_block_with_attributes( 'test/html', [
+			'content' => [
+				'type'   => 'string',
+				'source' => 'raw',
+			],
+		] );
+
+		$html = '
+			<!-- wp:test/html -->
+			<p>Node 1</p><p>Node 2</p>
+			<!-- /wp:test/html -->';
+
+		$expected_blocks = [
+			[
+				'name'       => 'test/html',
+				'attributes' => [
+					'content' => '<p>Node 1</p><p>Node 2</p>',
+				],
+			],
+		];
+
+		$content_parser = new ContentParser( $this->get_block_registry() );
+		$blocks         = $content_parser->parse( $html );
+		$this->assertArrayHasKey( 'blocks', $blocks, sprintf( 'Unexpected parser output: %s', wp_json_encode( $blocks ) ) );
+		$this->assertArraySubset( $expected_blocks, $blocks['blocks'], true );
+	}
 }

From 14858ee64cf015409c40ca2db552c3979b1e21f1 Mon Sep 17 00:00:00 2001
From: Alec Geatches <alec.geatches@automattic.com>
Date: Fri, 18 Oct 2024 10:58:34 -0600
Subject: [PATCH 2/4] Add failing test for internal whitespace

---
 tests/parser/sources/test-source-raw.php | 29 ++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/tests/parser/sources/test-source-raw.php b/tests/parser/sources/test-source-raw.php
index 813d6bc..59c60f5 100644
--- a/tests/parser/sources/test-source-raw.php
+++ b/tests/parser/sources/test-source-raw.php
@@ -107,4 +107,33 @@ public function test_parse_raw_source_multiple_top_level_nodes() {
 		$this->assertArrayHasKey( 'blocks', $blocks, sprintf( 'Unexpected parser output: %s', wp_json_encode( $blocks ) ) );
 		$this->assertArraySubset( $expected_blocks, $blocks['blocks'], true );
 	}
+
+	public function test_parse_raw_source_multiple_top_level_nodes_with_whitespace() {
+		$this->register_block_with_attributes( 'test/html', [
+			'content' => [
+				'type'   => 'string',
+				'source' => 'raw',
+			],
+		] );
+
+		$html = '
+			<!-- wp:test/html -->
+			<span class="highlight">This</span> <span>should</span> <span>retain</span>&nbsp;<span>inner</span>
+<span>whitespace</span>
+			<!-- /wp:test/html -->';
+
+		$expected_blocks = [
+			[
+				'name'       => 'test/html',
+				'attributes' => [
+					'content' => "<span class=\"highlight\">This</span> <span>should</span> <span>retain</span>&nbsp;<span>inner</span>\n<span>whitespace</span>",
+				],
+			],
+		];
+
+		$content_parser = new ContentParser( $this->get_block_registry() );
+		$blocks         = $content_parser->parse( $html );
+		$this->assertArrayHasKey( 'blocks', $blocks, sprintf( 'Unexpected parser output: %s', wp_json_encode( $blocks ) ) );
+		$this->assertArraySubset( $expected_blocks, $blocks['blocks'], true );
+	}
 }

From b6be8dcb3d183fbfa4fd29ab8082f8ce95e3f61b Mon Sep 17 00:00:00 2001
From: Alec Geatches <alec.geatches@automattic.com>
Date: Fri, 18 Oct 2024 10:59:05 -0600
Subject: [PATCH 3/4] Change `source_block_raw` to use parent body tag to
 preserve whitespace

---
 src/parser/content-parser.php | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/src/parser/content-parser.php b/src/parser/content-parser.php
index c5c4e94..bfad2f2 100644
--- a/src/parser/content-parser.php
+++ b/src/parser/content-parser.php
@@ -680,10 +680,15 @@ protected function source_block_raw( $crawler ) {
 		// Also see tag attribute parsing in Gutenberg:
 		// https://github.com/WordPress/gutenberg/blob/6517008/packages/blocks/src/api/parser/get-block-attributes.js#L131
 
-		$attribute_value = null;
-
 		if ( $crawler->count() > 0 ) {
-			$attribute_value = trim( $crawler->outerHtml() );
+			// $crawler's outerHtml() will only return the HTML of the first node in this raw HTML.
+			// If the raw HTML contains multiple top-level nodes, we need to use the inner HTML of the wrapping
+			// 'body' tag. This will also preserve internal whitespace in the HTML.
+			$body_node = $crawler->closest( 'body' );
+
+			if ( $body_node && $body_node->count() > 0 ) {
+				$attribute_value = trim( $body_node->html() );
+			}
 		}
 
 		return $attribute_value;

From a4939bbe953548544e670e74207d1973d1ab4ae0 Mon Sep 17 00:00:00 2001
From: Alec Geatches <alec.geatches@automattic.com>
Date: Fri, 18 Oct 2024 11:07:25 -0600
Subject: [PATCH 4/4] Fix removed default value

---
 src/parser/content-parser.php | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/parser/content-parser.php b/src/parser/content-parser.php
index bfad2f2..7c8535b 100644
--- a/src/parser/content-parser.php
+++ b/src/parser/content-parser.php
@@ -680,6 +680,8 @@ protected function source_block_raw( $crawler ) {
 		// Also see tag attribute parsing in Gutenberg:
 		// https://github.com/WordPress/gutenberg/blob/6517008/packages/blocks/src/api/parser/get-block-attributes.js#L131
 
+		$attribute_value = null;
+
 		if ( $crawler->count() > 0 ) {
 			// $crawler's outerHtml() will only return the HTML of the first node in this raw HTML.
 			// If the raw HTML contains multiple top-level nodes, we need to use the inner HTML of the wrapping